U.S. patent application number 11/016182 was filed with the patent office on 2005-09-22 for very large scale immobilized polymer synthesis.
This patent application is currently assigned to Affymetrix, INC.. Invention is credited to Fodor, Stephen P.A., Pirrung, Michael C., Read, J. Leighton, Stryer, Lubert.
Application Number | 20050208537 11/016182 |
Document ID | / |
Family ID | 27502770 |
Filed Date | 2005-09-22 |
United States Patent
Application |
20050208537 |
Kind Code |
A1 |
Fodor, Stephen P.A. ; et
al. |
September 22, 2005 |
Very large scale immobilized polymer synthesis
Abstract
A synthetic strategy for the creation of large scale chemical
diversity. Solid-phase chemistry, photolabile protecting groups,
and photolithography are used to achieve light-directed
spatially-addressable parallel chemical synthesis. Binary masking
techniques are utilized in one embodiment. A reactor system,
photoremovable protecting groups, and improved data collection and
handling techniques are also disclosed. A technique for screening
linker molecules is also provided.
Inventors: |
Fodor, Stephen P.A.; (Palo
Alto, CA) ; Stryer, Lubert; (Stanford, CA) ;
Pirrung, Michael C.; (Chapel Hill, NC) ; Read, J.
Leighton; (Palo Alto, CA) |
Correspondence
Address: |
AFFYMETRIX, INC
ATTN: CHIEF IP COUNSEL, LEGAL DEPT.
3380 CENTRAL EXPRESSWAY
SANTA CLARA
CA
95051
US
|
Assignee: |
Affymetrix, INC.
Santa Clara
CA
|
Family ID: |
27502770 |
Appl. No.: |
11/016182 |
Filed: |
December 16, 2004 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11016182 |
Dec 16, 2004 |
|
|
|
10928299 |
Aug 27, 2004 |
|
|
|
10928299 |
Aug 27, 2004 |
|
|
|
10190951 |
Jul 8, 2002 |
|
|
|
10190951 |
Jul 8, 2002 |
|
|
|
10014716 |
Dec 14, 2001 |
|
|
|
10014716 |
Dec 14, 2001 |
|
|
|
08348471 |
Nov 30, 1994 |
|
|
|
6420169 |
|
|
|
|
08348471 |
Nov 30, 1994 |
|
|
|
07805727 |
Dec 6, 1991 |
|
|
|
5424186 |
|
|
|
|
07805727 |
Dec 6, 1991 |
|
|
|
07492462 |
Mar 7, 1990 |
|
|
|
5143854 |
|
|
|
|
07492462 |
Mar 7, 1990 |
|
|
|
07362901 |
Jun 7, 1989 |
|
|
|
07805727 |
|
|
|
|
07624120 |
Dec 6, 1990 |
|
|
|
07624120 |
Dec 6, 1990 |
|
|
|
07492462 |
Mar 7, 1990 |
|
|
|
5143854 |
|
|
|
|
07492462 |
Mar 7, 1990 |
|
|
|
07362901 |
Jun 7, 1989 |
|
|
|
Current U.S.
Class: |
435/6.11 ;
382/128; 435/6.12; 702/20 |
Current CPC
Class: |
B01J 19/0046 20130101;
B01J 2219/00475 20130101; C40B 40/06 20130101; B01J 2219/00725
20130101; C07D 263/44 20130101; C07H 19/04 20130101; C07H 21/00
20130101; C07K 1/045 20130101; G01N 15/1475 20130101; B01J
2219/00529 20130101; Y02P 20/55 20151101; C07C 229/14 20130101;
G03F 7/265 20130101; B01J 2219/00434 20130101; B01J 2219/00648
20130101; B01J 2219/00722 20130101; B01J 2219/00608 20130101; C07D
317/62 20130101; C07K 17/14 20130101; G03F 7/38 20130101; G11C
13/0014 20130101; B01J 2219/005 20130101; B01J 2219/00659 20130101;
C07K 17/06 20130101; C12Q 1/6809 20130101; B01J 2219/00315
20130101; C40B 50/14 20130101; G11C 13/0019 20130101; B01J
2219/00675 20130101; B01J 2219/00436 20130101; B01J 2219/00527
20130101; B01J 2219/00695 20130101; C07K 7/06 20130101; B01J
2219/00585 20130101; C07K 1/047 20130101; B01J 2219/00641 20130101;
G03F 7/00 20130101; B01J 2219/00389 20130101; B01J 2219/00605
20130101; B01J 2219/00711 20130101; G01N 33/54373 20130101; B01J
2219/00626 20130101; B01J 2219/0061 20130101; B01J 2219/00689
20130101; B82Y 30/00 20130101; C12Q 1/6816 20130101; C07K 1/042
20130101; C40B 60/14 20130101; G01N 21/6452 20130101; B01J
2219/0059 20130101; B01J 2219/00637 20130101; B82Y 10/00 20130101;
C12Q 1/6874 20130101; B01J 2219/00617 20130101; B01J 2219/00612
20130101; G01N 21/6458 20130101; B01J 2219/00432 20130101; C07C
229/16 20130101; C07H 19/10 20130101; C12Q 1/6837 20130101; B01J
2219/00459 20130101; B01J 2219/00596 20130101; B01J 2219/00531
20130101; C07B 2200/11 20130101; B01J 2219/00468 20130101; G01N
21/6428 20130101; C40B 40/10 20130101 |
Class at
Publication: |
435/006 ;
702/020; 382/128 |
International
Class: |
C12Q 001/68; G06F
011/00; G06F 019/00; G01N 033/48; G01N 033/50 |
Claims
1. A method of detecting hybridization between biological polymers,
comprising the acts of: providing a substrate having a surface
including at least one biological polymer and at least one
fluorescent label associated with the biological polymer;
generating an excitation laser beam; scanning said laser beam
relative to said surface; collecting fluorescent radiation
responsive to said laser beam using optics; detecting said
collected fluorescent radiation; and autofocusing to bring into
focus with respect to said optics at least a portion of said
surface including the biological polymer.
2-62. (canceled)
63. A method for synthesizing a plurality of biopolymers on the
surface of a support, said method comprising: (a) placing said
support into a reaction chamber and applying to said surface said
biopolymers or precursors of said biopolymers, (b) removing said
support from said reaction chamber and placing said support into a
flow chamber, (c) introducing a liquid reagent for conducting said
synthesis into said flow chamber, (d) removing said liquid reagent
from said flow chamber wherein the pressure in said chamber is
maintained substantially atmospheric during said removing. (e)
removing said support from said flow chamber and (f) repeating
steps (a)-(e) to form said plurality of biopolymers on the surface
of said support.
64. A method according to claim 63 wherein liquid reagent is
removed from said flow chamber under vacuum.
65. A method according to claim 63 wherein liquid reagent is
removed from said flow chamber by simultaneously venting and
applying a vacuum to said flow chamber.
66. A method according to claim 89 wherein said venting and said
applying a vacuum are carried out at opposite ends of said flow
chamber.
67. A method according to claim 63 wherein said method further
comprises holding said liquid reagent in said flow chamber for a
predetermined period of time.
68. A method according to claim 63 wherein said support is
glass.
69. A method according to claim 63 further comprising introducing a
pressurized inert gas into said flow chamber after step (c) and
simultaneously evacuating said flow chamber.
70. A method according to claim 63 wherein said biopolymers are
polynucleotides.
71. A method according to claim 63 wherein said liquid reagent for
conducting said synthesis comprises an oxidizing agent or an agent
for removing a protecting group.
72. A method according to claim 63 wherein said biopolymers are
synthesized on said surface in multiple arrays and said support is
subsequently diced into individual arrays of biopolymers on a
support.
73. A method according to claim 72 further comprising exposing the
array to a sample and reading the array.
74. A method according to claim 73 comprising forwarding data
representing a result obtained from a reading of the array.
75. A method according to claim 74 wherein the data is transmitted
to a remote location.
76. A method according to claim 75 comprising receiving data
representing a result of an interrogation obtained by the reading
of the array.
77. A method for synthesizing an array of biopolymers on the
surface of a support wherein said synthesis comprises a plurality
of monomer additions, said method comprising after each of said
monomer additions: (a) placing said support into a flow chamber,
(c) introducing a liquid reagent for conducting said synthesis into
said flow chamber, (d) removing said reagent from said flow chamber
by simultaneously venting said chamber and applying a vacuum to the
interior of said chamber, (e) removing said support from said flow
chamber and (f) repeating steps (a)-(e) to form said plurality of
biopolymers on the surface of said support.
78. A method according to claim 77 wherein said venting and said
applying a vacuum are carried out at opposite ends of said flow
chamber.
79. A method according to claim 77 wherein said method further
comprises holding said liquid reagent in said flow chamber for a
predetermined period of time.
80. A method according to claim 77 wherein said support is
glass.
81. A method according to claim 77 further comprising introducing a
pressurized inert gas into said flow chamber after step (d) and
simultaneously evacuating said flow chamber.
82. A method according to claim 77 wherein said biopolymers are
polynucleotides.
83. A method according to claim 77 wherein said liquid reagent for
conducting said synthesis is an oxidizing agent or an agent foi
removing a protecting group.
84. A method according to claim 77 wherein said biopolymers are
synthesized on said surface in multiple arrays and said support is
subsequently diced into individual arrays of biopolymers on a
support.
85. A flow cell assembly for conducting at least one reaction in
the synthesis of an array of biopolymers on the surface of a
support, said flow cell comprising: (a) a flow cell chamber, (b) a
manifold in fluid communication with said chamber, said manifold
comprising at least a wash reagent inlet, an inlet for a reagent
for conducting a step of said synthesis, and a vent, and (c) a
vacuum source in fluid communication with said flow cell
chamber.
86. A flow cell assembly according to claim 85 further comprising a
fluid level sensor and a controller for controlling said inlets,
said vent and said vacuum source.
87. A flow cell assembly according to claim 85 further comprising a
gas inlet.
88. An apparatus for synthesizing an array of biopolymers on the
surface of a support, said apparatus comprising: (a) one or more
flow cell assemblies of claim 85, (b) one or more fluid dispensing
stations in fluid communication with one or more of said plurality
of flow cell assemblies, (c) a station for monomer addition to said
surface of said support, and (d) a mechanism for moving a support
to and from said station for monomer addition and a flow cell and
from one flow cell to another flow cell.
89. An apparatus according to claim 88 further comprising a
controller for controlling the movement of said mechanism.
90. An apparatus according to claim 88 wherein said mechanism is a
robotic arm.
91. A method comprising using an array, prepared by an apparatus
according to claim 88, by exposing the array to a sample and
reading the array.
92. A method according to claim 91 comprising forwarding data
representing a result obtained from a reading of the array.
93. A method according to claim 92 wherein the data is transmitted
to a remote location.
94. A method according to claim 93 comprising receiving data
representing a result of an interrogation obtained by the reading
of the array.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of U.S. Ser. No.
08/348,471, filed Nov. 30, 1994, which is a continuation of U.S.
Ser. No. 07/805,727 (filed Dec. 6, 1991, now U.S. Pat. No.
5,424,186); which is a continuation-in-part of U.S. Ser. No.
07/492,462 (filed Mar. 7, 1990, now U.S. Pat. No. 5,143,854); which
is a continuation-in-part of U.S. Ser. No. 07/362,901 (filed Jun.
7, 1989, now abandoned). U.S. Ser. No. 07/805,727 (filed Dec. 6,
1991, now U.S. Pat. No. 5,424,186) is also a continuation-in-part
of U.S. Ser. No. 07/624,120 (filed Dec. 6, 1990, now abandoned);
which is a continuation-in-part of U.S. Ser. No. 07/492,462 (filed
Mar. 3, 1990, now U.S. Pat. No. 5,143,854) and U.S. Ser. No.
07/362,901 (filed Jun. 7, 1989, now abandoned).
COPYRIGHT NOTICE
[0002] A portion of the disclosure of this patent document contains
material which is subject to copyright protection. The copyright
owner has no objection to the facsimile reproduction by anyone of
the patent document or the patent disclosure exactly as it appears
in the Patent and Trademark Office patent file or records, but
otherwise reserves all copyright rights whatsoever.
[0003] This application is also related to the following
applications: U.S. Ser. No. 07/626,730 (filed Dec. 6, 1990, now
U.S. Pat. No. 5,547,839); U.S. Ser. No. 07/624,114 (filed Dec. 6,
1990, now abandoned); U.S. Ser. No. 07/796,243 (filed Nov. 22,
1991, now U.S. Pat. No. 5,384,261); U.S. Ser. No. 07/796,947 (filed
Nov. 22, 1991, now U.S. Pat. No. 5,242,974); U.S. Ser. No.
07/796,727 (filed Nov. 22, 1991, now U.S. Pat. No. 5,242,974); and
PCT Publication No. WO 90/15070 (published Dec. 13, 1990).
[0004] The disclosures of all of these applications are
incorporated herein by reference in their entirety and for all
purposes.
BACKGROUND OF THE INVENTION
[0005] The present invention relates to the field of polymer
synthesis. More specifically, the invention provides a reactor
system, a masking strategy, photoremovable protecting groups, data
collection and processing techniques, and applications for light
directed synthesis of diverse polymer sequences on substrates.
[0006] Prior methods of preparing large numbers of different
polymers have been painstakingly slow when used at a scale
sufficient to permit effective rational or random screening. For
example, the "Merrifield" method (J. Am. Chem. Soc. (1963)
85:2149-2154, which is incorporated herein by reference for all
purposes) has been used to synthesize peptides on a solid support.
In the Merrifield method, an amino acid is covalently bonded to a
support made of an insoluble polymer. Another amino acid with an
alpha protecting group is reacted with the covalently bonded amino
acid to form a dipeptide. After washing, the protecting group is
removed and a third amino acid with an alpha protecting group is
added to the dipeptide. This process is continued until a peptide
of a desired length and sequence is obtained. Using the Merrifield
method, it is not economically practical to synthesize more than a
handful of peptide sequences in a day.
[0007] To synthesize larger numbers of polymer sequences, it has
also been proposed to use a series of reaction vessels for polymer
synthesis. For example, a tubular reactor system may be used to
synthesize a linear polymer on a solid phase support by automated
sequential addition of reagents. This method still does not enable
the synthesis of a sufficiently large number of polymer sequences
for effective economical screening, i.e., for purposes of drug
discovery.
[0008] Methods of preparing a plurality of polymer sequences are
also known in which a container encloses a known quantity of
reactive particles, the particles being larger in size than
foramina of the container. The containers may be selectively
reacted with desired materials to synthesize desired sequences of
product molecules. As with other methods known in the art, this
method cannot practically be used to synthesize a sufficient
variety of polypeptides for effective screening.
[0009] Other techniques have also been described. These methods
include the synthesis of peptides on 96 plastic pins which fit the
format of standard microtiter plates. Unfortunately, while these
techniques have been somewhat useful, substantial problems remain.
For example, these methods continue to be limited in the diversity
of sequences which can be economically synthesized and
screened.
[0010] From the above, it is seen that an improved method and
apparatus for synthesizing a variety of chemical sequences at known
locations is desired.
SUMMARY OF THE INVENTION
[0011] Methods, apparatus, and compositions for synthesis and use
of diverse polymer sequences on a substrate are disclosed, as well
as applications thereof.
[0012] In one preferred embodiment, linker molecules are provided
on a substrate. A terminal end of the linker molecules is provided
with a reactive functional group protected with a photoremovable
protective group. Using lithographic methods, the photoremovable
protecting group is exposed to light and removed from the linker
molecules in the first selected regions. The substrate is then
washed or otherwise contacted with a first monomer that reacts with
exposed functional groups on the linker molecules. In a preferred
embodiment, the monomer is an amino acid containing a
photoremovable protecting group at its amino or carboxy terminus
and the linker molecule terminates in an amino or carboxy acid
group bearing a photoremovable protective group.
[0013] A second set of selected regions is, thereafter, exposed to
light and the photoremovable protecting group on the linker
molecule/protected amino acid is removed at the second set of
regions. The substrate is then contacted with a second monomer
containing a photoremovable protecting group for reaction with
exposed functional groups. This process is repeated to selectively
apply monomers until polymers of a desired length and desired
chemical sequence are obtained. Photolabile groups are then
optionally removed and the sequence is, thereafter, optionally
capped. Side chain protective groups, if present, are also
removed.
[0014] By using the lithographic techniques disclosed herein, it is
possible to direct light to relatively small and precisely known
locations on the substrate. It is, therefore, possible to
synthesize polymers of a known chemical sequence at known locations
on the substrate. The general version of this technique is termed
very Large Scale Immobilized Polymer Synthesis (VLSIPS.TM.).
[0015] The resulting substrate will have a variety of uses
including, for example, screening large numbers of polymers for
biological activity. To screen for biological activity, the
substrate is exposed to one or more receptors such as an antibody,
whole cells, receptors on vesicles, lipids, or any one of a variety
of other receptors. The receptors are preferably labeled with, for
example, a fluorescent marker, radioactive marker, or a labeled
antibody reactive with the receptor. The location of the marker on
the substrate is detected with, for example, photon detection or
autoradiographic techniques. Through knowledge of the sequence of
the material at the location where binding is detected, it is
possible to quickly determine which sequence binds with the
receptor and, therefore, the technique can be used to screen large
numbers of peptides. Other possible applications of the inventions
herein include diagnostics in which various antibodies for
particular receptors would be placed on a substrate and, for
example, blood sera would be screened for immune deficiencies.
Still further applications include, for example, selective "doping"
of organic materials in semiconductor devices, i.e., the
introduction of selected impurities into the device, and the
like.
[0016] According to one aspect of the invention, an improved
reactor system for synthesis of diverse polymer sequences on a
substrate is provided. According to this embodiment the invention
provides for a reactor for contacting reaction fluids to a
substrate; a system for delivering selected reaction fluids to the
reactor; a translation stage for moving a mask or substrate from at
least a first relative location relative to a second relative
location; a light for illuminating the substrate through a mask at
selected times; and an appropriately programmed digital computer
for selectively directing a flow of fluids from the reactor system,
selectively activating the translation stage, and selectively
illuminating the substrate so as to form a plurality of diverse
polymer sequences on the substrate at predetermined locations.
[0017] The invention also provides a technique for selection of
linker molecules using the VLSIPS.TM. synthesis technique.
According to this aspect of the invention, the invention provides a
method of screening a plurality of linker polymers for use in
binding affinity studies. The invention includes the steps of
forming a plurality of linker polymers on a substrate in selected
regions, the linker polymers are formed by the steps of
recursively: (1) on a surface of a substrate, irradiating a portion
of the selected regions to remove a protecting group, and
contacting the surface with a monomer; (2) contacting the plurality
of linker polymers with a ligand; and (3) contacting the ligand
with a labeled receptor.
[0018] According to another aspect of the invention, improved
photoremovable protecting groups are provided. According to this
aspect of the invention a compound having the formula: 1
[0019] wherein n=0 or 1; Y is selected from the group consisting of
an oxygen of the carboxyl group of a natural or unnatural amino
acid, an amino group of a natural or unnatural amino acid, or the
C-5' oxygen group of a natural or unnatural deoxyribonucleic or
ribonucleic acid; R.sup.1 and R.sup.2 independently are a hydrogen
atom, a lower alkyl, aryl, benzyl, halogen, hydroxyl, alkoxyl,
thiol, thioether, amino, nitro, carboxyl, formate, formamido,
sulfido, or phosphido group; and R.sup.3 is a alkoxy, alkyl, aryl,
hydrogen, or alkenyl group is provided.
[0020] The invention also provides improved masking techniques for
VLSIPS. According to one aspect of the masking technique, the
invention provides an ordered method for forming a plurality of
polymer sequences by sequential addition of reagents comprising the
step of serially protecting and deprotecting portions of the
plurality of polymer sequences for addition of other portions of
the polymer sequences using a combinatorial synthesis strategy.
[0021] Improved data collection equipment and techniques are also
provided. According to one embodiment, the instrumentation provides
a system for determining affinity of a receptor to a ligand
comprising: means for applying light to a surface of a substrate,
the substrate comprising a plurality of ligands at predetermined
locations, the means for applying light providing simultaneous
illumination at a plurality of the predetermined locations; and an
array of detectors for detecting fluorescence at the plurality of
predetermined locations. The invention further provides for
improved data analysis techniques including the steps of exposing
fluorescently labelled receptors to a substrate, the substrate
comprising a plurality of ligands in regions at known locations; at
a plurality of data collection points within each of the regions,
determining an amount of fluorescence from the data collection
points; removing the data collection points deviating from a
predetermined statistical distribution; and determining a relative
binding affinity of the receptor from remaining data collection
points.
[0022] Protected amino acid N-carboxy anhydrides for use in polymer
synthesis are also disclosed. According to this aspect of the
invention, a compound having the following formula is provided:
2
[0023] where R is a side chain of a natural or unnatural amino acid
and X is a photoremovable protecting group.
[0024] A further understanding of the nature and advantages of the
inventions herein may be realized by reference to the remaining
portions of the specification and the attached drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0025] FIGS. 1 to 7 illustrate masking, irradiation, and coupling
of monomers;
[0026] FIGS. 8A and 8B are fluorescence curves for NVOC slides not
exposed and exposed to light respectively;
[0027] FIGS. 9A-9D are fluorescence plots of slides exposed through
100 .mu.m, 50 .mu.m, 20 .mu.m and 10 .mu.m masks;
[0028] FIG. 10 illustrates fluorescence of a slide with the peptide
YGGFL (SEQ. ID NO:1) on selected region on its surface which has
been exposed to labeled Herz antibodies specific for the
sequence;
[0029] FIG. 11 schematically illustrates one example of
light-directed peptide synthesis;
[0030] FIG. 12 is a fluorescence plot of YGGFL (SEQ. ID NO:1) and
PGGFL synthesized in a 50 .mu.m checkerboard pattern;
[0031] FIGS. 13A-13D illustrate formation and screening of a
checkerboard pattern of YGGFL (SEQ. ID NO:1) and GGFL (SEQ. ID
NO:15);
[0032] FIG. 14 is a fluorescence plot of YPGGFL (SEQ. ID NO:3) and
YGGFL (SEQ. ID NO:1) synthesized in a 50 .mu.m checkerboard
pattern;
[0033] FIGS. 15A and 15B illustrate the raping of 16 sequences
synthesized on two different glass slides;
[0034] FIG. 16 is a fluorescence plot of the slide illustrated in
FIG. 15A;
[0035] FIG. 17 is a fluorescence plot of the slide illustrated in
FIG. 15B;
[0036] FIG. 18 is a fluorescence plot of an experiment which
produced 4,096 compounds;
[0037] FIG. 19 is a fluorescence plot of a substrate on which
65,536 different compounds were formed;
[0038] FIGS. 20A and 20B show a tripeptide used in a fluorescence
energy-transfer substrate assay and that substrate after
cleavage;
[0039] FIGS. 21A and 21B are fluorescence plots generated with
fluorescence energy-transfer substrate assays;
[0040] FIGS. 22A and 22B illustrate alterative embodiments of a
reactor system for forming a plurality of polymers on a
substrate;
[0041] FIG. 23 schematically illustrates an automated system for
synthesizing diverse polymer sequences;
[0042] FIGS. 24A and 24B illustrate operation of a program for
polymer synthesis;
[0043] FIG. 25 is a schematic illustration of a "pure" binary
masking strategy;
[0044] FIG. 26 is a schematic illustration of a gray code binary
masking strategy;
[0045] FIG. 27 is a schematic illustration of a modified gray code
binary masking strategy;
[0046] FIG. 28A schematically illustrates a masking strategy for a
four step synthesis;
[0047] FIG. 28B schematically illustrates synthesis of 400 peptide
dimers of genetically coded amino acids;
[0048] FIG. 29 is a coordinate map for the ten-step binary
synthesis;
[0049] FIG. 30 is a fluorescence plot of a 4.times.10 array of
peptides having sequences similar to dynorphin B;
[0050] FIG. 31 illustrates a strategy for producing an array of
peptides related to the dynorphin B sequence;
[0051] FIG. 32 is a fluorescence plot of an array of peptides
produced according to the strategy illustrated in FIG. 31;
[0052] FIG. 33 is a fluorescence plot of an array of peptides
containing various deletions from the dynorphin B sequence;
[0053] FIG. 34 is a plot of the relative binding affinities of an
anti-dynorphin B monoclonal antibody to various sequences produced
on the substrate shown in FIG. 33;
[0054] FIG. 35 schematically illustrates a data collection
system;
[0055] FIG. 36 is a block diagram illustrating the architecture of
the data collection system;
[0056] FIG. 37 is a flow chart illustrating operation of software
for the data collection/analysis system;
[0057] FIG. 38 schematically illustrates one example of
light-directed oligonucleotide synthesis;
[0058] FIG. 39A-39C are fluorescence plots demonstrating
hybridization, dehybridization and rehybridization between
immobilized poly A and poly T;
[0059] FIGS. 40A-40E illustrate a synthesis strategy for forming
polysaccharides in accordance with the present invention; and
[0060] FIG. 41 illustrates the introduction of a 1reduced amide
bond into a growing peptide.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
Contents
[0061] I. Definitions
[0062] II. General
[0063] A. Deprotection and Addition
[0064] 1. Example--Polymer Synthesis
[0065] 2. Example
[0066] 3. Example--Slide Preparation
[0067] 4. Example--Synthesis of a Dimer of an Amino-propyl Group
and a Fluorescent Group
[0068] 5. Example--Removal of NVOC and Attachment of a Marker
[0069] 6. Example--Use of a Mask in Removal of NVOC
[0070] 7. Example
[0071] 8. Example
[0072] B. Antibody recognition
[0073] 1. Example--Attachment of YGGFL and Subsequent Exposure to
Herz Antibody and Goat Antimouse
[0074] 2. Example
[0075] 3. Example--Monomer-by-Monomer Formation of YGGFL and
Subsequent Exposure to Labeled Antibody
[0076] 4. Example--Monomer-by-Monomer Synthesis of YGGFL and
YPGGFL
[0077] 5. Example--Synthesis of an Array of Sixteen Different Amino
Acid Sequences and Estimation of Relative Binding Affinity to Herz
Antibody
[0078] 6. Example
[0079] 7. Example
[0080] C. Fluorescence Energy-Transfer Substrate Assays
[0081] III. Synthesis
[0082] A. Reactor System
[0083] B. Binary Synthesis Strategy
[0084] 1. Example
[0085] 2. Example
[0086] 3. Example
[0087] 4. Example
[0088] 5. Example
[0089] 6. Example
[0090] 7. Example
[0091] 8. Example
[0092] 9. Example
[0093] 10. Example
[0094] C. Linker Selection
[0095] D. Protecting Groups
[0096] 1. Use of Photoremovable Protecting Groups During
Solid-Phase Synthesis of Peptides
[0097] 2. Use of Photoremovable Protecting Groups During
Solid-Phase Synthesis of Oligonucleotides
[0098] E. Amino Acid N-Carboxy Anhydrides Protected with a
Photoremovable Group
[0099] IV. Data Collection
[0100] A. Data Collection System
[0101] B. Data Analysis
[0102] C. Alternative Embodiments
[0103] V. Other Representative Applications
[0104] A. Oligonucleotide Synthesis
[0105] 1. Example
[0106] 2. Example
[0107] 3. Example
[0108] B. Oligosaccharide Synthesis
[0109] 1. Example
[0110] C. Caged Binding Member System
[0111] D. Fingerprinting for Quality Control
[0112] E. .beta.-Amino Acid and D-Amino Acid Monomers
[0113] F. Reduced Amide Bonds
[0114] VI. Conclusion
[0115] I. Definitions
[0116] Certain terms used herein are intended to have the following
general definitions:
[0117] 1. Complementary: Refers to the topological compatibility or
matching together of interacting surfaces of a ligand molecule and
its receptor. Thus, the receptor and its ligand can be described as
complementary, and furthermore, the contact surface characteristics
are complementary to each other.
[0118] 2. Epitope: The portion of an antigen molecule which is
delineated by the area of interaction with the subclass of
receptors known as antibodies.
[0119] 3. Ligand: A ligand is a molecule that is recognized by a
particular receptor. Examples of ligands that can be investigated
by this invention include, but are not restricted to, agonists and
antagonists for cell membrane receptors, toxins and venoms, viral
epitopes, hormones (e.g., steroids, etc.), hormone receptors,
opiates, peptides, enzymes, enzyme substrates, cofactors, drugs,
lectins, sugars, oligonucleotides, nucleic acids, oligosaccharides,
proteins, and monoclonal antibodies.
[0120] 4. Monomer: A member of the set of small molecules which can
be joined together to form a polymer. The set of monomers includes
but is not restricted to, for example, the set of common L-amino
acids, the set of D-amino acids, the set of natural or synthetic
amino acids, the set of nucleotides and the set of pentoses and
hexoses. As used herein, monomer refers to any member of a basis
set for synthesis of a polymer. For example, dimers of the 20
naturally occurring L-amino acids form a basis set of 400 monomers
for synthesis of polypeptides. Different basis sets of monomers may
be used in any of the successive steps in the synthesis of a
polymer. Furthermore, each of the sets may include protected
members which are modified after synthesis.
[0121] 5. Peptide: A polymer in which the monomers are alpha amino
acids and which are joined together through amide bonds,
alternatively referred to as a polypeptide. In the context of this
specification it should be appreciated that the amino acids may,
for example, the L-optical isomer or the D-optical isomer. Peptides
are often two or more amino acid monomers long, and often 4 or more
amino acids long, often 5 or more amino acids long, often 10 or
more amino acids long, often 15 or more amino acids long, and often
20 or more amino acid monomers long, for example. Standard
abbreviations for amino acids are used (e.g., P for proline). These
abbreviations are included in Stryer, Biochemistry, Third Ed.,
1988, which is incorporated herein by reference for all
purposes.
[0122] 6. Radiation: Energy which may be selectively applied
including energy having a wavelength of between 10.sup.-14 and
10.sup.4 meters including, for example, electron beam radiation,
gamma radiation, x-ray radiation, ultra-violet radiation, visible
light, infrared radiation, microwave radiation, and radio waves.
"Irradiation" refers to the application of radiation to a
surface.
[0123] 7. Receptor: A molecule that has an affinity for a given
ligand. Receptors may be naturally-occurring or synthetic
molecules. Also, they can be employed in their unaltered state or
as aggregates with other species. Receptors may be attached,
covalently or noncovalently, to a binding member, either directly
or via a specific binding substance. Examples of receptors which
can be employed by this invention include, but are not restricted
to, antibodies, cell membrane receptors, monoclonal antibodies and
antisera reactive with specific antigenic determinants (such as on
viruses, cells, or other materials), drugs, polynucleotides,
nucleic acids, peptides, cofactors, lectins, sugars,
polysaccharides, cells, cellular membranes, and organelles.
Receptors are sometimes referred to in the art as anti-ligands. As
the term receptors is used herein, no difference in meaning is
intended. A "Ligand Receptor Pair" is formed when two
macromolecules have combined through molecular recognition to form
a complex.
[0124] Other examples of receptors which can be investigated by
this invention include but are not restricted to:
[0125] a) Microorganism receptors: Determination of ligands which
bind to receptors, such as specific transport proteins or enzymes
essential to survival of microorganisms, is useful for a new class
of antibiotics. Of particular value would be antibiotics against
opportunistic fungi, protozoa, and those bacteria resistant to the
antibiotics in current use.
[0126] b) Enzymes: For instance, determining the binding site of
enzymes such as the enzymes responsible for cleaving
neurotransmitters provides useful information. Determination of
ligands which bind to certain receptors to modulate the action of
the enzymes which cleave the different neurotransmitters is useful
in the development of drugs which can be used in the treatment of
disorders of neurotransmission.
[0127] c) Antibodies: For instance, the invention may be useful in
investigating the ligand-binding site on the antibody molecule
which combines with the epitope of an antigen of interest;
determining a sequence that mimics an antigenic epitope may lead to
the development of vaccines of which the immunogen is based on one
or more of such sequences or lead to the development of related
diagnostic agents or compounds useful in therapeutic treatments
such as for auto-immune diseases (e.g., by blocking the binding of
the "self" antibodies). "Antibody" as used herein can also include
antibody fragments, such as Fab fragments, which are composed of a
light chain and the variable region of a heavy chain.
[0128] d) Nucleic Acids: Sequences of nucleic acids may be
synthesized to establish DNA or RNA binding sequences.
[0129] e) Catalytic Polypeptides: Polymers, preferably
polypeptides, which are capable of promoting a chemical reaction
involving the conversion of one or more reactants to one or more
products. Such polypeptides generally include a binding site
specific for at least one reactant or reaction intermediate and an
active functionality proximate to the binding site, in which the
functionality is capable of chemically modifying the bound
reactant. Catalytic polypeptides are described in, for example,
U.S. application Ser. No. 07/404,920 (now U.S. Pat. No. 5,215,889),
which is incorporated herein by reference for all purposes.
[0130] f) Hormone receptors: For instance, the receptors for
insulin and growth hormone. Determination of the ligands which bind
with high affinity to a receptor is useful in the development of,
for example, an oral replacement of the daily injections which
diabetics must take to relieve the symptoms of diabetes, and in the
other case, a replacement for the scarce human growth hormone which
can only be obtained from cadavers or by recombinant DNA
technology. Other examples are the vasoconstrictive hormone
receptors; determination of those ligands which bind to a receptor
may lead to the development of drugs to control blood pressure.
[0131] g) Opiate receptors: Determination of ligands which bind to
the opiate receptors in the brain is useful in the development of
less-addictive replacements for morphine and related drugs.
[0132] 8. Substrate: A material having a rigid or semi-rigid
surface. In many embodiments, at least one surface of the substrate
will be substantially flat, although in some embodiments it may be
desirable to physically separate synthesis regions for different
polymers with, for example, wells, raised regions, etched trenches,
or the like. According to other embodiments, small beads may be
provided on the surface which may be released upon completion of
the synthesis.
[0133] 9. Protecting group: A material which is chemically bound to
a monomer unit and which may be removed upon selective exposure to
an activator such as electromagnetic radiation and, especially
ultraviolet and visible light. Examples of protecting groups with
utility herein include those comprising nitropiperonyl,
pyrenylmethoxy-carbonyl, nitroveratryl, nitrobenzyl, dimethyl
dimethoxybenzyl, 5-bromo-7-nitroindolinyl, o-hydroxy-.alpha.-methyl
cinnamoyl, and 2-oxymethylene anthraquinone.
[0134] 10. Predefined Region: A predefined region is a localized
area on a surface which is, was, or is intended to be activated for
formation of a polymer. The predefined region may have any
convenient shape, e.g., circular, rectangular, elliptical,
wedge-shaped, etc. For the sake of brevity herein, "predefined
regions" are sometimes referred to simply as "regions."
[0135] 11. Substantially Pure: A polymer is considered to be
"substantially pure" within a predefined region of a substrate when
it exhibits characteristics that distinguish it from other
predefined regions. Typically, purity will be measured in terms of
biological activity or function as a result of uniform sequence.
Such characteristics will typically be measured by way of binding
with a selected ligand or receptor. Preferably the region is
sufficiently pure such that the predominant species in the
predefined region is the desired sequence. According to preferred
aspects of the invention, the polymer is 5% pure, more preferably
more than 10% pure, preferably more than 20% pure, more preferably
more than 80% pure, more preferably more than 90% pure, more
preferably more than 95% pure, where purity for this purpose refers
to the ratio of the number of ligand molecules formed in a
predefined region having a desired sequence to the total number of
molecules formed in the predefined region.
[0136] 12. Activator: An energy source adapted to render a group
active and which is directed from a source to a predefined location
on a substrate. A primary illustration of an activator is light.
Other examples of activators include ion beams, electric fields,
magnetic fields, electron beams, x-ray, and the like.
[0137] 13. Combinatorial Synthesis Strategy: An ordered strategy
for parallel synthesis of diverse polymer sequences by sequential
addition of reagents which may be represented by a reactant matrix,
and a switch matrix, the product of which is a product matrix. A
reactant matrix is a 1 column by m row matrix of the building
blocks to be added. The switch matrix is all or a subset of the
binary numbers, preferably ordered, between 1 and m arranged in
columns. In preferred embodiments, a "binary strategy" is one in
which at least two successive steps illuminate half of a region of
interest on the substrate. In a binary synthesis strategy, all
possible compounds which can be formed from an ordered set of
reactants are formed. In most preferred embodiments, binary
synthesis refers to a synthesis strategy which also factors a
previous addition step. For example, a strategy in which a switch
matrix for a masking strategy halves regions that were previously
illuminated, illuminating about half of the previously illuminated
region and protecting the remaining half (while also protecting
about half of previously protected regions and illuminating about
half of previously protected regions). It will be recognized that
binary rounds may be interspersed with non-binary rounds and that
only a portion of a substrate may be subjected to a binary scheme.
A combinatorial "masking" strategy is a synthesis which uses light
or other deprotecting or activating agents to remove protecting
groups from materials for addition of other materials such as amino
acids. In some embodiments, selected columns of the switch matrix
are arranged in order of increasing binary numbers in the columns
of the switch matrix.
[0138] 14. Linker: A molecule or group of molecules attached to a
substrate and spacing a synthesized polymer from the substrate for
exposure/binding to a receptor.
[0139] 15. Abbreviations: The following abbreviations are intended
to have the following meanings:
[0140] BOC: benzyloxycarbonyl.
[0141] BOP: benzotriazol-1-yloxytris-(dimethylamino). phosphonium
hexafluorophosphate.
[0142] CCD: charge coupled device.
[0143] DCC: dicyclohexylcarbodiimide.
[0144] DCM: dichloromethane; methylene chloride.
[0145] DDZ: dimethoxydimethylbenzyloxy.
[0146] DIEA: N,N-diisopropylethylamine.
[0147] DMAP: 4-dimethylaminopyridine.
[0148] DMF: dimethyl formamide.
[0149] DMT: dimethoxytrityl.
[0150] FMOC: fluorenylmethyloxycarbonyl.
[0151] HBTU: 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium
hexafluorophosphate.
[0152] HOBT: 1-hydroxybenzotriazole hydrate.
[0153] NMP: N-methylpyrrolidone.
[0154] NV: nitroveratryl.
[0155] NVOC: 6-nitroveratryloxycarbonyl.
[0156] PG: protective group.
[0157] TFA: trifluoracetic acid.
[0158] THF: tetrahydrofuran.
[0159] II. General
[0160] The present invention provides synthetic strategies and
devices for the creation of large scale chemical diversity.
Solid-phase chemistry, photolabile protecting groups, and
photolithography are brought together to achieve light-directed
spatially-addressable parallel chemical synthesis in preferred
embodiments.
[0161] The invention is described herein for purposes of
illustration primarily with regard to the preparation of peptides
and nucleotides, but could readily be applied in the preparation of
other polymers. Such polymers include, for example, both linear and
cyclic polymers of nucleic acids, polysaccharides, phospholipids,
and peptides having either .alpha.-, .beta.-, or .omega.-amino
acids, heteropolymers in which a known drug is covalently bound to
any of the above, polyurethanes, polyesters, polycarbonates,
polyureas, polyamides, polyethyleneimines, polyarylene sulfides,
polysiloxanes, polyimides, polyacetates, or other polymers which
will be apparent upon review of this disclosure. It will be
recognized further, that illustrations herein are primarily with
reference to C- to N-terminal synthesis, but the invention could
readily be applied to N- to C-terminal synthesis without departing
from the scope of the invention. Methods for forming cyclic and
reversed polarity peptides and other polymers are described in U.S.
Pat. No. 5,242,974 and previously incorporated herein by
reference.
[0162] The prepared substrate may, for example, be used in
screening a variety of polymers as ligands for binding with a
receptor, although it will be apparent that the invention could be
used for the synthesis of a receptor for binding with a ligand. The
substrate disclosed herein will have a wide variety of other uses.
Merely by way of example, the invention herein can be used in
determining peptide and nucleic acid sequences which bind to
proteins, finding sequence-specific binding drugs, identifying
epitopes recognized by antibodies, and evaluation of a variety of
drugs for clinical and diagnostic applications, as well as
combinations of the above.
[0163] The invention preferably provides for the use of a substrate
"S" with a surface. Linker molecules "L" are optionally provided on
a surface of the substrate. The purpose of the linker molecules, in
some embodiments, is to facilitate receptor recognition of the
synthesized polymers.
[0164] Optionally, the linker molecules may be chemically protected
for storage purposes. A chemical storage protecting group such as
t-BOC (t-butoxycarbonyl) may be used in some embodiments. Such
chemical protective groups would be chemically removed upon
exposure to, for example, acidic solution and would serve to
protect the surface during storage and be removed prior to polymer
preparation.
[0165] On the substrate or a distal end of the linker molecules, a
functional group with a protecting group P.sub.0 is provided. The
protecting group P.sub.0 may be removed upon exposure to radiation,
electric fields, electric currents, or other activators to expose
the functional group.
[0166] In a preferred embodiment, the radiation is ultraviolet
(UV), infrared (IR), or visible light. As more fully described
below, the protecting group may alternatively be an
electrochemically-sensitive group which may be removed in the
presence of an electric field. In still further alternative
embodiments, ion beams, electron beams, or the like may be used for
deprotection.
[0167] A. Deprotection and Addition
[0168] Concurrently or after exposure of a known region of the
substrate to light, the surface is contacted with a first monomer
unit M.sub.1 which reacts with the functional group which has been
exposed by the deprotection step. The first monomer includes a
protecting group P.sub.1. P.sub.1 may or may not be the same as
P.sub.0.
[0169] Accordingly, after a first cycle, known first regions of the
surface may comprise the sequence:
[0170] S-L-M.sub.1-P.sub.1
[0171] while remaining regions of the surface comprise the
sequence:
[0172] S-L-P.sub.0.
[0173] Thereafter, second regions of the surface (which may include
the first region) are exposed to light and contacted with a second
monomer M.sub.2 (which may or may not be the same as M.sub.1)
having a protecting group P.sub.2. P.sub.2 may or may not be the
same as P.sub.0 and P.sub.1. After this second cycle, different
regions of the substrate may comprise one or more of the following
sequences:
1 S-L-M.sub.1-M.sub.2-P.sub.2 S-L-M.sub.2-P.sub.2
S-L-M.sub.1-P.sub.1 and/or S-L-P.sub.0.
[0174] The above process is repeated until the substrate includes
desired polymers of desired lengths. By controlling the locations
of the substrate exposed to light and the reagents exposed to the
substrate following exposure, the location of each sequence will be
known.
[0175] Thereafter, the protective groups are removed from some or
all of the substrate and the sequences are, optionally, capped with
a capping unit C. The process results in a substrate having a
surface with a plurality of polymers of the following general
formula:
[0176] S-[L]-(M.sub.i)-(M.sub.j)-(M.sub.k) . . . (M.sub.x)-[C]
[0177] where square brackets indicate optional groups, and M.sub.i
. . . M.sub.x indicates any sequence of monomers. The number of
monomers could cover a wide variety of values, but in a preferred
embodiment they will range from 2 to 100.
[0178] In some embodiments a plurality of locations on the
substrate polymers are to contain a common monomer subsequence. For
example, it may be desired to synthesize a sequence
S-M.sub.1-M.sub.2-M.sub.3 at first locations and a sequence
S-M.sub.4-M.sub.2-M.sub.3 at second locations. The process would
commence with irradiation of the first locations followed by
contacting with M.sub.1-P, resulting in the sequence S-M.sub.1-P at
the first location. The second locations would then be irradiated
and contacted with M.sub.4-P, resulting in the sequence S-M.sub.4-P
at the second locations. Thereafter both the first and second
locations would be irradiated and contacted with the dimer
M.sub.2-M.sub.3, resulting in the sequence
S-M.sub.1-M.sub.2-M.sub.3 at the first locations and
S-M.sub.4-M.sub.2-M.sub.3 at the second locations. Of course,
common subsequences of any length could be utilized including those
in a range of 2 or more monomers, 2 to 100 monomers, 2 to 20
monomers, and a most preferred range of 2 to 3 monomers.
[0179] According to other embodiments, a set of masks is used for
the first monomer layer and, thereafter, varied light wavelengths
are used for selective deprotection. For example, in the process
discussed above, first regions are first exposed through a mask and
reacted with a first monomer having a first protecting group
P.sub.1, which is removable upon exposure to a first wavelength of
light (e.g., IR). Second regions are masked and reacted with a
second monomer having a second protecive group P.sub.2 which is
removable upon exposure to a second wavelength of light (e.g. UV).
Thereafter, masks become unnecessary in the synthesis because the
entire substrate may be exposed alternatively to the first and
second wavelengths of light in the deprotection cycle.
[0180] The polymers prepared on a substrate according to the above
methods will have a variety of uses including, for example,
screening for biological activity. In such screening activities,
the substrate containing the sequences is exposed to an unlabeled
or labeled receptor such as an antibody, receptor on a cell,
phospholipid vesicle, or any one of a variety of other receptors.
In one preferred embodiment the polymers are exposed to a first,
unlabeled receptor of interest and, thereafter, exposed to a
labeled receptor-specific recognition element, which is, for
example, an antibody. This process will provide signal
amplification in the detection stage.
[0181] The receptor molecules may bind with one or more polymers on
the substrate. The presence of the labeled receptor and, therefore,
the presence of a sequence which binds with the receptor is
detected in a preferred embodiment through the use of
autoradiography, detection of fluorescence with a charge-coupled
device, fluorescence microscopy, or the like. The sequence of the
polymer at the locations where the receptor binding is detected may
be used to determine all or part of a sequence which is
complementary to the receptor.
[0182] Use of the invention herein is illustrated primarily with
reference to screening for biological activity. The invention will,
however, find many other uses. For example, the invention may be
used in information storage (e.g., on optical disks), production of
molecular electronic devices, production of stationary phases in
separation sciences, production of dyes and brightening agents,
photography, and in immobilization of cells, proteins, lectins,
nucleic acids, polysaccharides and the like in patterns on a
surface via molecular recognition of specific polymer sequences. By
synthesizing the same compound in adjacent, progressively differing
concentrations, a gradient will be established to control
chemotaxis or to develop diagnostic dipsticks which, for example,
titrate an antibody against an increasing amount of antigen. By
synthesizing several catalyst molecules in close proximity, more
efficient multistep conversions may be achieved by "coordinate
immobilization." Coordinate immobilization also may be used for
electron transfer systems, as well as to provide both structural
integrity and other desirable properties to materials such as
lubrication, wetting, etc.
[0183] According to alternative embodiments, molecular
biodistribution or pharmacokinetic properties may be examined. For
example, to assess resistance to intestinal or serum proteases,
polymers may be capped with a fluorescent tag and exposed to
biological fluids of interest.
[0184] FIG. 1 is a flow chart illustrating the process of forming
chemical compounds according to one embodiment of the invention.
Synthesis occurs on a solid support 2. A pattern of illumination
through a mask 4a using a light source 6 determines which regions
of the support are activated for chemical coupling. In one
preferred embodiment activation is accomplished by using light to
remove photolabile protecting groups from selected areas of the
substrate.
[0185] After deprotection, monomers indicated by "A" in FIG. 1,
each bearing a photolabile protecting group (indicated by "X"), are
exposed to the surface of the substrate and react with regions that
were addressed by light in the preceding step. The substrate is
then illuminated through a second mask 4b, which activates another
region for reaction with a second protected monomer "B." The
process is then repeated using desired masks and mask orientations
in combination with selected monomers. The pattern of masks used in
these illuminations and the sequence of reactants define the
ultimate products and their locations, resulting in diverse
sequences at predefined locations, as shown with the sequences ACEG
and BDFH in the lower portion of FIG. 1. Preferred embodiments of
the invention take advantage of combinatorial masking strategies to
form a large number of compounds in a small number of chemical
steps.
[0186] A high degree of miniaturization is possible because the
density of compounds is determined largely with regard to spatial
addressability of the activator, in one case the diffraction of
light. Each compound is physically accessible and its position is
precisely known. Hence, the array is spatially-addressable and its
interactions with other molecules can be assessed.
[0187] In a particular embodiment shown in FIG. 1, the substrate
contains amino groups that are blocked with a photolabile
protecting group. Amino acid sequences are made accessible for
coupling to a receptor by removal of the photoprotecting
groups.
[0188] When a polymer sequence to be synthesized is, for example, a
polypeptide, amino groups at the ends of linkers attached to a
glass substrate are derivatized with, for example,
nitroveratryloxycarbonyl (NVOC), a photoremovable protecting group.
The linker molecules may be, for example, aryl acetylene, ethylene
glycol oligomers containing from 2-10 monomers, diamines, diacids,
amino acids, or combinations thereof. Photodeprotection is effected
by illumination of the substrate through, for example, a mask
wherein the pattern has transparent regions with dimensions of, for
example, less than 1 cm.sup.2, 10.sup.-1 cm.sup.2, 10.sup.-2
cm.sup.2, 10.sup.-3 cm.sup.2, 10.sup.-4 cm.sup.2, 10.sup.-5
cm.sup.2, 10.sup.-6 cm.sup.2, 10.sup.-7 cm.sup.2, 10.sup.-8
cm.sup.2, or 10.sup.-10 cm.sup.2. In a preferred embodiment, the
regions are between about 10.times.10 .mu.m and 500.times.500
.mu.m. According to some embodiments, the masks are arranged to
produce a checkerboard array of polymers, although any one of a
variety of geometric configurations may be utilized.
[0189] 1. Example--Polymer Synthesis
[0190] FIG. 2 illustrates one embodiment of the invention disclosed
herein in which a substrate 8 is shown in cross-section.
Essentially, any conceivable substrate may be employed in the
invention. The substrate may be biological, nonbiological, organic,
inorganic, or a combination of any of these, existing as particles,
strands, precipitates, gels, sheets, tubing, spheres, containers,
capillaries, pads, slices, films, plates, slides, etc. The
substrate may have any convenient shape, such as a disc, square,
sphere, circle, etc. The substrate is preferably flat but may take
on a variety of alternative surface configurations. For example,
the substrate may contain raised or depressed regions on which the
synthesis takes place. The substrate and its surface preferably
form a rigid support on which to carry out the reactions described
herein. The substrate and its surface are also chosen to provide
appropriate light-absorbing characteristics. For instance, the
substrate may be a polymerized Langmuir Blodgett film,
functionalized glass, Si, Ge, GaAs, GaP, SiO.sub.2, SiN.sub.4,
modified silicon, any of the different crystal lattices made with
silicon or gallium arsenide that are commercially available and
used in semiconductor manufacturing, or any one of a wide variety
of gels or polymers such as (poly)tetrafluoroethylene,
(poly)vinylidenedifluoride, polystyrene, polycarbonate, or
combinations thereof. Other substrate materials will be readily
apparent to those of skill in the art upon review of this
disclosure. In a preferred embodiment the substrate is flat glass
or single-crystal silicon with surface relief features of less than
10 .ANG..
[0191] According to some embodiments, the surface of the substrate
is etched using well known techniques to provide for desired
surface features. For example, by way of the formation of trenches,
v-grooves, mesa structures, or the like, the synthesis regions may
be more closely placed within the focus point of impinging light,
be provided with reflective "mirror" structures for maximization of
light collection from fluorescent sources, or the like.
[0192] Surfaces on the solid substrate will usually, though not
always, be composed of the same material as the substrate. Thus,
the surface may be composed of any of a wide variety of materials,
for example, polymers, plastics, resins, polysaccharides, silica or
silica-based materials, carbon, metals, inorganic glasses,
membranes, or any of the above-listed substrate materials. In some
embodiments the surface may provide for the use of caged binding
members which are attached firmly to the surface of the substrate.
Preferably, the surface will contain reactive groups, which could
be carboxyl, amino, hydroxyl, or the like. Most preferably, the
surface will be optically transparent and will have surface Si--OH
functionalities, such as are found on silica surfaces.
[0193] The surface 10 of the substrate is preferably provided with
a layer of linker molecules 12, although it will be understood that
the linker molecules are not required elements of the invention.
The linker molecules are preferably of sufficient length to permit
polymers in a completed substrate to interact freely with molecules
exposed to the substrate. The linker molecules should be 6-50 atoms
long to provide sufficient exposure. The linker molecules may be,
for example, aryl acetylene, ethylene glycol oligomers containing
2-10 monomer units, diamines, diacids, amino acids, or combinations
thereof. Other linker molecules may be used in light of this
disclsoure.
[0194] According to alternative embodiments, the linker molecules
are selected based upon their hydrophilic/hydrophobic properties to
improve presentation of synthesized polymers to certain receptors.
For example, in the case of a hydrophilic receptor, hydrophilic
linker molecules will be preferred so as to permit the receptor to
more closely approach the synthesized polymer.
[0195] According to another alternative embodiment, linker
molecules are also provided with a photocleavable group at an
intermediate position. The photocleavable group is preferably
cleavable at a wavelength different from the protective group. This
enables removal of the various polymers following completion of the
synthesis by way of exposure to the different wavelengths of light.
One can also modify the linker molecule with a photocleavable
group, which, when removed, will induce a conformational change in
the polymer attached to the linker.
[0196] The linker molecules can be attached to the substrate via
carbon-carbon bonds using, for example,
(poly)trifluorochloroethylene surfaces, or preferably, by siloxane
bonds (using, for example, glass or silicon oxide surfaces).
Siloxane bonds with the surface of the substrate may be formed in
one embodiment via reactions of linker molecules bearing
trichlorosilyl groups. The linker molecules may optionally be
attached in an ordered array, i.e., as parts of the head groups in
a polymerized Langmuir Blodgett film. In alternative embodiments,
the linker molecules are adsorbed to the surface of the
substrate.
[0197] The linker molecules and monomers used herein are provided
with a functional group to which is bound a protective group.
Preferably, the protecting group is on the distal or terminal end
of the linker molecule opposite the substrate. The protecting group
may be either a negative protecting group (i.e., the protecting
group renders the linker molecules less reactive with a monomer
upon exposure) or a positive protecting group (i.e., the protecting
group renders the linker molecules more reactive with a monomer
upon exposure). In the case of negative protective groups an
additional step of reactivation will be required. In some
embodiments, this will be done by heating. Those of skill in the
art will also note that more than one functional group can be
employed on either the linker or the monomer, i.e., to facilitate
the synthesis of branched or "dendritic" structures.
[0198] The protecting group on the linker molecules may be selected
from a wide variety of positive light-reactive groups preferably
including nitro aromatic compounds such as o-nitrobenzyl
derivatives or benzylsulfonyl. In a preferred embodiment,
6-nitroveratryloxycarbonyl (NVOC), 2-nitrobenzyloxycarbonyl (NBOC)
or .alpha.,.alpha.-dimethyl-dimethoxybenz- yloxycarbonyl (DDZ) is
used. In one embodiment, a nitro aromatic compound containing a
benzylic hydrogen ortho to the nitro group is used, i.e., a
chemical of the form: 3
[0199] where R.sub.1 is alkoxy, alkyl, halo, aryl, alkenyl, or
hydrogen; R.sub.2 is alkoxy, alkyl, halo, aryl, nitro, or hydrogen;
R.sub.3 is alkoxy, alkyl, halo, nitro, aryl, or hydrogen; R.sub.4
is alkoxy, alkyl, hydrogen, aryl, halo, or nitro; and R.sub.5 is
alkyl, alkynyl, cyano, alkoxy, hydrogen, halo, aryl, or alkenyl.
Other materials which may be used include o-hydroxy-.alpha.-methyl
cinnamoyl derivatives.
[0200] In an alternative embodiment the positive reactive group is
activated for reaction with reagents in solution. For example, a
5-bromo-7-nitro indoline group, when bound to a carbonyl, undergoes
reaction upon exposure to light at 420 nm.
[0201] In a second alternative embodiment, the reactive group on
the linker molecule is selected from a wide variety of negative
light-reactive groups including a cinammate group.
[0202] Alternatively, the reactive group is activated or
deactivated by electron beam lithography, x-ray lithography, or any
other radiation. Suitable reactive groups for electron beam
lithography include sulfonyl. Other methods may be used including,
for example, exposure to a current source. Other reactive groups
and methods of activation may be used in light of this
disclosure.
[0203] As shown in FIG. 2, the linking molecules are preferably
exposed to, for example, irradiation, such as light, through a
suitable mask 14 using photolithographic techniques of the type
known in the semiconductor industry and described in, for example,
Sze, VLSI Technology, McGraw-Hill (1983), and Mead et al.,
Introduction to VLSI Systems, Addison-Wesley (1980), which are
incorporated herein by reference for all purposes. The light may be
directed at either the surface containing the protective groups or
at the back of the substrate, so long as the substrate is
transparent to the wavelength of light needed for removal of the
protective groups. In the embodiment shown in FIG. 2, light is
directed at the surface of the substrate containing the protective
groups. FIG. 2 illustrates the use of such masking techniques as
they are applied to a positive reactive group so as to activate
linking molecules and expose functional groups in areas 16a and
16b.
[0204] The mask 14 is in one embodiment a transparent support
material selectively coated with a layer of opaque material.
Portions of the opaque material are removed, leaving opaque
material in the precise pattern desired on the substrate surface.
The mask is brought into close proximity with, imaged on, or
brought directly into contact with the substrate surface as shown
in FIG. 2. "Openings" in the mask correspond to locations on the
substrate where it is desired to remove photoremovable protective
groups from the substrate. Alignment may be performed using
conventional alignment techniques in which alignment marks (not
shown) are used to accurately overlay successive masks with
previous patterning steps, or more sophisticated techniques may be
used. For example, interferometric techniques such as the one
described in Flanders et al., "A New Interferometric Alignment
Technique," App. Phys. Lett. (1977) 31:426-428, which is
incorporated herein by reference, may be used.
[0205] To enhance contrast of light applied to the substrate, it is
desirable to provide contrast enhancement materials between the
mask and the substrate according to some embodiments. This contrast
enhancement layer may comprise a molecule which is decomposed by
light such as quinone diazide or a material which is transiently
bleached at the wavelength of interest. Transient bleaching of
materials will allow greater penetration where light is applied,
thereby enhancing contrast. Alternatively, contrast enhancement may
be provided by way of a cladded fiber optic bundle.
[0206] The light may be from a conventional incandescent source, a
laser, a laser diode, or the like. If non-collimated sources of
light are used it may be desirable to provide a thick- or
multi-layered mask to prevent spreading of the light onto the
substrate. It may, further, be desirable in some embodiments to
utilize groups which are sensitive to different wavelengths to
control synthesis. For example, by using groups which are sensitive
to different wavelengths, it is possible to select branch positions
in the synthesis of a polymer or eliminate certain masking steps.
Several reactive groups along with their corresponding wavelengths
for deprotection are provided in Table 1.
2TABLE 1 Approximate Group Deprotection Wavelength Nitroveratryloxy
carbonyl (NVOC) UV (300-400 nm) Nitrobenzyloxy carbonyl (NBOC) UV
(300-350 nm) Dimethyl dimethoxybenzyloxy carbonyl UV (280-300 nm)
5-Bromo-7-nitroindolinyl UV (420 nm) o-Hydroxy-.alpha.-methyl
cinnamoyl UV (300-350 nm) 2-Oxymethylene anthraquinone UV (350
nm)
[0207] Note that different photoprotected monomers, such as amino
acids, can exhibit different photolysis rates. See, for example,
"The Peptides, Analysis, synthesis, Biology" Chapter 8, E. Gross
and J. Meienhofer, Eds., Academic Press, Inc. (1980); and PCT
application WO 89/10931. It may be desirable to utilize
photoprotected monomers with substantially similar photolysis rates
in a particular application. To obtain such a set of photoprotected
monomers, one merely needs to select the appropriate
photoprotecting group for each monomer in the set. In similar
fashion, one can prepare a set of photoprotected monomers with
substantially different photolysis rates (from monomer to monomer)
by appropriate choice of photoprotecting groups.
[0208] While the invention is illustrated primarily herein by way
of the use of a mask to illuminate selected regions the substrate,
other techniques may also be used. For example, the substrate may
be translated under a modulated laser or diode light source. Such
techniques are discussed in, for example, U.S. Pat. No. 4,719,615
(Feyrer et al.), which is incorporated herein by reference. In
alternative embodiments a laser galvanometric scanner is utilized.
In other embodiments, the synthesis may take place on or in contact
with a conventional liquid crystal (referred to herein as a "light
valve") or fiber optic light sources. By appropriately modulating
liquid crystals, light may be selectively controlled so as to
permit light to contact selected regions of the substrate.
Alternatively, synthesis may take place on the end of a series of
optical fibers to which light is selectively applied. Other means
of controlling the location of light exposure will be apparent to
those of skill in the art.
[0209] The substrate may be irradiated either in contact or not in
contact with a solution and is, preferably, irradiated in contact
with a solution. The solution contains reagents to prevent the
by-products formed by irradiation from interfering with synthesis
of the polymer according to some embodiments. Such by-products
might include, for example, carbon dioxide, nitrosocarbonyl
compounds, styrene derivatives, indole derivatives, and products of
their photochemical reactions. Alternatively, the solution may
contain reagents used to match the index of refraction of the
substrate. Reagents added to the solution may further include, for
example, acidic or basic buffers, thiols, substituted hydrazines
and hydroxylamines, reducing agents (e.g., NADH) or reagents known
to react with a given functional group (e.g., aryl
nitroso+glyoxylic acid.fwdarw.aryl formhydroxamate+CO.sub.2).
[0210] Either concurrently with or after the irradiation step, the
linker molecules are washed or otherwise contacted with a first
monomer, illustrated by "A" in regions 18a and 18b in FIG. 3. The
first monomer reacts with the activated functional groups of the
linker molecules which have been exposed to light. The first
monomer, which is preferably an amino acid, is also provided with a
photoprotective group. The photoprotecting group on the monomer may
be the same as or different than the protecting group used in the
linker molecules, and may be selected from any of the
above-described protective groups. In one embodiment, the
protective groups for the A monomer is selected from the group NBOC
and NVOC.
[0211] As shown in FIG. 4, the process of irradiating is thereafter
repeated, with a mask repositioned so as to remove linkage
protective groups and expose functional groups in regions 20a and
20b which are illustrated as being regions which were protected in
the previous masking step. As an alternative to repositioning of
the first mask, in many embodiments a second mask will be utilized.
In other alternative embodiments, some steps may provide for
illuminating a common region in successive steps. As shown in FIG.
4, it may be desirable to provide separation between irradiated
regions. For example, separation of about 1-5 .mu.m may be
appropriate to account for alignment tolerances.
[0212] As shown in FIG. 5, the substrate is then exposed to a
second protected monomer "B," producing B regions 22a and 22b.
Thereafter, the substrate is again masked so as to remove the
protective groups and expose reactive groups on A region 18a and B
region 18b. The substrate is again exposed to monomer B, resulting
in the formation of the structure shown in FIG. 6. The dimers B-A
and B-B have been produced on the substrate.
[0213] A subsequent series of masking and contacting steps similar
to those described above with A (not shown) provides the structure
shown in FIG. 7. The process provides all possible dimers of B and
A, i.e., B-A, A-B, A-A, and B-B.
[0214] The substrate, the area of synthesis, and the area for
synthesis of each individual polymer could be of any size or shape.
For example, squares, ellipsoids, rectangles, triangles, circles,
or portions thereof, along with irregular geometric shapes, may be
utilized. Duplicate synthesis areas may also be applied to a single
substrate for purposes of redundancy.
[0215] In some embodiments a single substrate supports more than
about 10 different monomer sequences and preferably more than about
100 different monomer sequences, although in some embodiments more
than about 10.sup.3, 10.sup.4, 10.sup.5, 10.sup.6, 10.sup.7, or
10.sup.8 different sequences are provided on a substrate. Of
course, within a region of the substrate in which a monomer
sequence is synthesized, it is preferred that the monomer sequence
be substantially pure. In some embodiments, regions of the
substrate contain polymer sequences which are at least about 1%,
5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 60%, 70%, 80%,
90%, 95%, 96%, 97%, 98%, or 99% pure.
[0216] According to some embodiments, several sequences are
intentionally provided within a single region so as to provide an
initial screening for biological activity, after which materials
within regions exhibiting significant binding are further
evaluated.
[0217] 2. Example
[0218] It is important to achieve good contrast between exposed and
non-exposed regions on a substrate. Otherwise, unwanted products
will be formed as, for example, when a monomer is added to a
polymer in an unexposed or dark region. It has been determined that
long exposure times will often result in more complete monomer
coupling in activated regions. However, if the substrate is exposed
for too long a period, the photolysis near the exposure edges will
decrease, i.e., the contrast between exposed and unexposed regions
will be reduced.
[0219] The deprotection of an NVOC-protected amine group was
employed to model masking resolution.
[0220] The deprotection of the NVOC protected amine by ultraviolet
light is a first order reaction of reactant A being converted to
product B: 1 A -> B A t = kA ln A A A 0 = ln A A 0 = - kt A A 0
= - kt = % NVOC protected amines after time t 1 - A A 0 = 1 - - kt
= % free amines at t
[0221] where k is rate of photolysis=(physical constants).times.I,
and I is light intensity at 365 nm=13 mw/cm.sup.2. A is the
concentration of the reactant and A.sub.0 is that concentration at
t=0.
[0222] The dark areas were modelled as if they were being
photolyzed with a fraction of the light intensity and a new rate of
photolysis to the areas beyond the photolysis site was defined. 2 I
' = CI k ' = Ck r = ( 1 - A ' A 0 ' ) = F A F B = 1 - - k ' t k ' =
- ln ( 1 - r ) t
[0223] F is the height of the histogram used to analyze the
experimental results. Thus, F is also the florescence intensity at
a distance from the edge of photolysis.
[0224] Contrast was investigated by photolysis through a binary
mask (12800 .mu.m.times.6400 .mu.m) for 660, 1320, and 9990
seconds. The contrast ratio was measured (as function of distance
from the photolysis edge) as the ratio of the work height of the
histogram in the dark area to the height of the histogram in the
light area, i.e. FA/FB. The results are presented below.
3 N time 0 .mu.M 50 .mu.M 100 .mu.M 200 .mu.M 1 660 sec. 1 0.4 0.1
0.01 2 1320 sec. 1 0.7 0.3 0.03 3 9900 sec. 1 0.92 0.7 0.39
[0225] Thus, it can be seen that the photolysis fidelity is a
function of both the time of exposure and the distance from the
edge.
[0226] 3. Example--Slide Preparation
[0227] Before attachment of reactive groups it is preferred to
clean the substrate which is, in a preferred embodiment, a glass
substrate such as a microscope slide or cover slip. According to
one embodiment the slide is soaked in an alkaline bath consisting
of, for example, 1 liter of 95% ethanol with 120 ml of water and
120 grams of sodium hydroxide for 12 hours. The slides are then
washed under running water and allowed to air dry, and rinsed once
with a solution of 95% ethanol.
[0228] The slides are then aminated with, for example,
aminopropyltriethoxysilane for the purpose of attaching amino
groups to the glass surface on linker molecules, although any omega
functionalized silane could also be used for this purpose. In one
embodiment 0.1% aminopropyltriethoxysilane is utilized, although
solutions with concentrations from 10.sup.-7% to 10% may be used,
with about 10.sup.-3% to 2% preferred. A 0.1% mixture is prepared
by adding to 100 ml of a 95% ethanol/5% water mixture, 100
microliters (.mu.l) of aminopropyltriethoxysilane. The mixture is
agitated at about ambient temperature on a rotary shaker for about
5 minutes. 500 .mu.l of this mixture is then applied to the surface
of one side of each cleaned slide. After 4 minutes, the slides are
decanted of this solution and rinsed three times by dipping in, for
example, 100% ethanol.
[0229] After the plates dry, they are placed in a 110-120.degree.
C. vacuum oven for about 20 minutes, and then allowed to cure at
room temperature for about 12 hours in an argon environment. The
slides are then dipped into DMF (dimethylformamide) solution,
followed by a thorough washing with dichloromethane.
[0230] The aminated surface of the slide is then exposed to about
500 .mu.l of, for example, a 30 millimolar (mM) solution of
NVOC-GABA (gamma amino butyric acid) NHS (N-hydroxysuccinimide) in
DMF for attachment of a NVOC-GABA to each of the amino groups.
[0231] The surface is washed with, for example, DMF,
dichloromethane, and ethanol.
[0232] Any unreacted aminopropyl silane on the surface--that is,
those amino groups which have not had the NVOC-GABA attached--are
now capped with acetyl groups (to prevent further reaction) by
exposure to a 1:3 mixture of acetic anhydride in pyridine for 1
hour. Other materials which may perform this residual capping
function include trifluoroacetic anhydride, formicacetic anhydride,
or other reactive acylating agents. Finally, the slides are washed
again with DMF, methylene chloride, and ethanol.
[0233] 4. Example--Synthesis of a Dimer of an Aminopropyl Group and
a Fluorescent Group
[0234] In synthesizing the dimer of an aminopropyl group and a
fluorescent group, a functionalized Durapore.TM. membrane was used
as a substrate. The Durapore.TM. membrane was a polyvinylidine
difluoride with aminopropyl groups. The aminopropyl groups were
protected with the DDZ group by reaction of the carbonyl chloride
with the amino groups, a reaction readily known to those of skill
in the art. The surface bearing these groups was placed in a
solution of THF and contacted with a mask bearing a checkerboard
pattern of 1 mm opaque and transparent regions. The mask was
exposed to ultraviolet light having a wavelength down to at least
about 280 nm for about 5 minutes at ambient temperature, although a
wide range of exposure times and temperatures may be appropriate in
various embodiments of the invention. For example, in one
embodiment, an exposure time of between about 1 and 5000 seconds
may be used at process temperatures of between -70 and +50.degree.
C.
[0235] In one preferred embodiment, exposure times of between about
1 and 500 seconds at about ambient pressure are used. In some
preferred embodiments, pressure above ambient is used to prevent
evaporation.
[0236] The surface of the membrane was then washed for about 1 hour
with a fluorescent label which included an active ester bound to a
chelate of a lanthanide. Wash times will vary over a wide range of
values from about a few minutes to a few hours. These materials
fluoresce in the red and the green visible region. After the
reaction with the active ester in the fluorophore was complete, the
locations in which the fluorophore was bound could be visualized by
exposing them to ultraviolet light and observing the red and the
green fluorescence. It was observed that the derivatized regions of
the substrate closely corresponded to the original pattern of the
mask.
[0237] 5. Example--Removal of NVOC and Attachment of a Fluorescent
Marker
[0238] NVOC-GABA groups were attached as described above, except
that the substrate was a glass slide. The entire surface of one
slide was exposed to light so as to expose a free amino group at
the end of the gamma amino butyric acid. This slide, and a
duplicate which was not exposed, were then exposed to fluorescein
isothiocyanate (FITC).
[0239] FIG. 8A illustrates the slide which was not exposed to
light, but which was exposed to FITC. Fluorescence on the surface
was measured by excitation using 488 nm laser light and
photomultiplier detection through appropriate fluuorescein emission
filters described in additional detail below. The units of the x
axis are time (msec) and the units of the y axis are counts. The
trace contains a certain amount of background fluorescence. The
duplicate slide was exposed to 350 nm broadband illumination for
about 1 minute (12 mW/cm.sup.2, .about.350 nm illumination), washed
and reacted with FITC. The fluorescence curves for this slide are
shown in FIG. 8B. A large increase in the level of fluorescence is
observed, which indicates photolysis has exposed a number of amino
groups on the surface of the slides for attachment of a fluorescent
marker.
[0240] 6. Example--Use of a Mask in Removal of NVOC
[0241] The next experiment was performed with a 0.1%
aminopropylated slide. Light from a Hg--Xe arc lamp was imaged onto
the substrate through a laser-ablated chrome-on-glass mask in
direct contact with the substrate.
[0242] This slide was illuminated for approximately 5 minutes, with
12 mW of 350 nm broadband light and then reacted with the 1 mM FITC
solution. It was put on the laser detection scanning stage and a
graph was plotted as a two-dimensional representation of position
versus fluorescence intensity. The fluorescence intensity (in
counts) as a function of location is given on the scale to the
right of FIG. 9A for a mask having 100.times.100 .mu.m squares.
[0243] The experiment was repeated a number of times through
various masks. The fluorescence pattern for a 50 .mu.m mask is
illustrated in FIG. 9B, for a 20 .mu.m mask in FIG. 9C, and for a
10 .mu.m mask in FIG. 9D. The mask pattern is distinct down to at
least about 10 .mu.m squares using this lithographic technique.
[0244] 7. Example
[0245] In one example of the invention, free amino groups were
fluorescently labelled by treatment of the entire substrate surface
with fluorescein isothiocyanate (FITC) after photodeprotection.
Glass microscope slides were cleaned, aminated by treatment with
0.1% aminopropyltriethoxysilane in 95% ethanol, and incubated at
110.degree. C. for 20 min. The aminated surface of the slide was
then exposed to a 30 mM solution of the N-hydroxysuccinimide ester
of NVOC-GABA (nitroveratryloxycarbonyl-.tau.-amino butyric acid) in
DMF. The NVOC protecting group was photolytically removed by
imaging the 365 nm output from a Hg arc lamp through a chrome on
glass 100 .mu.m checkerboard mask onto the substrate for 20 min at
a power density of 12 mW/cm.sup.2. The exposed surface was then
treated with 1 mM FITC in DMF. The substrate surface was scanned in
an epifluorescence microscope (Zeiss Axioskop 20) using 488 nm
excitation from an argon ion laser (Spectra-Physics model 2025).
The fluorescence emission above 520 nm was detected by a cooled
photomultiplier (Hamamatsu 943-02) operated in a photon counting
mode. Fluorescence intensity was translated into a color display
with red in the highest intensity and black in the lowest intensity
areas. The presence of a high-contrast fluorescent checkerboard
pattern of 100.times.100 .mu.m elements revealed that free amino
groups were generated in specific regions by spatially-localized
photodeprotection.
[0246] 8. Example
[0247] Slide preparation is illustrated below. Slides used in
synthesis may be detergent cleaned, glass slides. Such glass slides
may be, for example, 1".times.3" smooth cut, 0.7 mm thick,
anti-scratch coated, or 2".times.3" smooth cut, 0.7 mm thick,
anti-scratch coated from Erie Scientific. The slides are soaked in
10% Micro.TM. detergent (from Baxter), individually scrubbed, and
immersed in deionized 1H.sub.2O until all slides have been
scrubbed. The slides are then subjected to 10 minute sonication in
70.degree. C. "Micro" detergent and rinsed 10.times. with deionized
H.sub.2O. This process is followed by a 3 minute immersion in
70.degree. C. 10% (w/v) NaOH. The slides are then rinsed 10.times.
with deionized H.sub.2O, followed by a 1 minute immersion in 1%
HCl. The slides are then again rinsed 10.times. with deionized
H.sub.2O, followed by another 10 minute sonication in 70.degree. C.
deionized water and are rinsed 3-4.times. in deionized H.sub.2O.
The slides are then ethanol rinsed and dried with nitrogen or
argon. The slides are then inspected visually for spots and
scratches, preferably in a yellow light with a black
background.
[0248] Alternatively or in addition, the slides are acid cleaned.
The slides are loaded into teflon racks and subjected to a 30
minute immersion in Nochromix.TM. (Aldrich) solution with 36 g per
liter of concentrated H.sub.2SO.sub.4, which is regenerated if
discolored, filtered (glass fiber filter) to remove particulate
matter, and provided with occasional agitation. The slides are then
rinsed for 1 min. in deionized H.sub.2O with vigorous agitation.
The slides are then placed for 10 minutes in a rinse tank with 14
psi argon or nitrogen bubbling, a full open deionized water tap,
and occasional agitation.
[0249] The slides are then immersed for 3 minutes in 70.degree. C.
10% (w/v) NaOH, followed by a 1 minute rinse deionized H.sub.2O
with vigorous agitation, followed by 10 minutes in a rinse tank.
The slides are then immersed for 1 min. in 1% HCl, and rinsed for 5
minutes in a rinse tank. The slides are then ethanol rinsed, dried
with nitrogen or argon, and inspected visually for spots and
scratches.
[0250] tBOC aminopropyl derivatization is illustrated below. The
slides are loaded into plastic staining jars. Preferably the slides
are completely dry, with 9 slides per jar. Silation reagents are
then mixed as follows:
[0251] a pre-mix 1:10 mole ratio of
tBocaminoprbpyltriethoxysilane:methylt- riethoxysilane
[0252] tBOC-aminopropyltriethoxysilane:
[0253] MW=321.49
[0254] d=0.945 g/ml; and
[0255] methyltriethoxysilane:
[0256] NW=178.30
[0257] d=0.895 g/ml
[0258] 1:10 ratio=1 ml tBOC-aminopropyl to 5.86 ml
methyltriethoxysilane
[0259] The reagents are kept anhydrous and stored under argon. The
silation reagent is diluted to 1% (v/v) in dichloromethane (DCM),
mixed well, and 60 ml per jar are added. The jars are capped and
left overnight.
[0260] The silation solution is poured into a plastic container,
rinsed with dichloromethane (DCM), and the slides are rinsed with
toluene. The toluene is then poured off, and the slides are dried
immediately with argon. The slides are loaded into glass drying
racks, inspected for streaks, and allowed to stand for
approximately 30 minutes.
[0261] The slides are then baked for 1 hour in 100.degree. C. oven
with the glass racks in metal trays covered with foil. The oven is
preferably no hotter than 110.degree. C. The slides are then cooled
and numbered using an engraving tool.
[0262] Aminocaproic acid coupling is illustrated below. The
tBOC-aminopropyl slides are loaded into glass staining jars with 15
slides per jar. The slides are then deprotected and neutralized
with a 30 minute immersion in 50% TFA/DCM, a rinse for 2 minutes in
DCM, and a rinse 2.times. in 5% DIEA/DCM for 5 minutes each,
followed by a rinse with dichloromethane, and a rinse with ethanol.
The slides are then dried with argon, and derivatized within one
hour.
[0263] The volume of solution necessary is equal to 0.4 ml.times.#
slides. The concentration of the solution is 100 mM
NVOC-aminocaproic acid, 110 mM HOBT, 200 mM DIEA, and 100 mM
BOC.
[0264] The slides are placed face up on glass plates, and 0.4 ml
solution is layered per slide. The slides are then covered in
plastic trays and allowed to sit for 2 hours. The slides are then
rinsed with DMF or NMP, rinsed with DCM, and rinsed with EtOH.
[0265] The slides are capped with acetic anhydride by immersion for
1 hour in 25% acetic anhydride/pyridine and 0.1% DMAP. The slides
are then rinsed with DMF or NMP, rinsed with DCM, and rinsed with
EtOH. The slides are then dried with argon and stored in a
light-tight container.
[0266] Biotinylation of NVOC-aminocaproic slides may be desirable
in some instances (for example, see infra Section V.C.) and is
achieved as follows. The slides are photolyzed in 5 mM
H.sub.2SO.sub.4/dioxane with appropriate masking and a large area
mercury illuminator with a 350-450 nm dichroic reflector and a 12
minute exposure at 12-13 mW/cm.sup.2.
[0267] The slides are collected in dioxane until all slides have
been exposed, washed 2.times. in 5% DIEA/DMF for 5 minutes each,
and rinsed with DMF, DCM, and EtOH. The slides are then dried with
argon, and are preferably derivatized within one hour.
[0268] The volume of solution necessary is equal to 0.4 ml.times.#
slides. The concentration of the solutions is 100 mM Biotin, 110 mM
HOBT, and 200 mM DIEA and 100 mM BOC. A heat gun is used to help
dissolve biotin. 100 mM BOP (MW=442.29) is dissolved in {fraction
(1/10)}th final volume NMP or DMF. The solutions are mixed, capped,
and allowed to stand 10 minutes. The final volume is adjusted with
NMP or DMF and the solutions are mixed well.
[0269] The slides are placed face up on glass plates, and 0.4 ml
solution per slide is layered onto the slide. The slides are
covered in plastic trays, allowed to sit for 2 hours, and rinsed
with NMP or DMF, then rinsed with DCM, then rinsed with EtOH, and
dried with argon.
[0270] FITC labeling of amines is achieved as follows. First the
amines are deprotected and neutralized by photolyzing
NVOC-aminocaproic slides. The photolysis takes place in 5 mM
H.sub.2SO.sub.4/dioxane with appropriate masking using a large area
illuminator with a 350-450 nm dichroic reflector, 12-13
mW/cm.sup.2, and a 12 minute exposure. The slides are rinsed in
dioxane, and then rinsed 2.times. in 5% DIEA/DMF; 5 minutes each.
The slides are then rinsed with DMF.
[0271] Deprotected amines are labeled by immersing slides in 1 mM
FITC/DMF for 1 hour, rinsing with DMF, rinsing with DCM, rinsing
with ethanol, and drying slides with argon.
[0272] B. Antibody Recognition
[0273] In one preferred embodiment the substrate is used to
determine which of a plurality of amino acid sequences is
recognized by an antibody of interest.
[0274] For purposes of increasing the signal-to-noise ratio of the
technique, some embodiments of the invention provide for exposure
of the substrate to a first labeled or unlabeled receptor followed
by exposure of a labeled, second receptor (e.g., an antibody) which
binds at multiple sites on the first receptor. If, for example, the
first receptor is an antibody derived from a first species of an
animal, the second receptor is an antibody derived from a second
species directed to epitopes associated with the first species. In
the case of a mouse antibody, for example, fluorescently labeled
goat antibody or antiserum which is antimouse may be used to bind
at multiple sites on the mouse antibody, providing several times
the fluorescence compared to the attachment of a single mouse
antibody at each binding site. This process may be repeated again
with additional antibodies (e.g., goat-mouse-goat, etc.) for
further signal amplification.
[0275] 1. Example--Attachment of YGGFL and Subsequent Exposure to
Herz Antibody and Goat Antimouse
[0276] In order to establish that receptors to a particular
polypeptide sequence would bind to a surface-bound peptide and be
detected, t-BOC protected Leu enkephalin was coupled to the surface
and recognized by an antibody. A slide was derivatized with 0.1%
amino propyl-triethoxysilane and protected with NVOC-6-amino
caproic acid or NVOC-GABA. A 500 .mu.m checkerboard mask was used
to expose the slide in a flow cell using backside contact printing.
The Leu enkephalin sequence
(H.sub.2N-tyrosine,glycine,glycine,phenylalanine,leucine-CO.sub.2H,
otherwise referred to herein as YGGFL (SEQ. ID NO:1)) was attached
via its carboxy end to the exposed amino groups on the surface of
the slide. The peptide was added in DMF solution with the
BOP/HOBT/DIEA coupling reagents and recirculated through the flow
cell for 2 hours at room temperature.
[0277] A first antibody, known as the Herz antibody, was applied to
the surface of the slide for 45 minutes at 2 .mu.g/ml in a
supercocktail (containing 1% BSA and 1% ovalbumin also in this
case). A second antibody, goat anti-mouse fluorescein conjugate,
was then added at 2 .mu.g/ml in the supercocktail buffer, and
allowed to incubate for 2 hours.
[0278] The results of this experiment are obtained by taking a
fluorescence scan obtained using a fluorescence detection system.
Again, FIG. 10 illustrates fluorescence intensity as a function of
position. The fluorescence scale is shown on the right. This image
was taken at 10 .mu.m steps. This figure indicates that not only
can deprotection be carried out in a well defined pattern, but also
that (1) the method provides for successful coupling of peptides to
the surface of the substrate, (2) the surface of a bound peptide is
available for binding with an antibody, and (3) that the detection
apparatus capabilities are sufficient to detect binding of a
receptor.
[0279] 2. Example
[0280] FIG. 11 is a flow chart illustrating another example of the
invention. Carboxy-activated NVOC-leucine was allowed to react with
an aminated substrate. The carboxy activated HOBT ester of leucine
and other amino acids used in this synthesis was formed by mixing
0.25 mmol of the NVOC amino protected amino acid with 37 mg HOBT
(1-hydroxybenzotriazole), 111 mg BOP
(benzotriazolyl-n-oxy-tris(dimethylamino)-phosphoniumhexa-fluo-
rophosphate) and 86 .mu.l DIEA (diisopropylethylamine) in 2.5 ml
DMF. The NVOC protecting group was removed by uniform illumination.
Carboxy-activated NVOC-phenylalanine was coupled to the exposed
amino groups for 2 hours at room temperature, and then washed with
DMF and methylene chloride. Two unmasked cycles of
photodeprotection and coupling with carboxy-activated NVOC-glycine
were carried out. The surface was then illuminated through a chrome
on glass 50 .mu.m checkerboard pattern mask. Carboxy-activated
N.alpha.-tBOC-O-tButyl-L-tyrosine was then added. The entire
surface was uniformly illuminated to photolyze the remaining NVOC
groups. Finally, carboxy-activated NVOC-L-proline was added, the
NVOC group was removed by illumination, and the t-BOC and t-butyl
protecting groups were removed with TFA. After removal of the
protecting groups, the surface consisted of a 50 .mu.m checkerboard
array of Tyr-Gly-Gly-Phe-Leu (YGGFL) and Pro-Gly-Gly-Phe-Leu
(PGGFL). See also SEQ ID NO:1 and SEQ ID NO:2.
[0281] The array of pentapeptides was probed with a mouse
monoclonal antibody directed against .beta.-endorphin. This
antibody (called 3E7) is known to bind YGGFL and YGGFM (see also
SEQ ID NO:1 and SEQ ID NO:21) with nanomolar affinity and is
discussed in Meo et al., Proc. Natl. Acad. Sci. USA (1983) 80:4084,
which is incorporated by reference herein for all purposes. This
antibody requires the amino terminal tyrosine for high affinity
binding. The array of peptides formed as described in FIG. 11 was
incubated with a 2 .mu.g/ml mouse monoclonal antibody (3E7) known
to recognize YGGFL. See also SEQ ID NO:1. 3E7 does not bind PGGFL.
See also SEQ ID NO:2. A second incubation with fluoresceinated goat
anti-mouse antibody labeled the regions that bound 3E7. The surface
was scanned with an epi-fluorescence microscope. As shown in FIG.
12, results showed alternating bright and dark 50 .mu.m squares
indicating that YGGFL (SEQ ID NO:1) and PGGFL (SEQ ID NO:2) were
synthesized in a geometric array determined by the mask. A high
contrast (>12:1 intensity ratio) fluorescence checkerboard image
shows that (a) YGGFL (SEQ ID NO:1) and PGGFL (SEQ ID NO:2) were
synthesized in alternate 50 .mu.m squares, (b) YGGFL (SEQ ID NO:1)
attached to the surface is accessible for binding to antibody 3E7,
and (c) antibody 3E7 does not bind to PGGFL (SEQ ID NO:2).
[0282] A three-dimensional representation of the fluorescence
intensity data in a 2 square by 4 square rectangular portion of the
checkerboard was also produced. It showed that the border between
synthesis sites is sharp. The height of each spike in this display
is linearly proportional to the integrated fluorescence intensity
in a 2.5 .mu.m pixel. The transition between PGGFL and YGGFL occurs
within two spikes (5 .mu.m). There is little variation in the
fluorescence intensity of different YGGFL squares. The mean
intensity of sixteen YGGFL synthesis sites was 2.03.times.10.sup.5
counts and the standard deviation was 9.6.times.10.sup.3
counts.
[0283] 3. Example--Monomer-by-Monomer Formation of YGGFL and
Subsequent Exposure to Labeled Antibody
[0284] Monomer-by-monomer synthesis of YGGFL and GGFL in alternate
squares was performed on a slide in a checkerboard pattern and the
resulting slide was exposed to the Herz antibody. This experiment
and the results thereof are illustrated in FIGS. 13A, 13B, 13C, and
13D.
[0285] In FIG. 13A, a slide is shown which is derivatized with
t-BOC-aminopropyl-triethoxysilane. The slide was treated with TFA
to remove the t-BOC protecting group. t-BOC-6-aminocaproic acid was
then coupled onto the aminopropyl groups. The aminocaproic acid
serves as a spacer between the aminopropyl group and the peptide to
be synthesized. The amino end of the spacer was deprotected and
coupled to NVOC-Leucine. The entire slide was then illuminated with
12 mW of 325 nm broadband illumination. The slide was then coupled
with NVOC-Phenylalanine and washed. The entire slide was again
illuminated, then coupled to NVOC-Glycine and washed. The slide was
again illuminated and coupled to NVOC-Glycine to form the sequence
shown in the last portion of FIG. 13A.
[0286] As shown in FIG. 13B, alternating regions of the slide were
then illuminated using a projection print with a 500.times.500
.mu.m checkerboard mask; thus, the amino group of Glycine was
exposed only in the lighted areas. When the next coupling chemistry
step was carried out, NVOC-Tyrosine was added, and it coupled only
at those spots which had received illumination. The entire slide
was then illuminated to remove all the NVOC groups, leaving a
checkerboard of YGGFL in the lighted areas and in the other areas,
GGFL (SEQ. ID NO:15). The Herz antibody (which recognizes the
YGGFL, but not GGFL) was then added, followed by goat anti-mouse
fluorescein conjugate.
[0287] The resulting fluorescence scan is shown in FIG. 13C, and
the scale for the fluorescence intensity is again given on the
right. Dark areas contain the tetrapeptide GGFL, which is not
recognized by the Herz antibody (and thus there is no binding of
the goat anti-mouse antibody with fluorescein conjugate), and in
the lightly shaded areas YGGFL is present. The YGGFL pentapeptide
is recognized by the Herz antibody and, therefore, there is
antibody in the lighted regions for the fluorescein-conjugated goat
anti-mouse to recognize.
[0288] Similar patterns are shown for a 50 .mu.m mask used in
direct contact ("proximity print") with the substrate in FIG. 13D.
Note that the pattern is more distinct and the corners of the
checkerboard pattern are touching when the mask is placed in direct
contact with the substrate (which reflects the increase in
resolution using this technique).
[0289] 4. Example--Monomer-by-Monomer Synthesis of YGGFL and
YPGGFL
[0290] In order to further demonstrate the operability of the
invention, a 50 .mu.m checkerboard pattern of alternating YGGFL and
YPGGFL (SEQ. ID NO:3) was synthesized on a substrate using
techniques like those set forth above. The resulting fluorescence
plot is provided in FIG. 14. Again, it is seen that the antibody is
clearly able to recognize the YGGFL sequence and does not bind
significantly at the YPGGFL regions.
[0291] 5. Example--Synthesis of an Array of Sixteen Different Amino
Acid Sequences and Estimation of Relative Binding Affinity to Herz
Antibody
[0292] Using techniques similar to those set forth above, an array
of 16 different amino acid sequences (replicated four times) was
synthesized on each of two glass substrates. The sequences were
synthesized by attaching the sequence NVOC-GFL across the entire
surface of the slides. Using a series of masks, two layers of amino
acids were then selectively applied to the substrate. Each region
had dimensions of 0.25 cm.times.0.0625 cm. The first slide
contained amino acid sequences containing only L amino acids while
the second slide contained selected D amino acids. FIGS. 15A and
15B illustrate a map of the various regions on the first and second
slides, respectively. The patterns shown in FIGS. 15A and 15B were
duplicated four times on each slide. The slides were then exposed
to the Herz antibody and fluorescein-labeled goat anti-mouse.
[0293] FIG. 16 is a fluorescence plot of the first slide, which
contained only L amino acids. Light shading indicates strong
binding (149,000 counts or more) while black indicates little or no
binding of the Herz antibody (20,000 counts or less). The bottom
right-hand portion of the slide appears "cut off" because the slide
was broken during processing. The sequence YGGFL is clearly most
strongly recognized. The sequences YAGFL and YSGFL also exhibit
strong recognition of the antibody. By contrast, most of the
remaining sequences show little or no binding. The four duplicate
portions of the slide are extremely consistent in the amount of
binding shown therein.
[0294] FIG. 17 is a fluorescence plot of the second slide. Again,
strongest binding is exhibited by the YGGFL sequence. Significant
binding is also detected to YaGFL (SEQ. ID NO:22), YsGFL (SEQ. ID
NO:23), and YpGFL (SEQ. ID NO:24). The remaining sequences show
less binding with the antibody. Note the low binding efficiency of
the sequence yGGFL.
[0295] Table 2 lists the various sequences tested in order of
relative fluorescence, which provides information regarding
relative binding affinity. In the table, lower case letters
represent D-amino acids.
4TABLE 2 Apparent Binding to Herz Ab L-a.a. Set D-a.a. Set YGGFL
YGGFL YAGFL YaGFL YSGFL YsGFL LGGFL (SEQ. ID NO: 25) YpGFL FGGFL
(SEQ. ID NO: 26) fGGFL YPGFL yGGFL LAGFL (SEQ. ID NO: 27) faGFL
FAGFL (SEQ. ID NO: 28) wGGFL WGGFL (SEQ. ID NO: 29) yaGFL fpGFL
waGFL
[0296] 6. Example
[0297] A 4096 compound experiment was conducted similarly to
detailed Example 8. The building blocks used were: Y, G, P, A, F,
W, G, F, M, Q, L, and S. Since G and F are repeated there are 4072
peptides. This chip was stained with Herz 3E7 IgG and then
FITC-labelled goat anti-mouse IgG. A substrate used in this
experiment was 700 .mu.m thick. The results are shown in FIG.
18.
[0298] 7. Example
[0299] In order to generate 65,536 different compounds (including
one null compound) in a minimum number of chemical steps, a sixteen
step binary masking strategy was used. The building blocks chosen
were (from amino to carboxy): r, R, H, Q, P, F, Homophenylalanine,
N, Ornithine, A, V, v, T, S, and G (lower case letters represent
D-amino acids). This experiment shows that unnatural amino acids
can be used as building blocks.
[0300] Once the masking strategy and the building blocks were
chosen, the amino acids were weighed into cartridges obtained from
ABI (Applied Biosystems Inc., Foster City, Calif.). For this
experiment the flow cell volume was 0.5 ml, and thus 15 mg of HOBT
and 44 mg of BOP were used. During synthesis the amino acid was
dissolved in 1.5 ml of solvent so that the flow cell was full of
amino acid solution during coupling. The final amino acid
concentration was 60 mM.
[0301] Next a process file was generated on the program "PS" (copy
provided in Appendix 3). This was done by hitting F1 (if an IBM
computer was being used to run the program) to "initialize masking
sequence." "Binary process minimum movement" was chosen and then
the program asked for input the building blocks in order of C
terminus to N terminus. The first building block that was input
that goes onto the chip first, in this case S. The program allows
for input of the names in either one letter or three letter codes.
Using a binary process with minimum movement, one does not have to
select the mask that will be used, as the program will select the
mask. For a sixteen-step binary synthesis the following masks are
used in the order given: mask A offset 0, mask A offset 1, mask B
offset 0, mask B offset 1, mask C offset 0, mask C offset 1, mask D
offset 0, mask D offset 1, mask E offset 0, mask E offset 1, mask F
offset 0, mask F offset 1, mask G offset 0, mask G offset 1, mask H
offset 0, and mask H offset 1. The masking sequence was then saved
to disk so that it could be used during data workup.
[0302] The exposure lamp was turned on and the shutter timer set
for 11 minutes. The lamp power was set to about 12 mW/cm.sup.2 at
365 nm. The lamp was warmed up for about an hour before the
experiment began. Meanwhile, the mask was cleaned with methanol and
the mirror was aligned.
[0303] Next, the ABI peptide synthesizer was set up. Of course,
other peptide synthesizers can be used, e.g., one commercially
available from Milligen, Inc. All of the reagent bottles were
filled and an empty cartridge was placed under the injector. The
amino acid cartridges were then loaded from the N-terminus on the
left to the C-terminus on the right. The synthesizer modules to be
used were then entered through synthesizer interface. In this case
the following sequence was used: HEBDCDFDCCD. This sequence is
repeated sixteen times for a sixteen step synthesis.
[0304] H is the photolysis module. The slide was rinsed with
p-dioxane and then rinsed twice in 5 mM H.sub.2SO.sub.4 in
p-dioxane. The flow cell was then filled with 5 mM H.sub.2SO.sub.4
in p-dioxane for photolysis. A relay was sent to the personal
computer telling it to open the lamp shutter. The shutter opens and
remains open for 11 minutes. During this time a 5% DIEA in DMF
solution was made in the peptide synthesizer's activator vessel.
Module E begins before the photolysis is completed. The first part
of module E starts the dissolution of the amino acid in the
cartridge. A 7% DIEA/DMF solution was delivered to the cartridge
and the solution was mixed by Argon bubbling. The cartridge wass
mixed for about six minutes. Next the cartridge solution was
further diluted with anhydrous DMF and mixed some more. By this
point the photolysis was completed.
[0305] Module B, the substrate activation module, first rinsed the
slide with p-dioxane six times. Then it rinsed the slide with DMF.
Next, an aliquot of the 5% DIEA/DMF solution made in the activator
vessel was moved to the flow cell where it sat for 100 seconds.
This step was repeated six times and the flow cell was drained.
[0306] Module D washed the slide with DMF and washed the slide with
dichloromethane followed by ethanol. Hence, the slide was washed
with DMF, dichloromethane and ethanol, and finally DMF again.
[0307] Next chemical coupling occurred using module F. The amino
acid solution was taken from the cartridge and put into the flow
cell where it sat for 1.5 hours. After coupling was completed, the
slide was then washed using modules D, C, C, and D. This is the end
of one synthesis cycle. As mentioned above, this sequence of
modules was repeated fifteen more times for a sixteen step
synthesis.
[0308] In use, the flow cell was set up and a substrate was chosen.
In this case a 160 .mu.m thick slide derivatized with an
NVOC-6-aminocaproic acid linker was used. The slide was placed on
the flow cell and the vacuum was turned on. The flow cell transfer
lines were attached to the synthesizer. The slide and flow cell
were checked for leaks using a methylene chloride wash. This also
served to rinse the slide. Next, the outer surface of the slide was
cleaned with methanol. The flow cell was then attached to the
synthesis mount and placed flush against the mask.
[0309] The synthesis then began. "Begin synthesis" was pressed on
"PS," (copy provided in Appendix 3) with the exposure time set to
660 seconds. After the mask moved to its first position, the ABI
was started. This synthesis took about 48 hours, since each cycle
was three hours.
[0310] Once the synthesis was completed, the slide underwent a
final photolysis to remove all of the NVOC groups on the slide.
Modules HIBDCD were used, with I a "wait" step. Since it is
desirable to photolyze the entire slide, the flow cell with the
attached substrate was taken off of the synthesis mount and
physically placed under the lamp. After the photolysis was
complete, it was put back so that the remaining modules would go
smoothly. The flow cell was in a vertical position to ensure total
coverage of the substrate with solutions.
[0311] The amino groups were capped by final photolysis with acetic
anhydride. This process is called acetylation.
[0312] After final photolysis and capping, the side chain
protecting groups on the amino acids were removed. The slide was
taken off of the flow cell and treated with a trifluoracetic acid
solution containing phenol, thioanisole, and ethanedithiol as
scavengers. After side-group deprotection the slide was neutralized
in a 5% DIEA/methylene chloride solution twice for five minutes
each. The slide was then rinsed with methylene chloride, DMF, and
ethanol.
[0313] Next, the slide was incubated with 3 ml of anti-dynorphin B
antibody (8 micrograms/ml) and 1% BSA in PBS (containing 0.08%
Tween 20.TM.) for two hours. After rinsing twice with PBS, the
slide was stained with FITC-labelled goat anti-mouse antibody (10
micrograms/ml) in 1% BSA/PBS for 1.5 hours. After the second
staining, the slide was rinsed twice with PBS/Tween 20 and once
with deionized water.
[0314] Next the slide was scanned using the fluorescence detection
system. Scanning parameters depend on the type of image being
scanned. In this case since each synthesis site is only 50 .mu.m,
the slide was scanned very slowly with a small increment size.
Typical parameters were 3000.times.3000 @5 .mu.m steps, 5 .mu.m/ms,
220 .mu.m/ms.sup.2.
[0315] Once the fluorescence images were obtained, the data file
was converted to a tiff file and analyzed using a program called
"avi" (attached as Appendix 1) which has a module that integrates
each synthesis site. It then made a file containing fluorescence
versus location information. Then another program called "pepserch"
(attached as Appendix 2) was used to combine the fluorescence
information with compound identity. In this experiment the largest
peptide synthesis was 16 amino acids in length, and 65,536
(including the null) peptides (including monomers) were
synthesized. The results are shown in FIG. 19.
[0316] C. Fluorescence Energy-Transfer Substrate Assays
[0317] A different application of the present invention tests for
catalytic cleavage of various polymer sequences by an enzyme or
other catalyst. For example, aspartyl proteases such as renin, HIV
proteases, elastase, collagenase and some cathepsins can be tested
against an array of peptides. According to this aspect of the
invention, a variety of peptide sequences are synthesized on a
solid substrate by the protection-deprotection strategy outlined
above. The resulting array is probed with an enzyme which might
cleave one or more peptide elements of the array resulting in a
detectable chain.
[0318] In one embodiment, the peptides to be tested have a
fluorescence donor group such as 1-aminobenzoic acid (anthranilic
acid or ABZ) or aminomethylcoumarin (AMC) located at one position
on the peptide and a fluorescence quencher group such as lucifer
yellow, methyl red or nitrobenzo-2-oxo-1,3-diazole (NBD) at a
different position near the distal end of the peptide. Note, that
some "donor" groups can also serve as "quencher" groups, depending
on the relative excitation and emission frequencies of the
particular pair selected. The intramolecular resonance energy
transfer from the fluorescence donor molecule to the quencher will
quench the fluorescence of the donor molecule. Upon cleavage,
however, the quencher is separated from the donor group, leaving
behind a fluorescent fragment. Plus, a scan of the surface with an
epifluorescence microscope for example, will show bright regions
where the peptide has been cleaved. As an example, FIG. 20A shows a
tripeptide having a donor-quencher pair on a substrate. The
fluorescence donor molecule, 1-aminobenzoic acid (ABZ), is coupled
to the .epsilon.-amino group of lysine (Lys) on the P' side of the
substrate. The donor molecule could, of course, be attached to the
.alpha.-amine group. A fluorescence quencher, NBD caproic acid is
coupled to the P side of the substrate molecule. Upon cleavage by a
protease as shown in FIG. 20B, the quencher is released leaving the
fluorescent type fragment still bound to the solid substrate for
detection.
[0319] FIG. 21 demonstrates the fluorescence donor-quencher
resonance energy transfer assay for two quenchers. In FIG. 21A the
bright outer circle is produced by fluorescence from
benzyloxycarbonyl protected 1-aminobenzoic acid linked to a slide
through lysine and a linker. The dark inner circle shows the
quenching effect of methyl red. FIG. 21B shows a similar result,
with the dark inner circle resulting from the quenching effect of
NBD-caproic acid.
[0320] III. Synthesis
[0321] A. Reactor System
[0322] FIG. 22a schematically illustrates a preferred embodiment of
a reactor system 100 for synthesizing polymers on the prepared
substrate in accordance with one aspect of the invention. The
reactor system includes a body 102 with a cavity 104 on a surface
thereof. In preferred embodiments the cavity 104 is between about
50 and 1000 .mu.m deep with a depth of about 500 .mu.m
preferred.
[0323] The bottom of the cavity is preferably provided with an
array of ridges 106 which extend both into the plane of the Figure
and parallel to the plane of the Figure. The ridges are preferably
about 50 to 200 .mu.m deep and spaced at about 2 to 3 mm. The
purpose of the ridges is to generate turbulent flow for better
mixing. The bottom surface of the cavity is preferably light
absorbing so as to prevent reflection of impinging light.
[0324] A substrate 112 is mounted above the cavity 104. The
substrate is provided along its bottom surface 114 with a
photoremovable protecting group such as NVOC with or without an
intervening linker molecule. The substrate is preferably
transparent to a wide spectrum of light, but in some embodiments is
transparent only at a wavelength at which the protecting group may
be removed (such as UV in the case of NVOC). The substrate in some
embodiments is a conventional microscope glass slide or cover slip.
The substrate is preferably as thin as possible, while still
providing adequate physical support. Preferably, the substrate is
less than about 1 mm thick, more preferably less than 0.5 mm thick,
more preferably less than 0.1 mm thick, and can be less than 0.05
mm thick. In alternative preferred embodiments, the substrate is
quartz, silicon, or other compounds such as a silicon nitride.
[0325] The substrate and the body serve to seal the cavity except
for an inlet port 108 and an outlet port 110. The body and the
substrate may be mated for sealing in some embodiments with one or
more gaskets. According to one embodiment, the body is provided
with two concentric gaskets and the intervening space is held at
vacuum to ensure mating of the substrate to the gaskets.
[0326] Fluid is pumped through the inlet port into the cavity by
way of a pump 116 which may be, for example, a model no. B-120-S
made by Eldex Laboratories, from fluid supply 118. Selected fluids
are circulated into the cavity by the pump, through the cavity, and
out the outlet for recirculation or disposal. The reactor may be
subjected to ultrasonic radiation and/or heated to aid in agitation
in some embodiments.
[0327] Above the substrate 112, a lens 120 is provided which may
be, for example, a 2" 100 mm focal length fused silica lens. For
the sake of a compact system, a reflective mirror 122 may be
provided for directing light from a light source 124 onto the
substrate. Light source 124 may be, for example, a Hg(Xe) light
source manufactured by Oriel and having model no. 66024. A second
lens 126 may be provided for the purpose of projecting a mask image
onto the substrate in combination with lens 620. This form of
lithography is referred to herein as projection printing. As will
be apparent from this disclosure, proximity printing and the like
may also be used according to some embodiments.
[0328] Light from the light source is permitted to reach only
selected locations on the substrate as a result of mask 128. Mask
128 may be, for example, a glass slide having etched chrome
thereon. The mask 128 in one embodiment is provided with a grid of
transparent locations and opaque locations. Such masks may be
manufactured by, for example, Photo Sciences, Inc. Light passes
freely through the transparent regions of the mask, but is
reflected from or absorbed by other regions. Therefore, only
selected regions of the substrate are exposed to light.
[0329] As discussed above, light valves (LCD's) may be used as an
alternative to conventional masks to selectively expose regions of
the substrate. Alternatively, fiber optic faceplates such as those
available from Schott Glass, Inc, may be used for the purpose of
contrast enhancement of the mask or as the sole means of
restricting the region to which light is applied. Such faceplates
would be placed directly above or on the substrate in the reactor
shown in FIG. 22A. In still further embodiments, flys-eye lenses,
tapered fiber optic faceplates, or the like, may be used for
contrast enhancement.
[0330] In order to provide for illumination of regions smaller than
a wavelength of light, more elaborate techniques may be utilized.
For example, according to one preferred embodiment, light is
directed at the substrate by way of molecular microcrystals on the
tip of, for example, micropipettes. Such devices are disclosed in
Lieberman et al., "A Light Source Smaller Than the Optical
Wavelength," Science (1990) 247:59-61, which is incorporated herein
by reference for all purposes.
[0331] In operation, the substrate is placed on the cavity and
sealed thereto. All operations in the process of preparing the
substrate are carried out in a room lit primarily or entirely by
light of a wavelength outside of the light range at which the
protecting group is removed. For example, in the case of NVOC, the
room should be lit with a conventional dark room light which
provides little or no UV light. All operations are preferably
conducted at about room temperature.
[0332] A first, deprotection fluid (without a monomer) is
circulated through the cavity. The solution preferably is of 5 mM
sulfuric acid in dioxane solution which serves to keep exposed
amino groups protonated and decreases their reactivity with
photolysis by-products. Absorptive materials such as
N,N-diethylamino 2,4-dinitrobenzene, for example, may be included
in the deprotection fluid which serves to absorb light and prevent
reflection and unwanted photolysis.
[0333] The slide is, thereafter, positioned in a light ray path
from the mask such that first locations on the substrate are
illuminated and, therefore, deprotected. In preferred embodiments
the substrate is illuminated for between about 1 and 15 minutes
with a preferred illumination time of about 10 minutes at 10-20
mW/cm.sup.2 with 365 nm light. The slides are neutralized (i.e.,
brought to a pH of about 7) after photolysis with, for example, a
solution of diisopropylethylamine (DIEA) in methylene chloride for
about 5 minutes.
[0334] The first monomer is then placed at the first locations on
the substrate. After irradiation, the slide is removed, treated in
bulk, and then reinstalled in the flow cell. Alternatively, a fluid
containing the first monomer, preferably also protected by a
protective group, is circulated through the cavity by way of pump
116aF. If, for example, it is desired to attach the amino acid Y to
the substrate at the first locations, the amino acid Y (bearing a
protecting group on its .alpha.-nitrogen), along with reagents used
to render the monomer reactive, and/or a carrier, is circulated
from a storage container 118aF, through the pump, through the
cavity, and back to the inlet of the pump.
[0335] The monomer carrier solution is, in a preferred embodiment,
formed by mixing of a first solution (referred to herein as
solution "A") and a second solution (referred to herein as solution
"B"). Table 3 provides an illustration of a mixture which may be
used for solution A.
5TABLE 3 Representative Monomer Carrier Solution "A" 0.25 mMoles
NVOC amino protected amino acid 37 mg HOBT (1-Hydroxybenzotriazole)
250 .mu.l DMF (Dimethylformamide) 86 .mu.l DIEA
(Diisopropylethylamine)
[0336] The composition of solution B is illustrated in Table 4.
Solutions A and B are mixed and allowed to react at room
temperature for about 8 minutes, then diluted with 2 ml of DMF, and
500 .mu.l are applied to the surface of the slide or the solution
is circulated through the reactor system and allowed to react for
about 2 hours at room temperature. The slide is then washed with
DMF, methylene chloride and ethanol.
6TABLE 4 Representative Monomer Carrier Solution "B" 250 .mu.l DMF
111 mg BOP (Benzotriazolyl-n-oxy-tris(dimethylamino)
phosphoniumhexafluoroph- osphate)
[0337] As the solution containing the monomer to be attached is
circulated through the cavity, the amino acid or other monomer will
react at its carboxy terminus with amino groups on the regions of
the substrate which have been deprotected. Of course, while the
invention is illustrated by way of circulation of the monomer
through the cavity, the invention could be practiced by way of
removing the slide from the reactor and submersing it in an
appropriate monomer solution.
[0338] After addition of the first monomer, the solution containing
the first amino acid is then purged from the system. After
circulation of a sufficient amount of the DMF/dichloromethane such
that removal of the amino acid can be assured (e.g., about
50.times. times the volume of the cavity and carrier lines), the
mask or substrate is repositioned, or a new mask is utilized such
that second regions on the substrate will be exposed to light and
the light 124 is engaged for a second exposure. This will deprotect
second regions on the substrate and the process is repeated until
the desired polymer sequences have been synthesized.
[0339] The entire derivatized substrate is then exposed to a
receptor of interest, preferably labeled with, for example, a
fluorescent marker, by circulation of a solution or suspension of
the receptor through the cavity or by contacting the surface of the
slide in bulk. The receptor will preferentially bind to certain
regions of the substrate which contain complementary sequences.
[0340] Antibodies are typically suspended in what is commonly
referred to as "supercocktail," which may be, for example, a
solution of about 1% BSA (bovine serum albumin), 0.05% Tween 20.TM.
in PBS (phosphate buffered saline) buffer. The antibodies are
diluted into the supercocktail buffer to a final concentration of,
for example, about 0.1 to 4 .mu.g/ml.
[0341] FIG. 22B illustrates an alternative preferred embodiment of
the reactor shown in FIG. 22A. According to this embodiment, the
mask 128 is placed directly in contact with the substrate.
Preferably, the etched portion of the mask is placed face down so
as to reduce the effects of light dispersion. According to this
embodiment, the imaging lenses 120 and 126 are not necessary
because the mask is brought into close proximity with the
substrate.
[0342] FIG. 23 schematically illustrates a particularly preferred
device used to synthesize diverse polymer sequences on a substrate.
The device includes an automated peptide synthesizer 401. The
automated peptide synthesizer is a device which flows selected
reagents through a flow cell 402 under the direction of a computer
404. In a preferred embodiment the synthesizer is an ABI Peptide
Synthesizer, model no. 431A. The computer may be selected from a
wide variety of computers or discrete logic including for, example,
an IBM PC-AT or similar computer linked with appropriate internal
control systems in the peptide synthesizer. The PC is provided with
signals from the ABI computer indicative of, for example, the
beginning of a photolysis cycle. One can also modify the
synthesizer with a board that links the contacts of relays in the
computer in parallel with the switches to the keyboard of the
control panel of the synthesizer to eliminate some of the
keystrokes that would otherwise be required to operate the
synthesizer.
[0343] Substrate 406 is mounted on the flow cell, forming a cavity
between the substrate and the flow cell. Selected reagents flow
through this cavity from the peptide synthesizer at selected times,
forming an array of peptides on the face of the substrate in the
cavity. Mounted above the substrate, and preferably in contact with
the substrate is a mask 408. Mask 408 is transparent in selected
regions to a selected wavelength of light and is opaque in other
regions to the selected wavelength of light. The mask is
illuminated with a light source 410 such as a UV light source. In
one specific embodiment the light source 410 is a model no. 82420
made by Oriel. The mask is held and translated by an x-y
translation stage 412 such as a translation stage made by Newport
corp. The computer coordinates action of the peptide synthesizer,
translation stage, and light source. Of course, the invention may
be used in some embodiments with translation of the substrate
instead of the mask.
[0344] In operation, the substrate is mounted on the flow cell. The
substrate, with its surface protected by a suitable photo removable
protecting group, is exposed to light at selected locations by
positioning the mask and directing light from a light source,
through the mask, onto the substrate for a desired period of time
(such as, for example, 1 sec to 60 min in the case of peptide
synthesis). A selected peptide or other monomer/polymer is pumped
through the reactor cavity by the peptide synthesizer for binding
at the selected locations on the substrate. After a selected
reaction time (such as about 1 sec to 300 min in the case of
peptide reactions) the monomer is washed from the system, the mask
is appropriately repositioned or replaced, and the cycle is
repeated. In most embodiments of the invention, reactions may be
conducted at or near ambient temperature. Agitation can be used to
mix the reaction contents.
[0345] FIGS. 24A and 24B are flow charts of the software used in
operation of the reactor system. At step 502 the peptide synthesis
software (PS, attached as appendix 3) is initialized. At step 504
the system calibrates positioners on the x-y translation stage and
begins a main loop. At step 506 the system determines which, if
any, of the function keys on the computer have been pressed. If F1
has been pressed, the system prompts the user for input of a
desired synthesis process. If the user enters F2, the system allows
a user to edit a file for a synthesis process at step 510. If the
user enters F3 the system loads a process from a disk at step 512.
If the user enters F4 the system saves an entered or edited process
to disk at step 514. If the user selects F5 the current process is
displayed at step 516 while selection of F6 starts the main portion
of the program, i.e., the actual synthesis according to the
selected process. If the user selects F7 the system displays the
location of the synthesized peptides, while pressing F10 returns
the user to the disk operating system.
[0346] FIG. 24B illustrates the synthesis step 518 in greater
detail. The main loop of the program is started in which the system
first moves the mask to a next position at step 526. During the
main loop of the program, necessary chemicals flow through the
reaction cell under the direction of the on-board computer in the
peptide synthesizer. At step 528 the system then waits for an
exposure command and, upon receipt of the exposure command exposes
the substrate for a desired time at step 530. When an
acknowledgement of complete exposure is received at step 532 the
system determines if the process is complete at step 534 and, if
so, waits for additional keyboard input at step 536 and,
thereafter, exits the perform synthesis process.
[0347] A computer program ("PS") used for operation of the system
described above is written in Turbo C (Borland Int'l) and has been
implemented in an IBM compatible system. The motor control software
is adapted from software produced by Newport Corporation. It will
be recognized that a large variety of programming languages could
be utilized without departing from the scope of the invention
herein. Certain calls are made to a graphics program in "Programmer
Guide to PC and PS2 Video Systems" (Wilton, Microsoft Press, 1987),
which is incorporated herein by reference for all purposes.
[0348] Alignment of the mask is achieved by one of two methods in
preferred embodiments. In a first embodiment the system relies upon
relative alignment of the various components, which is normally
acceptable since x-y-z translation stages are capable of sufficient
accuracy for the purposes herein. In alternative embodiments,
alignment marks on the substrate are coupled to a CCD device for
appropriate alignment.
[0349] According to some embodiments, pure reagents are not added
at each step, or complete photolysis of the protecting groups is
not provided at each step. According to these embodiments, multiple
products will be formed in each synthesis site. For example, if the
monomers A and B are mixed during a synthesis step, A and B will
bind to deprotected regions, roughly in proportion to their
concentration in solution. Hence, a mixture of compounds will be
formed in a synthesis region. A substrate formed with mixtures of
compounds in various synthesis regions may be used to perform, for
example, an initial screening of a large number of compounds, after
which a smaller number of compounds in regions which exhibit high
binding affinity are further screened. Similar results may be
obtained by only partially photolyzing a region, adding a first
monomer, re-photolyzing the same region, and exposing the region to
a second monomer.
[0350] B. Combinatorial Synthesis Strategy
[0351] In preferred embodiments an ordered sequence of masks is
utilized. In some embodiments it is possible to use as few as a
single mask to synthesize all of the possible polymers of a given
monomer set.
[0352] If, for example, it is desired to synthesize all 16
dinucleotides from four bases, a 1 cm square synthesis region is
divided conceptually into 16 boxes, each 0.25 cm wide. Denote the
four monomer units by A, B, C, and D. The first reactions are
carried out in four vertical columns, each 0.25 cm wide. The first
mask exposes the left-most column of boxes, where A is coupled. The
second mask exposes the next column, where B is coupled; followed
by a third mask, for the C column; and a final mask that exposes
the right-most column, for D. The first, second, third, and fourth
masks may be a single "all-purpose" mask translated to different
locations. Such an "all-purpose" mask can be useful in any
synthesis strategy, whether binary or not.
[0353] The process is repeated in the horizontal direction for the
second unit of the dimer. This time, the masks allow exposure of
horizontal rows, again 0.25 cm wide. A, B, C, and D are
sequentially coupled using masks that expose horizontal fourths of
the reaction area. The resulting substrate contains all 16
dinucleotides of four bases.
[0354] The eight masks used to synthesize the dinucleotides or
other dimers are related to one another by translation or rotation.
In fact, one mask can be used in all eight steps if it is suitably
rotated and translated. For example, in the example above, a mask
with a single transparent region could be sequentially used to
expose each of the vertical columns, translated 90.degree., and
then sequentially used to allow exposure of the horizontal
rows.
[0355] Tables 5 and 6 provide a simple computer program in Quick
Basic for planning a masking program and a sample output,
respectively, for the synthesis of a polymer chain of three
monomers ("residues") having three different monomers in the first
level, four different monomers in the second level, and five
different monomers in the third level in a striped pattern. The
output of the program is the number of cells, the number of
"stripes" (light regions) on each mask, and the amount of
translation required for each exposure of the mask.
7TABLE 5 Mask Strategy Program DEFINT A-Z DIM b(20), w(20), l(500)
F$ = "LPT1:" OPEN f$ FOR OUTPUT AS #1 jmax = 3 `Number of residues
b(1) = 3: b(2) = 4: b(3) = 5 `Number of building blocks for res
1,2,3 g = 1: lmax(1) = 1 FOR j = 1 TO jmax: g= g * b(j): NEXT j
w(0) = 0: w(1) = g / b(1) PRINT #1, "MASK2.BAS ", DATES, TIMES:
PRINT #1, PRINT #1, USING "Number of residues=##"; jmax FOR j = l
TO jmax PRINT #1, USING " Residue ## ## building blocks"; j; b(j)
NEXT j PRINT #1, " PRINT #1, USING "Number of cells=####"; g: PRINT
#1, FOR j = 2 TO jmax lmax(j) = lmax(j - 1) * b(j - 1) w(j) = w(j -
l) / b(j) NEXT j FOR j = l TO jmax PRINT #1, USING "Mask for
residue ##"; j: PRINT #1, PRINT #1, USING " Number of stripes=###";
lmax(j) PRINT #1, USING " Width of each stripe=###"; w(j) FOR 1 = 1
TO lmax(j) a = l + (l - l) * w(j - 1) ae = a + w(j) - l PRINT #l,
USING " Stripe ## begins at location ### and ends at ###"; l; a; ae
NEXT l PRINT #1, PRINT #1, USING " For each of ## building blocks,
translate mask by ## cell(s)"; b(j); w(j), PRINT #1, : PRINT #1, :
PRINT #1, NEXT j .COPYRGT.Copyright 1990, Affymax Technologies
N.V.
[0356]
8TABLE 6 Masking Strategy Output Number of residues = 3 Residue 1 3
building blocks Residue 2 4 building blocks Residue 3 5 building
blocks Number of cells = 60 Mask for residue 1 Number of stripes =
1 Width of each stripe = 20 Stripe 1 begins at location 1 and ends
at 20 For each of 3 building blocks, translate mask by 20 cell(s)
Mask for residue 2 Number of stripes = 3 Width of each stripe = 5
Stripe 1 begins at location 1 and ends at 5 Stripe 2 begins at
location 21 and ends at 25 Stripe 3 begins at location 41 and ends
at 45 For each of 4 building blocks, translate mask by 5 cell(s)
Mask for residue 3 Number of stripes = 12 Width of each stripe = 1
Stripe 1 begins at location 1 and ends at 1 Stripe 2 begins at
location 6 and ends at 6 Stripe 3 begins at location 11 and ends at
11 Stripe 4 begins at location 16 and ends at 16 Stripe 5 begins at
location 21 and ends at 21 Stripe 6 begins at location 26 and ends
at 26 Stripe 7 begins at location 31 and ends at 31 Stripe 8 begins
at location 36 and ends at 36 Stripe 9 begins at location 41 and
ends at 41 Stripe 10 begins at location 46 and ends at 46 Stripe 11
begins at location 51 and ends at 51 Stripe 12 begins at location
56 and ends at 56 For each of 5 building blocks, translate mask by
1 cell(s) .COPYRGT.Copyright 1990, Affymax Technologies N.V.
[0357] In preferred embodiments of the invention herein a binary
synthesis strategy is utilized. The binary synthesis strategy is
illustrated herein primarily with regard to a masking strategy,
although it will be applicable to other polymer synthesis
strategies such as the pin strategy, and the like.
[0358] In a binary synthesis strategy, the substrate is irradiated
with a first mask, exposed to a first building block, irradiated
with a second mask, exposed to a second building block, etc. Each
combination of masked irradiation and exposure to a building block
is referred to herein as a "cycle."
[0359] In a preferred binary masking strategy, the masks for each
cycle allow illumination of half of a region of interest on the
substrate and no illumination of the remaining half of the region
of interest. By "half" it is intended herein not to mean exactly
one-half the region of interest, but instead a large fraction of
the region of interest such as from about 30 to 70 percent of the
region of interest. It will be understood that the entire masking
strategy need not take a binary form; instead non-binary cycles may
be introduced as desired between binary cycles.
[0360] In preferred embodiments of the binary masking strategy, a
given cycle illuminates only about half of the region which was
illuminated in a previous cycle, while not illuminating the
remaining half of the illuminated portion from the previous cycle.
Conversely, in such preferred embodiments, a given cycle
illuminates half of the region which was not illuminated in the
previous cycle and does not illuminate half the region which was
not illuminated in a previous cycle.
[0361] The synthesis strategy is most readily illustrated and
handled in matrix notation. At each synthesis site, the
determination of whether to add a given monomer is a binary
process. Therefore, each product element P.sub.j in a product
matrix P is given by the dot product of two vectors, a chemical
reactant vector, (CRV) e.g., CRV=[A,B,C,D], and a binary vector
.sigma..sub.j. Inspection of the products in the example below for
a four-step synthesis, shows that in one four-step synthesis
.sigma..sub.1[1,0,1,0], .sigma..sub.2=[1,0,0,1],
.sigma..sub.3=[0,1,1,0], and .sigma..sub.4=[0,1,0,1], where a 1
indicates illumination and a 0 indicates no illumination.
Therefore, it becomes possible to build a "switch matrix" S from
the column vectors .sigma..sub.j (j=1, k where k is the number of
products). 3 S = 1 2 3 4 1 1 0 0 0 0 1 1 1 0 1 0 0 1 0 1
[0362] The outcome P of a synthesis is simply P=CS, the product of
the chemical reactant matrix and the switch matrix.
[0363] The switch matrix for an n-cycle synthesis yielding k
products has n rows and k columns. An important attribute of S is
that each row specifies a mask. A two-dimensional mask m.sub.j for
the jth chemical step of a synthesis is obtained directly from the
jth row of S by placing the elements s.sub.jl, . . . s.sub.jk into,
for example, a square format. The particular arrangement below
provides a square format, although linear or other arrangements may
be utilized. 4 S = s 11 s 12 s 13 s 14 s 21 s 22 s 23 s 24 s 31 s
32 s 33 s 34 s 41 s 42 s 43 s 44 m j = s j1 s j2 s j3 s j4
[0364] Of course, compounds formed in a light-activated synthesis
can be positioned in any defined geometric array. A square or
rectangular matrix is convenient but not required. The rows of the
switch matrix may be transformed into any convenient array as long
as equivalent transformations are used for each row.
[0365] For example, the masks in the four-step synthesis below are
then denoted by: 5 m 1 = 1 1 0 0 m 2 = 0 0 1 1 m 3 = 1 0 1 0 m 4 =
0 1 0 1
[0366] where 1 denotes illumination (activation) and 0 denotes no
illumination.
[0367] The matrix representation is used to generate a desired set
of products and product maps in preferred embodiments. Each
compound is defined by the product of the chemical vector and a
particular switch vector. Therefore, for each synthesis address,
one simply saves the switch vector, assembles all of them into a
switch matrix, and extracts each of the rows to form the masks.
[0368] In some cases, particular product distributions or a maximal
number of products are desired. For example, for CRV=[A,B,C,D], any
switch vector (.sigma..sub.j) consists of four bits. Sixteen
four-bit vectors exist. Hence, a maximum of 16 different products
can be made by sequential addition of the reagents [A,B,C,D]. These
16 column vectors can be assembled in 16! different ways to form a
switch matrix. The order of the column vectors defines the masking
patterns, and therefore, the spatial ordering of products but not
their makeup. One ordering of these columns gives the following
switch matrix (in which "null" (.O slashed.) additions are included
in brackets for the sake of completeness, although such null
additions are elsewhere ignored herein): 6 S = 1 16 CRV 1 1 1 1 1 1
1 1 0 0 0 0 0 0 0 0 A [ 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 ] .0. 1 1 1
1 0 0 0 0 1 1 1 1 0 0 0 0 B [ 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 1 ] .0.
1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 C [ 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1
] .0. 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 D [ 0 1 0 1 0 1 0 1 0 1 0 1 0
1 0 1 ] .0.
[0369] The columns of S according to this aspect of the invention
are the binary representations of the numbers 15 to 0. The sixteen
products of this binary synthesis are ABCD, ABC, ABD, AB, ACD, AC,
AD, A, BCD, BC, BD, B, CD, C, D, and .O slashed. (null). Also note
that each of the switch vectors from the four-step synthesis masks
above (and hence the synthesis products) are present in the four
bit binary switch matrix. (See columns 6, 7, 10, and 11). Note that
if the desired compounds comprise only dimers, then one could
extract the switch vectors for compounds AB, AC, AD, BC, BD, and CD
for this synthesis.
[0370] This synthesis procedure provides an easy way for mapping
the completed products. The products in the various locations on
the substrate are simply defined by the columns of the switch
matrix (the first column indicating, for example, that the product
ABCD will be present in the upper left-hand location of the
substrate). Furthermore, if only selected desired products are to
be made, the mask sequence can be derived by extracting the columns
with the desired sequences. For example, to form the product set
ABCD, ABD, ACD, AD, BCD, BD, CD, and D, the masks are formed by use
of a switch matrix with only the 1st, 3rd, 5th, 7th, 9th, 11th,
13th, and 15th columns arranged into the switch matrix: 7 S = 1 1 1
1 0 0 0 0 1 1 0 0 1 1 0 0 1 0 1 0 1 0 1 0 1 1 1 1 1 1 1 1
[0371] To form all of the polymers of length 4, the reactant matrix
[ABCDABCDABCDABCD] is used. The switch matrix will be formed from a
matrix of the binary numbers from 0 to 2.sup.16 arranged in
columns. The columns having four monomers are then selected and
arranged into a switch matrix. Therefore, it is seen that the
binary switch matrix in general will provide a representation of
all the products which can be made from an n-step synthesis, from
which the desired products are then extracted.
[0372] The rows of the binary switch matrix will, in preferred
embodiments, have the property that each masking step illuminates
half of the synthesis area. Each masking step also factors the
preceding masking step; that is, half of the region that was
illuminated in the preceding step is again illuminated, whereas the
other half is not. Half of the region that was not illuminated in
the preceding step is also illuminated, whereas the other half is
not. Thus, masking is recursive. The masks are constructed, as
described previously, by extracting the elements of each row and
placing them in a square array. For example, the four masks in S
for a four-step synthesis are: 8 m 1 = 1 1 1 1 1 1 1 1 0 0 0 0 0 0
0 0 m 2 = 1 1 1 1 0 0 0 0 1 1 1 1 0 0 0 0 m 3 = 1 1 0 0 1 1 0 0 1 1
0 0 1 1 0 0 m 4 = 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0
[0373] The recursive factoring of masks allows the products of a
light-directed synthesis to be represented by a polynomial. (Some
light activated syntheses can only be denoted by irreducible, i.e.,
prime polynomials.) For example, the polynomial corresponding to
the top synthesis of FIG. 8a (discussed below) is
[0374] P=(A+B)(C+D)
[0375] A reaction polynomial may be expanded as though it were an
algebraic expression, provided that the order of joining of
reactants X.sub.1 and X.sub.2 is preserved
(X.sub.1X.sub.2.noteq.X.sub.2X.sub.1), i.e., the products are not
commutative. The product then is AC+AD+BC+BD. The polynomial
explicitly specifies the reactants and implicitly specifies the
mask for each step. Each pair of parentheses demarcates a round of
synthesis. The chemical reactants of a round (e.g., A and B) react
at nonoverlapping sites and hence cannot combine with one another.
The synthesis area is divided equally among the elements of a round
(e.g., A is directed to one-half of the area and B to the other
half). Hence, the masks for a round (e.g., the masks mA and
m.sub.B) are orthogonal and form an orthonormal set. The polynomial
notation also signifies that each element in a round is to be
joined to each element of the next round (e.g., A with C, A with D,
B with C, and B with D). This is accomplished by having m.sub.C
overlap m.sub.A and m.sub.B equally, and likewise for m.sub.D.
Because C and D are elements of a round, m.sub.C and m.sub.D are
orthogonal to each other and form an orthonormal set.
[0376] The polynomial representation of the binary synthesis
described above, in which 16 products are made from 4 reactants,
is
[0377] (A+.O slashed.)(B+.O slashed.)(C+.O slashed.)(D+.O
slashed.)
[0378] which gives ABCD, ABC, ABD, AB, ACD, AC, AD, A, BCD, BC, BD,
B, CD, C, D, and .O slashed. when expanded (with the rule that .O
slashed.X=X and X.O slashed.=X, and remembering that joining is
ordered). In a binary synthesis, each round contains one reactant
and one null (denoted by .O slashed.). Half of the synthesis area
receives the reactant and the other half receives nothing. Each
mask overlaps every other mask equally.
[0379] Binary rounds and non-binary rounds can be interspersed as
desired, as in
[0380] P=(A+.O slashed.)(B)(C+D+.O slashed.)(E+F+G)
[0381] The 18 compounds formed are ABCE, ABCF, ABCG, ABDE, ABDF,
ABDG, ABE, ABF, ABG, BCE, BCF, BCG, BDE, BDF, BDG, BE, BF, and BG.
The switch matrix S for this 7-step synthesis is 9 S = 1 1 1 1 1 1
1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0
0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 1 1 1 0 0 0 1 0
0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0
0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1
[0382] The round denoted by (B) places B in all products because
the reaction area was uniformly activated (the mask for B consisted
entirely of 1's).
[0383] The number of compounds k formed in a synthesis consisting
of r rounds, in which the ith round has b.sub.i chemical reactants
and z.sub.i nulls, is
[0384] k=.SIGMA.(b.sub.i+z.sub.i)
[0385] and the number of chemical steps n is
[0386] n=.SIGMA.b.sub.i
[0387] The number of compounds synthesized when b=a (the number of
chemical building blocks) and z=0 in all rounds is a.sup.n/a,
compared with 2.sup.n for a binary synthesis. For n=20 and a=5, 625
compounds (all tetramers) would be formed, compared with
1.049.times.10.sup.6 compounds in a binary synthesis with the same
number of chemical steps.
[0388] It should also be noted that rounds in a polynomial can be
nested, as in
[0389] (A+(B+.O slashed.)(C+.O slashed.))(D+.O slashed.)
[0390] The products are AD, BCD, BD, CD, D, A, BC, B, C, and .O
slashed..
[0391] Binary syntheses are attractive for two reasons. First, they
generate the maximal number of products (2.sup.n) for a given
number of chemical steps (n). For four reactants, 16 compounds are
formed in the binary synthesis, whereas only 4 are made when each
round has two reactants. A 10-step binary synthesis yields 1,024
compounds, and a 20-step synthesis yields 1,048,576. Second,
products formed in a binary synthesis are a complete nested set
with lengths ranging from 0 to n. All compounds that can be formed
by deleting one or more units from the longest product (the n-mer)
are present. Contained within the binary set are the smaller sets
that would be formed from the same reactants using any other set of
masks (e.g., AC, AD, BC, and BD formed in the synthesis shown in
FIG. 5 are present in the set of 16 formed by the binary
synthesis). In some cases, however, the experimentally achievable
spatial resolution may not suffice to accommodate all the compounds
that could be formed on a single substrate. Therefore, practical
limitations may require one to select a particular subset of the
possible switch vectors for a given synthesis.
[0392] 1. Example
[0393] FIG. 25 illustrates a synthesis with a binary masking
strategy. The binary masking strategy provides the greatest number
of sequences for a given number of cycles. According to this
embodiment, a mask m.sub.1 allows illumination of half of the
substrate. The substrate is then exposed to the building block A,
which binds at the illuminated regions.
[0394] Thereafter, the mask m.sub.2 allows illumination of half of
the previously illuminated region, while it does not illuminate
half of the previously illuminated region. The building block B is
then added, which binds at the illuminated regions from
m.sub.2.
[0395] The process continues with masks m.sub.3, m.sub.4, and
m.sub.5, resulting in the product array shown in the bottom portion
of the figure. The process generates 32 (2 raised to the power of
the number of monomers) sequences with 5 (the number of monomers)
cycles.
[0396] 2. Example
[0397] FIG. 26 illustrates another preferred binary masking
strategy which is referred to herein as the gray code masking
strategy. According to this embodiment, the masks m.sub.1 to
m.sub.5 are selected such that a side of any given synthesis region
is defined by the edge of only one mask. The site at which the
sequence BCDE is formed, for example, has its right edge defined by
m.sub.5 and its left side formed by mask m.sub.4 (and no other mask
is aligned on the sides of this site). Accordingly, problems
created by misalignment, diffusion of light under the mask and the
like will be minimized.
[0398] 3. Example
[0399] FIG. 27 illustrates another binary masking strategy.
According to this scheme, referred to herein as a modified gray
code masking strategy, the number of masks needed is minimized. For
example, the mask m.sub.2 could be the same mask as m.sub.1 and
simply translated laterally. Similarly, the mask m.sub.4 could be
the same as mask m.sub.3 and simply translated laterally.
[0400] 4. Example
[0401] A four-step synthesis is shown in FIG. 28A. The reactants
are the ordered set {A,B,C,D}. In the first cycle, illumination
through m.sub.1 activates the upper half of the synthesis area.
Building block A is then added to give the distribution 602.
Illumination through mask m.sub.2 (which activates the lower half),
followed by addition of B yields the next intermediate distribution
604. C is added after illumination through m.sub.3 (which activates
the left half) giving the distribution 604, and D after
illumination through M.sub.4 (which activates the right half), to
yield the final product pattern 608 {AC,AD,BC,BD}.
[0402] 5. Example
[0403] The above masking strategy for the synthesis may be extended
for all 400 dipeptides from the 20 naturally occurring amino acids
as shown in FIG. 28B. The synthesis consists of two rounds, with 20
photolysis and chemical coupling cycles per round. In the first
cycle of round 1, mask 1 activates {fraction (1/20)}th of the
substrate for coupling with the first of 20 amino acids. Nineteen
subsequent illumination/coupling cycles in round 1 yield a
substrate consisting of 20 rectangular stripes each bearing a
distinct member of the 20 amino acids. The masks of round 2 are
perpendicular to round 1 masks and therefore a single
illumination/coupling cycle in round 2 yields 20 dipeptides. The 20
illumination/coupling cycles of round 2 complete the synthesis of
the 400 dipeptides.
[0404] 6. Example
[0405] The power of the binary masking strategy can be appreciated
by the outcome of a 10-step synthesis that produced 1,024 peptides.
The polynomial expression for this 10-step binary synthesis
was:
[0406] (f+.O slashed.)(Y+.O slashed.)(G+.O slashed.)(A+.O
slashed.)(G+.O slashed.)(T+.O slashed.)(F+.O slashed.)(L+.O
slashed.)(S+.O slashed.)(F+.O slashed.)
[0407] Each peptide occupied a 400.times.400 .mu.m square. A
32.times.32 peptide array (1,024 peptides, including the null
peptide and 10 peptides of l=1, and a limited number of duplicates)
was clearly evident in a fluorescence scan following side group
deprotection and treatment with the antibody 3E7 and
fluoresceinated antibody. Each synthesis site was a 400.times.400
.mu.m square.
[0408] The scan showed a range of fluorescence intensities, from a
background value of 3,300 counts to 22,400 counts in the brightest
square (x=20, y=9). Only 15 compounds exhibited an intensity
greater than 12,300 counts. The median value of the array was 4,800
counts.
[0409] The identity of each peptide in the array could be
determined from its x and y coordinates (each range from 0 to 31)
and the map of FIG. 29. The chemical units at positions 2, 5, 6, 9,
and 10 are specified by the y coordinate and those at positions 1,
3, 4, 7, 8 by the x coordinate. All but one of the peptides was
shorter than 10 residues. For example, the peptide at x=12 and y=3
is YGAGF (SEQ ID NO:30; positions 1, 6, 8, 9, and 10 are nulls).
YGAFLS (SEQ ID NO:4), the brightest element of the array, is at
x=20 and y=9.
[0410] It is often desirable to deduce a binding affinity of a
given peptide from the measured fluorescence intensity.
Conceptually, the simplest case is one in which a single peptide
binds to a univalent antibody molecule. The fluorescence scan is
carried out after the slide is washed with buffer for a defined
time. The order of fluorescence intensities is then a measure
primarily of the relative dissociation rates of the
antibody-peptide complexes. If the on-rate constants are the same
(e.g., if they are diffusion-controlled), the order of fluorescence
intensities will typically correspond to the order of binding
affinities. However, the situation is sometimes more complex
because a bivalent primary antibody and a bivalent secondary
antibody are used. The density of peptides in a synthesis area
corresponded to a mean separation of -7 nm, which would allow
multivalent antibody-peptide interactions. Hence, fluorescence
intensities obtained according to the method herein will often be a
qualitative indicator of binding affinity. For a more complete
analysis of how the present invention can be extended to the
binding affinity of an immobilized ligand to a receptor, see U.S.
Ser. No. 07/796,947, filed Nov. 22, 1991, and incorporated herein
by reference.
[0411] Another important consideration is the fidelity of
synthesis. Deletions are produced by incomplete photodeprotection
or incomplete coupling. The coupling yield per cycle in these
experiments is typically between 85% and 95%. The contribution to
the net coupling yield from photodeprotection and chemical coupling
has been assessed in the following ways. The photolysis rate for
NVOC-amino acids was experimentally determined and illumination
conditions that ensure greater than 99% of the amino acids have
been photodeprotected were chosen. The chemical coupling efficiency
of selected amino acids on substrates employed in this invention
has also been determined. For example, in order to determine the
coupling efficiency of Leu to Leu, NVOC was first selectively
photolyzed from one region of a NVOC-Leu derivitized surface. The
photochemically deprotected amino groups in this region were then
coupled to a FMOC-Leu-OBt. At this stage, incomplete Leu to Leu
coupling would leave unreacted amino groups. A second photolysis
step was then used to photolyze a different region of the
substrate. Treatment of the substrate with FITC would label the
free amino groups that remain from incomplete chemical coupling in
the first region and free amino groups exposed by photolysis in the
second region. Direct comparison of the quantitative fluorescence
signal from both regions indicates the extent of chemical coupling.
If the chemical coupling yield is high, the ratio of the signals of
the first to the second photolysis regions is low. This technique
has been used in order to develop the experimental conditions that
maximize chemical coupling.
[0412] Implementing the switch matrix by masking is imperfect
because of light diffraction, internal reflection, and scattering.
Consequently, stowaways (chemical units that should not be on
board) arise by unintended illumination of regions that should be
dark. A binary synthesis array contains many of the controls needed
to assess the fidelity of a synthesis. For example, the
fluorescence signal from a synthesis area nominally containing a
tetrapeptide ABCD could come from a tripeptide deletion impurity
such as ACD. Such an artifact would be ruled out by the finding
that the fluorescence intensity of the ACD site is less than that
of the ABCD site.
[0413] The fifteen most highly fluorescent peptides in the array
obtained with the synthesis of 1,024 peptides described above, were
YGAFLS (SEQ ID NO:4), YGAFS (SEQ ID NO:5), YGAFL (SEQ ID NO:6),
YGGFLS (SEQ ID NO:7), YGAF (SEQ ID NO:8), YGALS (SEQ ID NO:9),
YGGFS (SEQ ID NO:10), YGAL (SEQ ID NO:11), YGAFLF (SEQ ID NO:12),
YGAF (SEQ ID NO:8), YGAFF (SEQ ID NO:13), YGGLS (SEQ ID NO:14),
YGGFL (SEQ ID NO:1 and SEQ ID NO:15), YGAFSF (SEQ ID NO:16), and
YGAFLSF (SEQ ID NO:17). A striking feature is that all fifteen
begin with YG, which agrees with previous work showing that an
amino-terminal tyrosine is a key determinant of binding to 3E7.
Residue 3 of this set is either A or G, and residue 4 is either F
or L. The exclusion of S and T from these positions is clear cut.
The finding that the preferred sequence is YG (A/G) (F/L) fits
nicely with the outcome of a study in which a very large library of
peptides on phage generated by recombinant DNA methods was screened
for binding to antibody 3E7 (see Cwirla et al., Proc. Natl. Acad.
Sci. USA, (1990) 87:6378, incorporated herein by reference).
Additional binary syntheses based on leads from peptides on phage
experiments show that YGAFMQ (SEQ ID NO:18), YGAFM (SEQ ID NO:19),
and YGAFQ (SEQ ID NO:20) give stronger fluorescence signals than
does YGGFM (SEQ ID NO:21), the immunogen used to obtain antibody
3E7.
[0414] Variations on the above masking strategy will be valuable in
certain circumstances. For example, if a "kernel" sequence of
interest consists of PQR separated from XYZ, the aim is to
synthesize peptides in which these units are separated by a
variable number of different residues. The kernel can be placed in
each peptide by using a mask that has 1's everywhere. The
polynomial representation of a suitable synthesis is:
[0415] (P)(O)(R)(A+.O slashed.)(B+.O slashed.)(C+.O slashed.)(D+.O
slashed.)(X)(Y)(Z)
[0416] Sixteen peptides will be formed, ranging in length from the
6-mer PQRXYZ to the 10-mer PQRABCDXYZ.
[0417] Several other masking strategies will also find value in
selected circumstances. By using a particular mask more than once,
two or more reactants will appear in the same set of products. For
example, suppose that the mask for an 8-step synthesis is
[0418] A 11110000
[0419] B 00001111
[0420] C 11001100
[0421] D 00110011
[0422] E 10101010
[0423] F 01010101
[0424] G 11110000
[0425] H 00001111
[0426] The products are ACEG, ACFG, ADEG, ADFG, BCEH, BCFH, BDEH,
and BDFH. A and G always appear in the same product, although not
necessarily next to each other, because their additions were
directed by the same mask, and likewise for B and H.
[0427] 7. Example
[0428] The synthesis strategies shown above are useful in many
different applications. To aid in applying the present invention to
any desired synthesis, the following illustrative example is
provided. Assume one wishes to synthesize polymers up to 4 monomers
in length. A given polymer can be designated as Y.sub.1 Y.sub.2
Y.sub.3 Y.sub.4. If the monomer set contains 20 members, then the
set (5) can be represented as follows:
[0429] S={M.sub.1, M.sub.2, M.sub.3, . . . M.sub.20},
[0430] where, for example, Y.sub.1 may be M.sub.1, Y.sub.2 may be
M.sub.16, etc. Then, if one desires to synthesize all polymers in
which each position is varied through the entire set of monomers,
then the synthesis can be represented as:
[0431] Product of Synthesis
SY.sub.2Y.sub.3Y.sub.4=Y.sub.1SY.sub.3Y.sub.4+-
Y.sub.1Y.sub.2SY.sub.4+Y.sub.1Y.sub.2Y.sub.3S.
[0432] In the above polynomial, there are four terms, and each term
represents 20 different compounds. If one desires to synthesize all
polymers in which two positions are varied through the entire set
of monomers, then the synthesis can be represented as:
[0433] Product of Synthesis
SSY.sub.3Y.sub.4+SY.sub.2SY.sub.4+SY.sub.2SY.s-
ub.3S+Y.sub.1SSY.sub.4+Y.sub.1SY.sub.3S+Y.sub.1Y.sub.2SS.
[0434] In the above polynomial, there are six terms, and each term
represents 400 compounds. If one desires to synthesize all polymers
in which three positions are varied through the entire set of
monomers, then the synthesis can be represented as:
[0435] Product of
Synthesis=SSSY.sub.4+SSY.sub.3S=SY.sub.2SS+Y.sub.1SSS.
[0436] In the above polynomial, there are four terms and each term
represents 8,000 compounds. If one desires to synthesize all
polymers in which four positions are varied through the entire set
of monomers, then the synthesis can be represented as:
[0437] Product of Synthesis SSSS
[0438] In the above polynomial, there is one term, which represents
160,000 compounds.
[0439] By modeling the synthesis as a polynomial expression, one
can more easily discern the appropriate masking strategy required
to effect the synthesis.
[0440] 8. Example
[0441] One example of the power of this strategy involved the
mapping of a binding epitope on dynorphin B. The sequence of this
epitope was demonstrated (as shown below) to be RQFKVVT (SEQ. ID
NO:31). An array of peptides synthesized using the general
protection-deprotection technology outlined above. Referring to
FIG. 30, the following peptides were synthesized: row 1=RXFKVVT;
row 2=RQXFKVVT; row 3=RQXKVVT; and row 4=RXKVVT. In each row, "X"
represents a group of four amino acids, which were simultaneously
added to the immobilized peptide. The particular group of amino
acids used as "X" on a given block are identified by column number
as follows: column 1 [G,A,R,K in a ratio 1:1:1:1], column 2 [null];
column 3 [C,M,S,T in the ratio 1:1:1:5]; column 4 [null]; column 5
[F,Y,W,H in the ratio 1:1:1:1]; column 6 [null]; column 7 [D,E,N,Q
in the ratio 1:1:1:1]; column 8 [null]; column 9 [V,L,I,P in the
ratio 2:1:4:1]; and column 10 [null].
[0442] After these peptides were synthesized, they were screened
with an anti-dynorphin B murine monoclonal antibody and then
exposed to a fluorescently labeled goat anti-mouse antibody. The
brightest region (column 2, row 2) corresponds to the binding
epitope, RQFKVVT. Other strongly labeled regions, include column 7,
row 1 and column 5, row 3, each of which contains some peptides
having the above binding epitope. It should be noted that the
varying ratios of some of the amino acids within a given group were
necessary to obtain roughly equal binding efficiencies among the
four amino acids.
[0443] 9. Example
[0444] A much larger array of compounds was prepared as shown in
FIG. 31. In the first four steps, KVVT (SEQ. ID NO:36) was
synthesized over the entire substrate. Next, F was added over 50%
of the substrate via a column mask as displayed in step 5. The next
20 steps involved addition of a row of each of the
genetically-coded amino acids in a stepwise fashion down the
substrate. One-twentieth of the substrate was exposed on each pass.
Next, Q was added over one-half of the entire substrate as shown at
step 26. Finally, R was added over the entire substrate.
[0445] Four classes of peptide were produced: (1) RYKVVT (SEQ. ID
NO:32), (2) RQYKVVT (SEQ. ID NO:33), (3) RQYFKVVT (SEQ. ID NO:34),
and (4) RYFKVVT (SEQ. ID NO:35). In each case Y represents all 20
L-amino acids. FIG. 32 is an image of a fluorescence scan prepared
after the final array of peptides was exposed to anti-dynorphin B
mouse monocolonal antibody followed by goat anti-mouse antibody.
The top one-fourth of the image was prepared according to the
synthesis procedure outlined above.
[0446] 10. Example
[0447] The binding epitope on dynorphin B for the anti-dynorphin B
antibody described above was determined by deleting various amino
acids and combinations of amino acids from the overall sequence of
dynorphin B. The peptides containing these various deletions were
prepared on a substrate by the general protection-deprotection
methods described above. Then the substrate was exposed to the
anti-dynorphin B monoclonal antibody and a fluorescence image was
produced as shown in FIG. 33. Each quadrant of the image contained
at least one site for each of the subject peptides (identified in
Table 7).
[0448] From the intensity distribution contained in this plot, it
was possible to determine the relative binding affinity to various
of the peptides. This information is summarized in Table 7 below
where N is the number of sites of which the peptide was
synthesized.
9TABLE 7 SEQUENCE RELATIVE ADJUSTED N INTENSITY F 8 4.9 .+-. 3.2 F
L 4 2.8 .+-. 2.2 F L R 8 4.7 .+-. 2.1 F L R R 4 6.0 .+-. 2.7 F L R
R Q 4 8.2 .+-. 2.6 F L R R Q F 4 8.5 .+-. 3.4 F L R R Q F K 4 10.2
.+-. 2.8 F L R R Q F K V 8 13.7 .+-. 3.0 F L R R Q F K V V 4 37.4
.+-. 14.6 (30.5 .+-. 5.4) F L R R Q F K V V T 4 84.2 .+-. 28.2
(98.3 .+-. 1.2) L R R Q F K V V T 4 86.9 .+-. 13.7 R R Q F K V V T
4 98.8 .+-. 0.9 R Q F K V V T 8 93.6 .+-. 10.0 (96.2 .+-. 7.1) Q F
K V V T 4 36.2 .+-. 16.0 F K V V T 8 12.9 .+-. 3.7 K V V T 4 10.7
.+-. 1.8 V V T 4 8.2 .+-. 1.2 V T 8 7.9 .+-. 3.0 T 4 7.0 .+-.
3.9
[0449] FIG. 34 is a bar graph showing the relative binding
affinities for each of the peptides.
[0450] C. Linker Selection
[0451] According to preferred embodiments the linker molecules used
as an intermediary between the synthesized polymers and the
substrate are selected for optimum length and/or type for improved
binding interaction with a receptor. According to this aspect of
the invention diverse linkers of varying length and/or type are
synthesized for subsequent attachment of a ligand. Through
variations in the length and type of linker, it becomes possible to
optimize the binding interaction between an immobilized ligand and
its receptor.
[0452] The degree of binding between a ligand (peptide, inhibitor,
hapten, drug, etc.) and its receptor (enzyme, antibody, etc.) when
one of the partners is immobilized onto a substrate will in some
embodiments depend on the accessibility of the receptor in solution
to the immobilized ligand. The accessibility in turn will depend on
the length and/or type of linker molecule employed to immobilize
one of the partners. Preferred embodiments of the invention
therefore employ the VLSIPS.RTM. synthesis technique described
herein to generate an array of, preferably, inactive or inert
linkers of varying length and/or type, using photochemical
protecting groups to selectively expose different regions of the
substrate and to build upon chemically-active groups.
[0453] In the simplest embodiment of this concept, the same unit is
attached to the substrate in varying multiples or lengths in known
locations on the substrate via VLSIPS synthesis techniques to
generate an array of polymers of varying length. A single ligand
(peptide, drug, hapten, etc.) is attached to each of them, and an
assay is performed with the binding site to evaluate the degree of
binding with a receptor that is known to bind to the ligand. In
cases where the linker length impacts the ability of the receptor
to bind to the ligand, varying levels of binding will be observed.
In general, the linker which provides the highest binding will then
be used to assay other ligands synthesized in accordance with the
techniques herein.
[0454] According to other embodiments the binding between a single
ligand/receptor pair is evaluated for linkers of diverse monomer
sequence. According to these embodiments, the linkers are
synthesized in an array in accordance with the techniques herein
and have different monomer sequences (and, optionally, different
lengths). Thereafter, all of the linker molecules are provided with
a ligand known to have at least some binding affinity for a given
receptor. The given receptor is then exposed to the ligand and
binding affinity is deduced. Linker molecules which provide
adequate binding between the ligand and receptor are then utilized
in screening studies.
[0455] D. Protecting Groups
[0456] As discussed above, selectively removable protecting groups
allow creation of well defined areas of substrate surface having
differing reactivities. Preferably, the protecting groups are
selectively removed from the surface by applying a specific
activator, such as electromagnetic radiation of a specific
wavelength and intensity. More preferably, the specific activator
exposes selected areas of the surface to remove the protecting
groups in the exposed areas.
[0457] Protecting groups of the present invention are used in
conjunction with solid phase oligomer syntheses, such as peptide
syntheses using natural or unnatural amino acids, nucleotide
syntheses using deoxyribonucleic and ribonucleic acids,
oligosaccharide syntheses, and the like. In addition to protecting
the substrate surface from unwanted reaction, the protecting groups
block a reactive end of the monomer to prevent self-polymerization.
For instance, attachment of a protecting group to the amino
terminus of an activated amino acid, such as an
N-hydroxysuccinimide-activated ester of the amino acid, prevents
the amino terminus of one monomer from reacting with the activated
ester portion of another during peptide synthesis. Alternatively,
the protecting group may be attached to the carboxyl group of an
amino acid to prevent reaction at this site. Most protecting groups
can be attached to either the amino or the carboxyl group of an
amino acid, and the nature of the chemical synthesis will dictate
which reactive group will require a protecting group. Analogously,
attachment of a protecting group to the 5'-hydroxyl group of a
nucleoside during synthesis using for example, phosphate-triester
coupling chemistry, prevents the 5'-hydroxyl of one nucleoside from
reacting with the 3'-activated phosphate-triester of another.
[0458] Regardless of the specific use, protecting groups are
employed to protect a moiety on a molecule from reacting with
another reagent. Protecting groups of the present invention have
the following characteristics: they prevent selected reagents from
modifying the group to which they are attached; they are stable
(that is, they remain attached to the molecule) to the synthesis
reaction conditions; they are removable under conditions that do
not adversely affect the remaining structure; and once removed,
they do not react appreciably with the surface or surface-bound
oligomer. The selection of a suitable protecting group will depend,
of course, on the chemical nature of the monomer unit and oligomer,
as well as the specific reagents they are to protect against.
[0459] In a preferred embodiment, the protecting groups are
photoactivatable. The properties and uses of photoreactive
protecting compounds have been reviewed. See, McCray et al., Ann.
Rev. of Biophys. and Biophys. Chem. (1989) 18:239-270, which is
incorporated herein by reference. Preferably, the photosensitive
protecting groups will be removable by radiation in the ultraviolet
(UV) or visible portion of the electromagnetic spectrum. More
preferably, the protecting groups will be removable by radiation in
the near UV or visible portion of the spectrum. In some
embodiments, however, activation may be performed by other methods
such as localized heating, electron beam lithography, laser
pumping, oxidation or reduction with microelectrodes, and the like.
Sulfonyl compounds are suitable reactive groups for electron beam
lithography. Oxidative or reductive removal is accomplished by
exposure of the protecting group to an electric current source,
preferably using microelectrodes directed to the predefined regions
of the surface which are desired for activation. Other methods may
be used in light of this disclosure.
[0460] Many, although not all, of the photoremovable protecting
groups will be aromatic compounds that absorb near-UV and visible
radiation. Suitable photoremovable protecting groups are described
in, for example, McCray et al., Patchornik, J. Amer. Chem. Soc.
(1970) 92:6333, and Amit et al., J. Org. Chem. (1974) 39:192, which
are incorporated herein by reference.
[0461] A preferred class of photoremovable protecting groups has
the general formula: 4
[0462] where R.sup.1, R.sup.2, R.sup.3, and R.sup.4 independently
are a hydrogen atom, a lower alkyl, aryl, benzyl, halogen,
hydroxyl, alkoxyl, thiol, thioether, amino, nitro, carboxyl,
formate, formamido or phosphido group, or adjacent substituents
(i.e., R.sup.1-R.sup.2, R.sup.2-R.sup.3, R.sup.3-R.sup.4) are
substituted oxygen groups that together form a cyclic acetal or
ketal; R.sup.5 is a hydrogen atom, a alkoxyl, alkyl, halo, aryl, or
alkenyl group, and n=0 or 1.
[0463] A preferred protecting group, 6-nitroveratryl (NV), which is
used for protecting the carboxyl terminus of an amino acid or the
hydroxyl group of a nucleotide, sugar, or carbohydrate for example,
is formed when R.sup.2 and R.sup.3 are each a methoxy group,
R.sup.1, R.sup.4 and R.sup.5 are each a hydrogen atom, and n=0:
5
[0464] A preferred protecting group, 6-nitroveratryloxycarbonyl
(NVOC), which is used to protect the amino terminus of an amino
acid, or the bydroxyl group of a nucleotide, sugar, or carbohydrate
for example, is formed when R.sup.2 and R.sup.3 are each a methoxy
group, R.sup.1, R.sup.4 and R.sup.5 are each a hydrogen atom, and
n=1: 6
[0465] Another preferred protecting group, 6-nitropiperonyl (NP),
which is used for protecting the carboxyl terminus of an amino acid
or the hydroxyl group of a nucleotide, sugar, or carbohydrate for
example, is formed when R.sup.2 and R.sup.3 together form a
methylene acetal, R.sup.1, R.sup.4 and R.sup.5 are each a hydrogen
atom, and n=0: 7
[0466] Another preferred protecting group,
6-nitropiperonyloxycarbonyl (NPOC), which is used to protect the
amino terminus of an amino acid, or the hydroxyl group of a
nucleotide, sugar, or carbohydrate or example, is formed when
R.sup.2 and R.sup.3 together form a ethylene acetal, R.sup.1,
R.sup.4 and R.sup.5 are each a hydrogen atom, and n 1: 8
[0467] A most preferred protecting group, methyl-6-nitroveratryl
(MeNV), which is used for protecting the carboxyl terminus of an
amino acid or the hydroxyl group of a nucleotide, sugar or
carbohydrate for example, is formed when R.sup.2 and R.sup.3 are
each a methoxy group, R.sup.1 and R.sup.4 are each a hydrogen atom,
R.sup.5 is a methyl group, and n=0: 9
[0468] Another most preferred protecting group,
methyl-6-nitroveratryloxyc- arbonyl (MeNVOC), which is used to
protect the amino terminus of an amino acid, or the hydroxyl group
of a nucleotide, sugar, or carbohydrate for example, is formed when
R.sup.2 and R.sup.3 are each a methoxy group, R.sup.1 and R.sup.4
are each a hydrogen atom, R.sup.5 is a methyl group, and n=1:
10
[0469] Another most preferred protecting group,
methyl-6-nitropiperonyl (MeNP), which is used for protecting the
carboxyl terminus of an amino acid or the hydroxyl group of a
nucleotide, sugar or carbohydrate for example, is formed when
R.sup.2 and R.sup.3 together form a methylene acetal, R.sup.1 and
R.sup.4 are each a hydrogen atom, R.sup.5 is a methyl group, and
n=0: 11
[0470] Another most preferred protecting group,
methyl-6-nitropiperonyloxy- carbonyl (MeNPOC), which is used to
protect the amino terminus of an amino acid or to protect the 5'
hydroxyl of nucleosides, nucleotides, carbohydrates, or sugars for
example, is formed when R.sup.2 and R.sup.3 together form a
methylene acetal, R.sup.1 and R.sup.4 are each a hydrogen atom,
R.sup.5 is a methyl group, and n 1: 12
[0471] A protected amino acid having a photoactivatable oxycarbonyl
protecting group, such NVOC or NPOC or their corresponding methyl
derivatives, MeNVOC or MeNPOC, respectively, on the amino terminus
is formed by acylating the amine of the amino acid or 5'hydroxyl of
a nucleotide, sugar or carbohydrate with an activated oxycarbonyl
ester of the protecting group. Examples of activated oxycarbonyl
esters of NVOC and MeNVOC have the general formula: 13
[0472] where X is halogen, mixed anhydride, phenoxy,
p-nitrophenoxy, N-hydroxysuccinimide, and the like.
[0473] A protected amino acid or nucleotide having a
photoactivatable protecting group, such as NV or NP or their
corresponding methyl derivatives, MeNV or MeNP, respectively, on
the carboxy terminus of the amino acid or 5'-hydroxy terminus of
the nucleotide, is formed by acylating the carboxy terminus or
5'-OH with an activated benzyl derivative of the protecting group.
Examples of activated benzyl derivatives of MeNV and MeNP have the
general formula: 14
[0474] where X is halogen, hydroxyl, tosyl, mesyl, trifluoromethyl,
diazo, azido, and the like.
[0475] Another method for generating protected monomers is to react
the benzylic alcohol derivative of the protecting group with an
activated ester of the monomer. For example, to protect the
carboxyl terminus of an amino acid, an activated ester of the amino
acid is reacted with the alcohol derivative of the protecting
group, such as 6-nitroveratrol (NVOH). Examples of activated esters
suitable for such uses include halo-formate, mixed anhydride,
imidazoyl formate, acyl halide, and also include formation of the
activated ester in situ the use of common reagents such as DCC and
the like. See Atherton et al. for other examples of activated
esters.
[0476] A further method for generating protected monomers is to
react the benzylic alcohol derivative of the protecting group with
an activated carbon of the monomer. For example, to protect the
5'-hydroxyl group of a nucleic acid, a derivative having a
5'-activated carbon is reacted with the alcohol derivative of the
protecting group, such as methyl-6-nitropiperonol (MeNPOH).
Examples of nucleotides having activating groups attached to the
5'-hydroxyl group have the general formula: 15
[0477] where Y is a halogen atom, a tosyl, mesyl, trifluoromethyl,
azido, or diazo group, and the like.
[0478] Another class of preferred photochemical protecting groups
has the formula: 16
[0479] where R.sup.1, R.sup.2, and R.sup.3 independently are a
hydrogen atom, a lower alkyl, aryl, benzyl, halogen, hydroxyl,
alkoxyl, thiol, thioether, amino, nitro, carboxyl, formate,
formamido, sulfanates, sulfido or phosphido group, R.sup.4 and
R.sup.5 independently are a hydrogen atom, an alkoxy, alkyl, halo,
aryl, or alkenyl group, and n=0 or 1.
[0480] A preferred protecting group, 1-pyrenylmethyloxycarbonyl
(PyROC), which is used to protect the amino terminus of an amino
acid, or the hydroxyl group of nucleotide, sugar or carbohydrate
for example, is formed when R.sup.1 through R.sup.5 are each a
hydrogen atom and n=1: 17
[0481] Another preferred protecting group, 1-pyrenylmethyl (PyR),
which is used for protecting the carboxy terminus of an amino acid
or the hydroxyl group of a nucleotide, sugar or carbohydrate for
example, is formed when R.sup.1 through R.sup.5 are each a hydrogen
atom and n=0: 18
[0482] An amino acid having a pyrenylmethyloxycarbonyl protecting
group on its amino terminus is formed by acylation of the free
amine of amino acid with an activated oxycarbonyl ester of the
pyrenyl protecting group. Examples of activated oxycarbonyl esters
of PyROC have the general formula: 19
[0483] where X is halogen, or mixed anhydride, p-nitrophenoxy, or
N-hydroxysuccinimide group, and the like.
[0484] A protected amino acid or nucleotide having a
photoactivatable protecting group, such as PyR, on the carboxy
terminus of the amino acid or 5'-hydroxy terminus of the nucleic
acid, respectively, is formed by acylating the carboxy terminus or
5'-OH with an activated pyrenylmethyl derivative of the protecting
group. Examples of activated pyrenylmethyl derivatives of PyROC
have the general formula: 20
[0485] where X is a halogen atom, a hydroxyl, diazo, or azido
group, and the like.
[0486] Another method of generating protected monomers is to react
the pyrenylmethyl alcohol moiety of the protecting group with an
activated ester of the monomer. For example, an activated ester of
an amino acid can be reacted with the alcohol derivative of the
protecting group, such as pyrenylmethyl alcohol (PyROH), to form
the protected derivative of the carboxy terminus of the amino acid.
Examples of activated esters include halo-formate, mixed anhydride,
imidazoyl formate, acyl halide, and also include formation of the
activated ester in situ and the use of common reagents such as DCC
and the like.
[0487] Clearly, many photosensitive protecting groups are suitable
for use in the present invention.
[0488] In preferred embodiments, the substrate is irradiated to
remove the photoremovable protecting groups and create regions
having free reactive moieties and side products resulting from the
protecting group. The removal rate of the protecting groups depends
on the wavelength and intensity of the incident radiation, as well
as the physical and chemical properties of the protecting group
itself. Preferred protecting groups are removed at a faster rate
and with a lower intensity of radiation. For example, at a given
set of conditions, MeNVOC and MeNPOC are photolytically removed
from the N-terminus of a peptide chain faster than their
unsubstituted parent compounds, NVOC and NPOC, respectively.
[0489] Removal of the protecting group is accomplished by
irradiation to separate the reactive group and the degradation
products derived from the protecting group. Not wishing to be bound
by theory, it is believed that irradiation of an NVOC- and
MeNVOC-protected oligomers occurs by the following reaction
schemes:
[0490]
NVOC-AA->3,4-dimethoxy-6-nitrosobenzaldehyde+CO.sub.2+AA
[0491]
MeNVOC-AA->3,4-dimethoxy-6-nitrosoacetophenone+CO.sub.2+AA
[0492] where AA represents the N-terminus of the amino acid
oligomer.
[0493] Along with the unprotected amino acid, other products are
liberated into solution: carbon dioxide and a
2,3-dimethoxy-6-nitrosophenylcarbonyl compound, which can react
with nucleophilic portions of the oligomer to form unwanted
secondary reactions. In the case of an NVOC-protected amino acid,
the degradation product is a nitrosobenzaldehyde, while the
degradation product for the other is a nitrosophenyl ketone. For
instance, it is believed that the product aldehyde from NVOC
degradation reacts with free amines to form a Schiff base (imine)
that affects the remaining polymer synthesis. Preferred
photoremovable protecting groups react slowly or reversibly with
the oligomer on the support.
[0494] Again not wishing to be bound by theory, it is believed that
the product ketone from irradiation of a MeNVOC-protected oligomer
reacts at a slower rate with nucleophiles on the oligomer than the
product aldehydes from irradiation of the same NVOC-protected
oligomer. Although not unambiguously determined, it is believed
that this difference in reaction rate is due to the difference in
general reactivity between aldehydes and ketones towards
nucleophiles due to steric and electronic effects.
[0495] The photoremovable protecting groups of the present
invention are readily removed. For example, the photolysis of
N-protected L-phenylalanine in solution having different
photoremovable protecting groups was analyzed, and the results are
presented in the following table:
10TABLE 9 Photolysis of Protected L-Phe-OH t.sub.1/2 in seconds
Solvent NBOC NVOC MeNVOC MeNPOC Dioxane 1288 110 24 19 5 mM
H.sub.2SO.sub.4/Dioxane 1575 98 33 22
[0496] The half life, t.sub.1/2, is the time in seconds required to
remove 50% of the starting amount of protecting group. NBOC is the
6-nitrobenzyloxycarbonyl group, NVOC is the
6-nitroveratryloxycarbonyl group, MeNVOC is the
methyl-6-nitroveratryloxycarbonyl group, and MeNPOC is the
methyl-6-nitropiperonyloxycarbonyl group. The concentration of each
protected phenylalanine was 0.10 mM.
[0497] Table 9 shows that deprotection of NVOC-, MeNVOC-, and
MeNPOC-protected phenylalanine proceeded faster than the
deprotection of NBOC. Furthermore, it shows that the deprotection
of the two derivatives that are substituted on the benzylic carbon,
MeNVOC and MeNPOC, were photolyzed at the highest rates in both
dioxane and acidified dioxane.
[0498] 1. Use of Photoremovable Groups During Solid-Phase Synthesis
of Peptides
[0499] The formation of peptides on a solid-phase support requires
the stepwise attachment of an amino acid to a substrate-bound
growing chain. In order to prevent unwanted polymerization of the
monomeric amino acid under the reaction conditions, protection of
the amino terminus of the amino acid is required. After the monomer
is coupled to the end of the peptide, the N-terminal protecting
group is removed, and another amino acid is coupled to the chain.
This cycle of coupling and deprotecting is continued for each amino
acid in the peptide sequence. See Merrifield, J. Am. Chem. Soc.
(1963) 85:2149, and Atherton et al., "Solid Phase Peptide
Synthesis" 1989, IRL Press, London, both incorporated herein by
reference for all purposes. As described above, the use of a
photoremovable protecting group allows removal of selected portions
of the substrate surface, via patterned irradiation, during the
deprotection cycle of the solid phase synthesis. This selectively
allows spatial control of the synthesis--the next amino acid is
coupled only to the irradiated areas.
[0500] In one embodiment, the photoremovable protecting groups of
the present invention are attached to an activated ester of an
amino acid at the amino terminus: 21
[0501] where R is the side chain of a natural or unnatural amino
acid, X is a photoremovable protecting group, and Y is an activated
carboxylic acid derivative. The photoremovable protecting group, X,
is preferably NVOC, NPOC, PyROC, MeNVOC, MeNPOC, and the like as
discussed above. The activated ester, Y, is preferably a reactive
derivative having a high coupling efficiency, such as an acyl
halide, mixed anhydride, N-hydroxysuccinimide ester, HOBT ester
perfluorophenyl ester, or urethane protected acid, and the like.
Other activated esters and reaction conditions are well known (See
Atherton et al.).
[0502] 2. Use of Photoremovable Protecting
[0503] Groups During Solid-Phase Synthesis of Oligonucleotides
[0504] The formation of oligonucleotides on a solid-phase support
requires the stepwise attachment of a nucleotide to a
substrate-bound growing oligomer. In order to prevent unwanted
polymerization of the monomeric nucleotide under the reaction
conditions, protection of the 5'-hydroxyl group of the nucleotide
is required. After the monomer is coupled to the end of the
oligomer, the 5'-hydroxylprotecting group is removed, and another
nucleotide is coupled to the chain. This cycle of coupling and
deprotecting is continued for each nucleotide in the oligomer
sequence. See Gait, "Oligonucleotide Synthesis: A Practical
Approach" 1984, IRL Press, London, incorporated herein by reference
for all purposes. As described above, the use of a photoremovable
protecting group allows removal, via patterned irradiation, of
selected portions of the substrate surface during the deprotection
cycle of the solid phase synthesis. This selectively allows spatial
control of the synthesis--the next nucleotide is coupled only to
the irradiated areas.
[0505] Oligonucleotide synthesis generally involves coupling an
activated phosphorous derivative on the 3'-hydroxyl group of a
nucleotide with the 5'-hydroxyl group of an oligomer bound to a
solid support. Two major chemical methods exist to perform this
coupling: the phosphate-triester and phosphoramidite methods (See
Gait). Protecting groups of the present invention are suitable for
use in either method.
[0506] In a preferred embodiment, a photoremovable protecting group
is attached to an activated nucleotide on the 5'-hydroxyl group:
22
[0507] where B is the base attached to the sugar ring; R is a
hydrogen atom when the sugar is deoxyribose or R is a hydroxyl
group when the sugar is ribose; P represents an activated
phosphorous group; and X is a photoremovable protecting group. The
photoremovable protecting group, X, is preferably NV, NP, PyR,
MeNV, MeNP, NVOC, NPOC, PyROC, MeNVOC, MeNPOC, and the like as
described above. The activated phosphorous group, P, is preferably
a reactive derivative having a high coupling efficiency, such as a
phosphate-triester, phosphoramidite or the like. Other activated
phosphorous derivatives, as well as reaction conditions, are well
known (See Gait).
[0508] E. Amino Acid N-Carboxy Anhydrides Protected with a
Photoremovable Group
[0509] During Merrifield peptide synthesis, an activated ester of
one amino acid is coupled with the free amino terminus of a
substrate-bound oligomer. Activated esters of amino acids suitable
for the solid phase synthesis include halo-formate, mixed
anhydride, imidazoyl formate, acyl halide, and also includes
formation of the activated ester in situ and the use of common
reagents such as DCC and the like (See Atherton et al.). A
preferred protected and activated amino acid has the general
formula: 23
[0510] where R is the side chain of the amino acid and X is a
photoremovable protecting group. This compound is a
urethane-protected amino acid having a photoremovable protecting
group attached to the amine. A more preferred activated amino acid
is formed when the photoremovable protecting group has the general
formula: 24
[0511] where R.sup.1, R.sup.2, R.sup.3, and R.sup.4 independently
are a hydrogen atom, a lower alkyl, aryl, benzyl, halogen,
hydroxyl, alkoxyl, thiol, thioether, amino, nitro, carboxyl,
formate, formamido or phosphido group, or adjacent substituents
(i.e., R.sup.1-R.sup.2, R.sup.2-R.sup.3, R.sup.3-R.sup.4) are
substituted oxygen groups that together form a cyclic acetal or
ketal; and R.sup.5 is a hydrogen atom, alkoxyl, alkyl, halo, aryl,
or alkenyl group.
[0512] A preferred activated amino acid is formed when the
photoremovable protecting group is 6-nitroveratryloxycarbonyl. That
is, R.sup.1 and R.sup.4 are each a hydrogen atom, R.sup.2 and
R.sup.3 are each a methoxy group, and R.sup.5 is a hydrogen atom.
Another preferred activated amino acid is formed when the
photoremovable group is 6-nitropiperonyl: R.sup.1 and R.sup.4 are
each a hydrogen atom, R.sup.2 and R.sup.3 together form a methylene
acetal, and R.sup.5 is a hydrogen atom. Other protecting groups are
possible. Another preferred activated ester is formed when the
photoremovable group is methyl-6-nitroveratryl or
methyl-6-nitropiperonyl- .
[0513] Another preferred activated amino acid is formed when the
photoremovable protecting group has the general formula: 25
[0514] where R.sup.1, R.sup.2, and R.sup.3 independently are a
hydrogen atom, a lower alkyl, aryl, benzyl, halogen, hydroxyl,
alkoxyl, thiol, thioether, amino, nitro, carboxyl, formate,
formamido, sulfanate, sulfido or phosphido group, and R.sup.4 and
R.sup.5 independently are a hydrogen atom, an alkoxy, alkyl, halo,
aryl, or alkenyl group. The resulting compound is a
urethane-protected amino acid having a pyrenylmethyloxycarbonyl
protecting group attached to the amine. A more preferred embodiment
is formed when R.sup.1 through R.sup.5 are each a hydrogen
atom.
[0515] The urethane-protected amino acids having a photoremovable
protecting group of the present invention are prepared by
condensation of an N-protected amino acid with an acylating agent
such as an acyl halide, anhydride, chloroformate and the like (See
Fuller et al., U.S. Pat. No. 4,946,942 and Fuller et al., J. Amer.
Chem. Soc. (1990) 112:7414-7416, both herein incorporated by
reference for all purposes).
[0516] Urethane-protected amino acids having photoremovable
protecting groups are generally useful as reagents during
solid-phase peptide synthesis, and because of the spatial
selectivity possible with the photoremovable protecting groups, are
especially useful for the spatially addressable peptide synthesis.
These amino acids are difunctional: the urethane group first serves
to activate the carboxy terminus for reaction with the amine bound
to the surface, and, once the peptide bond is formed, the
photoremovable protecting group protects the newly formed amino
terminus from further reaction. These amino acids are also highly
reactive to nucleophiles, such as deprotected amines on the surface
of the solid support, and due to this high reactivity, the
solid-phase peptide coupling times are significantly reduced, and
yields are typically higher.
[0517] IV. Data Collection
[0518] A. Data Collection System
[0519] Substrates prepared in accordance with the above description
are used in one embodiment to determine which of the plurality of
sequences thereon bind to a receptor of interest. FIG. 10
illustrates one embodiment of a device used to detect regions of a
substrate which contain fluorescent markers. This device would be
used, for example, to detect the presence or absence of a
fluorescently labeled receptor such as an antibody which has bound
to a synthesized polymer on a substrate.
[0520] Light is directed at the substrate from a light source 1002
such as a laser light source of the type well known to those of
skill in the art such as a model no. 2025 made by Spectra Physics.
Light from the source is directed at a lens 1004 which is
preferably a cylindrical lens of the type well known to those of
skill in the art. The resulting output from the lens 1004 is a
linear beam rather than a spot of light. Thus, data can be detected
substantially simultaneously along a linear array of pixels rather
than on a pixel-by-pixel basis. It will be understood that while a
cylindrical lens is used herein as an illustration of one technique
for generating a linear beam of light on a surface, other
techniques could also be utilized.
[0521] The beam from the cylindrical lens is passed through a
dichroic mirror or prism and directed at the surface of the
suitably prepared substrate 1008. Substrate 1008 is placed on an
x-y translation stage 1009 such as a model no. PM500-8 made by
Newport. Certain locations on the substrate will fluoresce and
fluorescence will be transmitted along the path indicated by dashed
lines back through the dichroic mirror, and focused with a suitable
lens 1010 such as an f/1.4 camera lens on a linear detector 1012
via a variable f stop focusing lens 1014. Through use of a linear
light beam, it becomes possible to generate data over a line of
pixels (such as about 1 cm) along the substrate, rather than from
individual points on the substrate. In alternative embodiments,
light is directed at a 2-dimensional area of the substrate and
fluorescence is detected by a 2-dimensional CCD array. Linear
detection is preferred because substantially higher power densities
are obtained.
[0522] Detector 1012 detects the amount of fluorescence emitted
from the substrate as a function of position. According to one
embodiment the detector is a linear CCD array of the type commonly
known to those of skill in the art. The x-y translation stage, the
light source, and the detector 1012 are all operably connected to a
computer 1016 such as an IBM PC-AT or equivalent for control of the
device and data collection from the CCD array.
[0523] In operation, the substrate is appropriately positioned by
the translation stage. The light source is then illuminated, and
fluorescence intensity data are gathered with the computer via the
detector.
[0524] In a preferred embodiment, the substrate and x/y translation
table are placed under a microscope which includes one or more
objectives. Light (488 nm) from a laser, which in some embodiments
is a model no. 2020-05 argon ion laser manufactured by Spectra
Physics, is directed at the substrate by a dichroic mirror which
passes greater than about 520 nm light but reflects 488 nm light.
The dichroic mirror may be, for example, a model no. FT510
manufactured by Carl Zeiss. Light reflected from the mirror then
enters the microscope which may be, for example, a model no.
Axioskop 20 manufactured by Carl Zeiss. Fluorescein-marked
materials on the substrate will fluoresce >520 nm light, and the
fluoresced light will be collected by the microscope and passed
through the mirror. The fluorescent light from the substrate is
then directed through a wavelength filter and, thereafter through
an aperture plate. The wavelength filter may be, for example, a
model no. OG530 manufactured by Melles Griot and the aperture plate
may be, for example, a model no. 477352/477380 manufactured by Carl
Zeiss.
[0525] The fluoresced light then enters a photomultiplier tube
which in some embodiments is a model no. R943-02 manufactured by
Hamamatsu, the signal is amplified in a preamplifier and photons
are counted by a photon counter. The number of photons is recorded
as a function of the location in the computer. The pre-amp may be,
for example, a model no. SR445 manufactured by Stanford Research
Systems and the photon counter may be a model no. SR430
manufactured by Stanford Research Systems. The substrate is then
moved to a subsequent location and the process is repeated. In
preferred embodiments the data are acquired every 1 to 100 .mu.m
with a data collection diameter of about 0.8 to 10 .mu.m preferred.
In embodiments with sufficiently high fluorescence, a CCD detector
with broadfield illumination is utilized.
[0526] FIG. 36 illustrates the architecture of the data collection
system in greater detail. Operation of the system occurs under the
direction of the photon counting program 1102. The user inputs the
scan dimensions, the number of pixels or data points in a region,
and the scan speed to the counting program. Via a GPIB bus 1104 the
program (in an IBM PC compatible computer, for example) interfaces
with a multichannel scaler 1106 such as a Stanford Research SR 430
and an x-y stage controller 1108 such as a Newport PM500. The
signal from the light from the fluorescing substrate enters a
photomultiplier 1110, providing output to the scaler 1106. Data are
output from the scaler indicative of the number of counts in a
given region. After scanning a selected area, the stage controller
is activated with commands for acceleration and velocity, which in
turn drives the scan stage 1112 such as a Newport PM500-A to
another region.
[0527] Data are collected in an image data file 1114 and processed
in a scaling program 1116. A scaled image is output for display on,
for example, a VGA display 1118. The image is scaled based on an
input of the percentage of pixels to clip and the minimum and
maximum pixel levels to be viewed. The system outputs for use the
min and max pixel levels in the raw data.
[0528] B. Data Analysis
[0529] The output from the data collection system is an array of
data indicative of fluorescence intensity versus location on the
substrate. The data are typically taken over regions substantially
smaller than the area in which synthesis of a given polymer has
taken place. Merely by way of example, if polymers were synthesized
in squares on the substrate having dimensions of 500 microns by 500
microns, the data may be taken over regions having dimensions of 5
microns by 5 microns. In most preferred embodiments, the regions
over which florescence data are taken across the substrate are less
than about 1/2 the area of the regions in which individual polymers
are synthesized, preferably less than {fraction (1/10)} the area in
which a single polymer is synthesized, and most preferably less
than {fraction (1/100)} the area in which a single polymer is
synthesized. Hence, within any area in which a given polymer has
been synthesized, a large number of fluorescence data points are
collected.
[0530] A plot of the number of pixels versus fluorescence intensity
for a scan of a cell when it has been exposed to, for example, a
labeled antibody will typically take the form of a bell curve, but
spurious data are observed, particularly at higher intensities.
Since it is desirable to use an average of fluorescence intensity
over a given synthesis region in determining relative binding
affinity, these spurious data will tend to undesirably skew the
data.
[0531] Accordingly, in one embodiment of the invention the data are
corrected for removal of these spurious data points, and an average
of the data points is thereafter utilized in determining relative
binding efficiency.
[0532] FIG. 37 illustrates one embodiment of a system for removal
of spurious data from a set of fluorescence data such as data used
in affinity screening studies. A user or the system inputs data
relating to the chip location and cell corners at step 1302. From
this information and the image file, the system creates a computer
representation of a histogram at step 1304, the histogram (at least
in the form of a computer file) plotting number of data pixels
versus intensity.
[0533] For each cell, a main data analysis loop is then performed.
For each cell, at step 1306, the system calculates the total
fluorescence intensity or number of pixels for the bandwidth
centered around varying intensity levels. For example, as shown in
the plot to the right of step 1306, the system calculates the
number of pixels within the band of width w. The system then
"moves" this bandwidth to a higher center intensity, and again
calculates the number of pixels in the bandwidth. This process is
repeated until the entire range of intensities have been scanned,
and at step 1308 the system determines which band has the highest
total number of pixels. The data within this bandwidth are used for
further analysis. Assuming the bandwidth is selected to be
reasonably small, this procedure will have the effect of
eliminating spurious data located at the higher intensity levels.
The system then repeats at step 1310 if ail cells have been
evaluated, or repeats for the next cell.
[0534] At step 1312 the system then integrates the data within the
bandwidth for each of the selected cells, sorts the data at step
1314 using the synthesis procedure file, and displays the data to a
user on, for example, a video display or a printer.
C. Alternative Embodiments
[0535] Alternative embodiments of the detection system will be used
according to some embodiments of the invention. According to one
embodiment of the invention, a slit scanning fluorescence detection
system is used in imaging VLSIPS.TM. chips. Such systems may have
improved sensitivity, resolution, contrast, speed of data
acquisition, etc. as compared to a pinhole system. Such systems
have improved speed of data acquisition, since the image of the
VLSIPS.TM. chip will be constructed of strips rather than scan
lines. High resolution scans of VLSIPS.TM. chips can take over an
hour to acquire with a pinhole system. The advantages of the slit
scanning approach depends on the size of the imaged illumination
slit (limited in practice by the size of the detector and the
magnification of the optical system) and the sensitivity and dark
noise of the linear detector used (i.e. how fast the detector
is).
[0536] Better optical sectioning and hence reduction of background
have been reported for a beam-scanning version of this approach.
Theoretical calculations and experimental measurements for a
scanning mirror/slit microscope using a divided aperture indicate
that the contribution to image formation by out-of-plane light
scattering/emitting elements will fall off faster with distance
than in the case of a pinhole aperture system (Koester (1989), in
Handbook of Biological Confocal Microscopy, pp 207-214, edited by
J. Pawley, Plenum Press, NY, incorporated herein by reference for
all purposes). Beam scanning arrangements are unlikely to be of use
in near future because of the limited field of view obtained.
[0537] Slit scanning systems have been built by others (see
references in Handbook of Biological Confocal Microscopy, Chapters
1 and 19, incorporated herein by reference).
[0538] Other improvements may be made to the system described
elsewhere herein. For example, improved image contrast may be
obtained by using dielectric barrier filters and a higher numerical
aperture microscope objective. These combined modifications improve
the detection limit by reducing background fluorescence and laser
light scattering. A difficulty in using higher numerical aperture
objectives is the shallow depth-of-field, which leads to
out-of-focus scans if the thickness of the substrates is
nonuniform. As a first step towards solving this problem, a
multiple focus and extrapolated x-axis control mechanism may be
utilized.
[0539] An alternative solution to this problem involves the use of
a piezoelectric focusing system which moves the monitor focus (e.g.
by bouncing a focused laser off the surface and detecting
positional variation with a diode array), a piezoelectric control
device may be used to provide real-time autofocus capability.
[0540] Alternate fluorophores would also be beneficial in some
embodiments. Fluorophores should be evaluated in terms of relative
quantum yield, photobleaching stability, and detection sensitivity
achieved under the scan conditions we use. To optimize detection
sensitivity, fluorophores with larger Stokes' shifts allow better
discrimination between emitted light and scattered laser light. In
addition, they are less subject to self-quenching phenomena at high
packing densities and hence should provide better quantitation of
the relative number of fluorophores bound to the surface.
Fluorophores with other excitation parameters may also be desirable
in some embodiments. For example, Pharmingen makes available of a
fluorophore which can be excited with the argon-ion laser and emits
above 670 nm.
[0541] According to the work of Hirschfeld (Appl. Optics (1976)
15:3135-3139, incorporated herein by reference), the integrated
fluorescent emission obtained upon complete bleaching of a
fluorescent tag is independent of fluorescence quantum efficiency,
absorption cross-section, and illumination intensity. Hence, this
approach offers high sensitivity and better quantitation (a greater
number of fluorophores can be attached to an antibody or packed
into a small area without loss of signal due to quenching), and
measurements should be less sensitive to errors in focusing than
the scanning approach. The method is probably most suitable for
sequential sampling of a small number of sites on a VLSIPS.TM.
surface, although it should also be possible to build a device with
a two-dimensional detector for simultaneous readout of many sites,
provided that good rejection of excitation light and background
fluorescence is achieved.
[0542] Time-resolved fluorescence provides an additional approach
to enhanced sensitivity through background reduction. Background
fluorescence in biological samples usually decays on the time scale
of nano- to microseconds. Pulsed excitation of a fluorescent tag
having a long lifetime can be detected with high sensitivity by
gating the detector so that emitted light is measured after the
background has decayed. Immunoassays have been developed using this
approach with sensitivities that are reported to approach that of
radioisotopic methods (Soini and Kojola (1983) Clin. Chem. (1983)
29:65-68, incorporated herein by reference). This mode of detection
is particularly attractive for two reasons. First, considerable
experience with rare-earth chelates having long fluorescence
lifetimes is available (on the order of 1 msec), and may be able to
provide novel additional compounds. Second, this approach allows
one to image the VLSIPS.TM. surface using a two-dimensional
detector (e.g., on CCD), resulting in reduced data acquisition time
and an instrument that is likely to be much easier to utilize. The
use of two-dimensional detectors in our present system is precluded
by the necessity of using a pinhole aperture in front of the
detector to achieve the confocal condition necessary for reduced
background. For example, terbium and europium chelates are
typically excited in the UV (320-340 nm), and emit at much longer
wavelengths (545 nm for terbium and 615 nm for europium). Thus,
they offer excellent wavelength discrimination as well as
time-resolved discrimination against background fluorescence. To
utilize these compounds a pulsed excitation source is utilized
(e.g. an acousto-optically modulated argon-ion laser, flash lamp,
or pulsed laser), a gated detector, and timing instrumentation.
[0543] Chemiluminescence has been reported to provide detection
sensitivities comparable to that achieved using radioisotopes, and
several products that can be chemically or enzymatically triggered
to emit light are commercially available. High detection
sensitivity results from measurement of signals against "zero"
background (i.e. there's no background excitation light). In
preliminary measurements using a commercially-available
chemiluminescent substrate for alkaline phosphatase (Lumi-Phos 530;
Boehringer-Mannheim #1275-470), it was possible to detect
approximately 10 .mu.moles of enzyme in a 600 microliter sample
using the detection end of the Aminco spectrofluorimeter. The
lifetime of the unstable dioxetane intermediate in those
experiments appeared to be too long to permit useful imaging of a
VLSIPS.TM. surface in some embodiments (diffusion of the substrate
would create a resolution problem if one used immobilized enzyme),
but other compounds may have shorter lifetimes. Alternatively,
using tethered substrate (one photon maximum released per surface
site) may provide enough light if the collection efficiency were
extremely high. In some embodiments, sandwiching a chemiluminescent
probe-labeled VLSIPS.TM. chip between a CCD detector and a mirror
to maximize collection efficiency may be utilized, while a fiber
optic faceplate between chip and detector may be utilized to
minimize cross-talk.
[0544] Other approaches may be utilized in detection of
receptor-ligand interactions in some embodiments, e.g. ChemFETS.
ChemFETS are semiconductor devices in which the current flowing
through the device is modulated by electrostatic interactions
between ions in solution and a region of the surface called the
"gate". A multiple gate device may be utilized in some embodiments
in which different macromolecules (e.g. receptors or antibodies)
are immobilized at each gate. Detection of ligand binding may then
be possible, either directly for charged ligands, or by using an
enzyme-antibody conjugate that gives rise to local pH changes, by
monitoring current for each gate region. A related device has
recently been commercialized by Molecular Devices.
[0545] V. Other Representative Applications
[0546] A. Oligonucleotide Synthesis
[0547] The generality of light directed spatially addressable
parallel chemical synthesis is demonstrated by application to
nucleic acid synthesis.
[0548] 1. Example
[0549] Light activated formation of a thymidine-cytidine dimer was
carried out. A three dimensional representation of a fluorescence
scan showing a 7 square by 4 square checkerboard pattern generated
by the light-directed synthesis of a dinucleotide was produced.
5'-nitroveratryl thymidine was attached to a synthesis substrate
through the 3' hydroxyl group. The nitroveratryl protecting groups
were removed by illumination through a 500 .mu.m checkerboard mask.
The substrate was then treated with phosphoramidite activated
2'-deoxycytidine. In order to follow the reaction fluorometrically,
the deoxycytidine had been modified with an FMOC protected
aminohexyl linker attached to the exocyclic amine
(5'-O-dimethoxytrityl-4-N-(6-N-fluorenylmethylcarbamoyl-hexylcarboxy)-2'--
deoxycytidine). After removal of the FMOC protecting group with
base, the regions which contained the dinucleotide were
fluorescently labelled by treatment of the substrate with 1 mM FITC
in DMF for one hour.
[0550] The three-dimensional representation of the fluorescence
intensity data showing alternating squares of bright raised pixels
reproduces the checkerboard illumination pattern used during
photolysis of the substrate. This result demonstrates that
oligonucleotides as well as peptides can be synthesized by the
light-directed method.
[0551] 2. Example
[0552] In another example the light-activated formation of
thymidine-cytidine-cytidine was carried out as shown in FIG. 38.
Here, as in the previous example, 5'-nitroveratryl thymidine was
attached to the substrate, via phosphoramidite chemistry to a
surface containing [Bis (2-hydroxyethyl)-3-aminopropylsiloxane].
The slide was then uniformly illuminated (362 nm at .about.14
mW/cm.sup.2) for 10 minutes in the presence of dioxane. After
drying, the surface was then treated with
N,4-dimethoxytrityl-5'-nitroveratryl-2'-deoxycytidine-3'-O-(2-cyanoethyl)-
-N,N-diisopropylphosphoramidite in the presence of tetrazole
(standard phosphoramidite coupling chemistry). After oxidizing and
drying, the plate was again illuminated as before except that a 500
.mu.m checkerboard mask was placed between the light source and the
slide. The surface was then exposed to
5'-O-(4,4'-Dimethoxy)-N-4-(6-((Biotinoyl)amin-
o)hexanoyl)amino)hexanoyl,
aminohexyl)-5-methyl-2'-deoxycytidine-3'-O-(2-c-
yanoethyl)-N,N-diisopropylphosphoramidite with tetrazole. After
oxidizing and drying, the areas which contained the trinucleotide
were fluorescently labelled by treatment with FITC labeled
streptavidin. A resulting representation of the fluorescence
intensity data showed alternating bright and dark squares
corresponding to the 500 .mu.m and checkerboard illumination
pattern used during photolysis.
[0553] 3. Example
[0554] In another example, an 8 nucleotide, poly-adenine oligomer
was prepared and later hybridized with a poly thymidine probe. The
synthesis was carried out as follows.
[0555] Ten 1.times.3" slides were incubated in a plastic jar filled
with 1% bis(2-hydroxylethyl)aminopropyltriethoxysilane in 95%
ethanol overnight at room temperature. The slides were then rinsed
thoroughly with ethanol, dried with N.sub.2 and baked at
110.degree. C. for 1 hour and put in a vacuum desiccator to cool. A
surface linker for coupling was prepared by mixing equal volumes of
0.2M monodimethoxytritylpentaethylene-
glycol-.beta.-cyanoethylphosphoramidite in anhydrous acetonitrile
and 0.45M tetrazole/CH.sub.3CN in a glass vial. 0.35 mL of this
solution was then dispensed onto the surface of each slide and
incubated for 3 min. The slide was rinsed briefly with CH.sub.3CN
and coupling was repeated with freshly prepared phosphoramidite.
Next, the phosphite-triester bond was oxidized to a phosphotriester
by dipping the slides into a jar filled with 0.1M iodine solution
(2.6 g iodine+80 mL tetrahydrofuran+20 mL 2,6-lutidine+2 mL
h.sub.2O) for 1 min, followed by rinsing thoroughly with CH.sub.3CN
and drying with N.sub.2. The dimethoxytrityl protecting groups were
removed by dipping the slides in a staining jar filled with 3%
dichloroacetic acid in methylene chloride for 30 sec followed by
rinsing with CH.sub.3CH and drying with N.sub.2. Steps C and D were
then repeated with 0.2M
5'-nitroveratyl-deoxythymidine-3'-.beta.-cyanoethylpho-
sphoramidite. The slides were then incubated for 1 hour in a
staining jar filled with capping solution (75 mLs 6.5% DMAP/THF+25
mLs 40% acetic anhydride/60% 2,6-lutidine), and then rinsed
thoroughly with CH.sub.3CN, dried with N.sub.2, and stored in the
dark under vacuum.
[0556] Polydeoxyadenine (poly-dA.sub.12) was prepared on an ABI
synthesizer using 1 .mu.mole 3'-amine-ON CPG (Glen Research) and
the standard ABI 1 .mu.mole coupling cycle having final DMT on. The
CPG was transferred to a 1.5 mL plastic screw-cap vial where 11.0
mL conc. NH.sub.4OH was added. The mixture was incubated for 18
hours at 55.degree. C. to cleave the oligomer from the resin and
remove the exocyclic amine protecting groups. The crude oligomer
was purified using a PolyPak cartridge (Glen Research) according to
the protocol supplied with the columns. The appropriate fractions
were pooled and dried by speed-vac. The oligomer was dissolved in
0.9 mL H.sub.2O, to which 0.1 mL 10.times. labelling buffer (1.0M
NaHCO.sub.3/Na.sub.2CO.sub.3, pH 9.0) and 0.25 mL freshly prepared
100 mg/mL FITC in DMF were added. The solution was then vortexed
and incubated overnight at room temperature. The reaction mixture
was purified on a 1.times.30 cm Sephadex G-25.TM. column using
H.sub.2O as the mobile phase, and the appropriate fractions were
pooled and dried by speed-vac. PAGE analysis showed that the
reaction was not 100% complete. The fluoresceinylated oligomer was
further purified by reverse phase HPLC (Hamilton PRP-1
semi-preparative column) using a linear gradient of 10-40%
CH.sub.3CN in 0.1M Triethylamine acetate, pH 7.6 over 45 min. at a
flow rate of 2 mL/min. Fractions were analyzed by PAGE, and the
appropriate fractions were pooled and dried by speed-vac. PAGE
analysis of the final product showed purity of approximately
99%.
[0557] A suitably derivatized slide was placed in a four-chamber
flow cell (wells approximately 1.5 cm dia. circles). One well was
filled with dioxane. All other wells were covered with black
electrical tape. The slide was then exposed to 365 nm light at 11.8
mW/cm.sup.2 for a period of 12 min. to remove the photoprotecting
groups. The flow cell was then attached to an ABI DNA synthesizer
(model 392), and 5'-NV-dT-OCEP coupled to the surface using the
modified cycle attached (cyc03 user). No capping step was performed
due to the excess phosphoramidite used. The photolysis/synthesis
cycles were repeated until poly-(dT).sub.12-OH was synthesized in
the well. The slide was removed from the flow cell, rinsed
thoroughly with CH.sub.3CN and dried with N.sub.2. The slide was
then incubated in 6.times.SSPE containing 3% BSA and 0.025% triton
X-100 for 30 min. at room temperature to block non-specific binding
sites. Next, the slide was transferred to a plastic container
filled with 20 ng/mL 5'-HO-poly(dA).sub.12-fluorescein in the same
buffer and incubated at room temperature for 1 hour. The slide was
briefly rinsed in a 20 mM NaCl solution, dried with N.sub.2, and
detected using the confocal microscopy system previously described.
Average photon counts in the well were 8-fold higher than the
background producing the image shown in FIG. 39A. The slide was
incubated overnight in 500 mLs 1.times.SSPE at 40.degree. C.,
rinsed and scanned. FIG. 39A shows a bright circle in the center
indicating that this wash removed the signal as shown in FIG. 39B.
Reprobing as before resulted in a signal with the same intensity as
that originally obtained as shown in FIG. 39C.
[0558] For sample 59-8b-1, -2, and -3, the average intensities were
347+/-25, 211+/-23 and 223+/-13, respectively, in the background.
For samples 59-8b-1 and 59-8b-3 the signal in the well was
4545+/-476 and 237+/-308, respectively. Therefore, for sample 1 the
signal/background ratio was 13, and it was 11 for sample number 3.
The sample number 59-8b-1 refers to the first probe. The sample
number 59-8b-2 refers to the scan after incubation at 40.degree. C.
The sample 59-8b-3 is the reprobe sample.
[0559] B. Oligosaccharide Synthesis
[0560] The present invention will find application in a wide
variety of additional applications including oligosaccharide
synthesis.
[0561] 1. Example
[0562] The potential synthesis of a C-glycoside unit that will be
attached to the surface is outlined below. FIG. 40A illustrates the
overall synthesis strategy.
[0563] The methyl glycoside of D-glucose is first converted through
a standard protection-deprotection sequence to the diacetate.
Selective esterification of the primary alcohol is then conducted,
followed by etherification of the secondary alcohol with a
photolabile protecting group and C-allylation, which provides the
C-glycoside. Oxidation of the olefin effects conversion to the
carboxylic acid which is then attached to the surface through
standard amide coupling. Photodeprotection will free a hydroxyl
group on the monosaccharide for the purposes of examining
glycosidic bond formation on the chip surface.
[0564] FIG. 40B illustrates formation of a simple activated
building block used especially for purposes of examining chemical
glycosidic bond formation. The building block is formed
D-glucose.
[0565] Conversion to the 6-amino derivative is effected through
fairly simple methods, such as shown in the first step of the
Figure. Protection of the amino function followed by exhaustive
acetylation provides the fully protected methyl glycoside.
Conversion to a activated glycosyl donor is done by hydrolysis and
conversion to the bromide. This material may be attached to a
substrate as shown in FIG. 40C. Coupling of the activated bromide
to the surface attached glycosyl acceptor will be monitored through
the agency of protected 6-amino function on the donor.
Photodeprotection followed by labeling with fluorescein
isothiocyanate will provide a sensitive assay for the formation of
the glycosidic bond.
[0566] Examination of enzymatic glycosidic bond formation requires
the synthesis of a nucleoside diphosphate sugar. One possible
scheme is illustrated in FIG. 40D. The introduction of the amino
group and exhaustive acetylation is accomplished by standard
methods such as those shown in the Figure. Protection of the amine
and glycoside hydrolysis leads to the free sugar which is now
activated via conversion to the bromide followed by displacement to
the anomeric phosphate derivative. Conversion to the uridine
diphosphate (UDP) derivative is accomplished through fairly
standard methodology, such as the one shown. The product may then
be attached to the substrate as shown in FIG. 40E.
[0567] Formation of the glycosidic bond between the nucleotide
diphosphate sugar and the immobilized saccharide on the VLSIPS.TM.
chip with galactosyl transferase may also be examined. Coupling of
the nucleoside glycoside donor to the surface attached glycosyl
acceptor will be monitored through the agency of the 6-amino
function on the donor. Labeling of the free amine with fluorescein
isothiocyanate will provide a sensitive assay for the formation of
the glycosidic bond.
[0568] C. Caged Binding Member System
[0569] According to an alternative embodiment of the invention, the
methods provide for attaching to the surface a caged binding member
which in its caged form has a relatively low affinity for other
potentially binding species, such as receptors and specific binding
substances. Such techniques are more fully described in copending
application U.S. Ser. No. 404,920, filed Sep. 8, 1989, and
incorporated herein by reference for all purposes.
[0570] According to this alternative embodiment, the invention
provides methods for forming predefined regions on a surface of a
solid support, wherein the predefined regions are capable of
immobilizing receptors. The methods make use of caged binding
members attached to the surface to enable selective activation of
the predefined regions. The caged binding members are liberated to
act as binding members ultimately capable of binding receptors upon
selective activation of the predefined regions. The activated
binding members are then used to immobilize specific molecules such
as receptors on the predefined region of the surface. The above
procedure is repeated at the same or different sites on the surface
so as to provide a surface prepared with a plurality of regions on
the surface containing, for example, the same or different
receptors. When receptors immobilized in this way have a
differential affinity for one or more ligands, screenings and
assays for the ligands can be conducted in the regions of the
surface containing the receptors.
[0571] The alternative embodiment may make use of novel caged
binding members attached to the substrate. Caged (unactivated)
members have a relatively low affinity for receptors of substances
that specifically bind to uncaged binding members when compared
with the corresponding affinities of activated binding members.
Thus, the binding members are protected from reaction until a
suitable source of energy is applied to the regions of the surface
desired to be activated. Upon application of a suitable energy
source, the caging groups labilize, thereby presenting the
activated binding member. A typical energy source will be
light.
[0572] Once the binding members on the surface are activated they
may be attached to a receptor. The receptor chosen may be a
monoclonal antibody, a nucleic acid sequence, a drug receptor, etc.
The receptor will usually, though not always, be prepared so as to
permit attaching it, directly or indirectly, to a binding member.
For example, a specific binding substance having a strong binding
affinity for the binding member and a strong affinity for the
receptor or a conjugate of the receptor may be used to act as a
bridge between binding members and receptors if desired. The method
uses a receptor prepared such that the receptor retains its
activity toward a particular ligand.
[0573] Preferably, the caged binding member attached to the solid
substrate will be a photoactivatable biotin complex, i.e., a biotin
molecule that has been chemically modified with photoactivatable
protecting groups so that it has a significantly reduced binding
affinity for avidin or avidin analogs than does natural biotin. In
a preferred embodiment, the protecting groups localized in a
predefined region of the surface will be removed upon application
of a suitable source of radiation to give binding members, that are
biotin or a functionally analogous compound having substantially
the same binding affinity for avidin or avidin analogs as does
biotin.
[0574] In another preferred embodiment, avidin or an avidin analog
is incubated with activated binding members on the surface until
the avidin binds strongly to the binding members. The avidin so
immobilized on predefined regions of the surface can then be
incubated with a desired receptor or conjugate of a desired
receptor. The receptor will preferably be biotinylated, e.g., a
biotinylated antibody, when avidin is immobilized on the predefined
regions of the surface. Alternatively, a preferred embodiment will
present an avidin/biotinylated receptor complex, which has been
previously prepared, to activated binding members on the
surface.
[0575] D. Fingerprinting for Quality Control
[0576] An alternative aspect of this invention involves testing a
therapeutic compound with an array of peptides or other biological
polymers to determine a characteristic binding pattern. Such a
characteristic binding pattern or "fingerprint" is used to monitor
the "constancy" of the compound over time by repeated testing with
the same array. As long as the fingerprint remains unchanged from
lot-to-lot, the bioprocess is producing the same product. If the
binding pattern of the therapeutic changes at any time, it would be
assumed that a subtle (or not so subtle) change in the therapeutic
compound had occurred. For example, changes in the glycosylation or
secondary structure of the protein could be detected. This method
promises to be particularly valuable with recombinant and other
products produced by fermentation processes where quality control
is problematic.
[0577] Preferably, the method would be performed with a very large
array of biological polymers (thousands or tens of thousands of
elements). A VLSIPS.TM. or caged biotin chip with binding
components to a very broad spectrum of polymer characteristics
would be employed. The elements of the array could thus be arranged
in a quasi-random manner. Alternatively, they might be organized in
a rational order. For example, certain physical properties (e.g.,
charge or hydrophobicity) of the constituent polymers might vary
gradually along a given spatial dimension. Chips having this
configuration could then be used for the QC of various biological
products. Custom chips specific for a single product, such as for
example tissue plasminogen activator (tPA), might also be
useful.
[0578] E. .beta.-Amino Acid and D-Amino Acid Monomers
[0579] The peptide diversity available in the present invention is
greatly increased by including non-natural amino acids (i.e. amino
acids which are not genetically coded) among the set of building
blocks. In particular, peptides containing at least one
.beta.-amino acids or D-amino acid residue may be synthesized by
the methods of the present invention. In addition, L-amino or
D-amino acids having modified side chains may be employed to
diversify the peptide products available. Cyclic .beta.-amino acid
monomers may be employed to reduce the increased conformational
mobility associated with acyclic .beta.-amino acids. In fact, an
ordered series of dihedral angles between the amido and carboxamido
groups of the peptide backbone may be obtained by changing the
number and constituent atoms of the cyclic .beta.-amino acid ring.
Such an ordered series of compounds may be useful in optimizing the
biological activity of a peptide drug. Since D-amino acids and
.beta.-amino acids of any sort are not subject to proteolytic
cleavage, incorporation of these residues into peptide drugs should
confer favorable properties of biological stability.
[0580] Amino acid monomers (D-, L- or .beta.-) with side chains
that are not found on the genetically coded amino acids may also be
used in preparing peptides according to the present invention.
Amino acids containing aromatic residues are of particular interest
because they are commonly present in biologically active binding
sites and in drugs. Although they are generally hydrophobic,
aromatic side chains can adopt a variety of electronic
configurations depending upon the substituents present.
[0581] Using phenylalanine as an example, a variety of
modifications to the side chain are available, some of which are
represented below. Each amino acid containing the groups below can
be employed as a monomer without requiring side chain protecting
groups often necessary for peptide synthesis. 26
[0582] Many of these compounds are commercially available. Those
that are not may be synthesized by a variety of well-known
procedures. A preferred choice is alkylation of acetamidomalonate
with the appropriate benzylic halide, followed by hydrolysis,
decarboxylation and enzymatic resolution with Aspergillus acylase I
as shown below and described in Chenault et al., J. Am. Chem. Soc.
(1989) 111:6354-6364 which is incorporated herein by reference for
all purposes. 27
[0583] Like genetically coded L-amino acids, D-amino acids must be
protected at the .alpha.-amino group during peptide synthesis. In
addition, the side chain may also have to be protected against
unwanted side reactions. The methods of protection set forth above
for L-amino acids can also be applied to D-isomers and amino acids
with modified side chains. The resulting protected monomers can
then be employed to synthesize an array of peptides (or peptide
analogs) using the protection-deprotection methods outlined
above.
[0584] Polymer backbones comprised of .beta.-amino acids have
several advantages. For example, they retain amide bonds, which
permits hydrogen bonding in all directions normal to the main
chain. The side chain density will be high for a given oligomer
length so long as the side chains are placed at both .alpha. and
.beta. carbons. Considerable control over the properties of the
peptide is gained by selecting appropriate substituents. For
example, attaching an alkyl group to the .beta. carbon, makes the
backbone more hydrophobic than the corresponding .alpha.-amino
acids. Additional control of the peptide conformation is also
possible. Although extra conformational freedom is permitted by the
.beta. carbon in straight chain .beta.-amino acids, conformational
restriction within the individual monomer units is also possible by
selecting appropriate side chains and cyclic groups, as will be
shown below.
[0585] Simple .beta.-amino acids which may be used in the present
invention will have an amino acid side chain at either the .alpha.
or .beta. carbon, and a methyl group or hydrogen atom at the other
(types I through IV shown below). Each of the methyl compounds
(types I and II below) will include four distinct stereoisomers for
each amino acid chain, representing a total of 160 compounds when
all of the genetically-coded side chains are employed. Replacing
the methyl group with a hydrogen atom (groups III and IV), reduces
the number of stereoisomers to two for each side chain. 28
[0586] Three methods for preparing .beta.-amino acids shown below
are summarized in Griffith, Annu. Rev. Biochem. (1989) 55:855-878
which is incorporated herein by reference for all purposes. The
condensation of cyanoacetic ester with carbonyl compounds or alkyl
halides followed by reduction provides structures of type III. The
Arndt-Eistert homologation of protected amino acids will give
compounds having the structures of type IV. Conjugate addition of
ammonia to .alpha.,.beta.-unsaturated esters will produce compounds
of types I-IV, depending on substituents in the ester starting
material. 29
[0587] Two other synthetic routes (shown below) which may be
employed to synthesize .beta.-amino acids take advantage of the
functional relation of .beta.-lactams (which have well-known
chemistries) and .beta.-amino acids. These methods are detailed in
various references, including Kamal et al., Heterocycles (1987)
26:1051-1076 and Hart et al., Chem. Rev. (1989) 89,:1447-1465, both
of which are incorporated herein by reference for all purposes. The
first synthetic route exploits the cycloaddition reactions of
chlorosulfonyl isocyanate (CSI) with alkenes to give, after
hydrolysis, .beta.-lactams. These can then be hydrolyzed to give
the corresponding the .beta.-amino acids. Because of the polar
mechanism for the CSI cycloaddition, it is not possible to use this
reaction to prepare type III compounds which have no substitution
on the carbon atom adjacent to the nitrogen atom. The second route
employs the condensation of enolates with imines, to produce
.beta.-lactams. Optically active compounds are provided this
method, but the basic reaction with diazomethane gives only type IV
structures. Type II structures may be prepared if diazomethane is
substituted with diazoethane and the resulting diastereomers are
separated. This synthesis route has the added advantage that it may
directly provide protected amino acids for peptide synthesis.
30
[0588] In situations where the most convenient methods do not offer
the opportunity for asymmetric synthesis (all but the Arndt-Eistert
and enolate-imine reactions described above), the .beta.-amino
acids must be resolved and assayed for optical purity. If the
methyl-substituted compounds (types I or II above) are prepared in
a non-stereoselective route, the diastereomers must also be
separated. Amides of type IV .beta.-amino acids may be
enantioselectively hydrolyzed by benzylpenicillinacylase. In
addition, .beta.-lactamases may be employed to enantiospecifically
hydrolyze .beta.-lactams to .beta.-amino acids. 31
[0589] Other methods of separating the isomers will be known to
those of skill in the art. In any event, analysis of the optical
purity of the products can be accomplished by chiral chromatography
or the Mosher method. The absolute configuration, which must be
determined for each compound prepared, can be assigned by
chiroptical methods.
[0590] One skilled in the art will be able to readily determine an
appropriate strategy for synthesizing peptides from various
.beta.-amino acid monomers employed in the present invention.
However, it should be noted that some classical methods of peptide
coupling (mixed anhydride, DCC) will not work with some
.beta.-amino acids. While not wishing to be bound by theory, it is
believed that the lack of a group at the .alpha.-carbon causes
increased side reactions of the activated carboxyl group. Another
possible side reaction which should be avoided is formation of a
dihydrooxazinone, which has been observed in some cases and which
may participate in the coupling reaction.
[0591] Successful synthesis strategies for some homopolymers (e.g.,
poly (.beta.-amino butyrate)) include polymerization of
.beta.-lactams. See e.g. Chen et al., Macromolecules (1974) 7:779
which is incorporated herein by reference for all purposes. Similar
methods may be employed in some instances with methods of the
present invention. The homopolymers so produced have been found to
adopt a .beta.-conformation analogous to the poly(.beta.-hydroxy
butyrate) polymers produced by bacteria. The .alpha.-esters of
aspartic acid also have been oligomerized to form 1-peptides, and
they form .beta.-sheets when containing eight or more units.
Directed synthesis in solution of a tripeptide formed from
.beta.-amino butyrate has been accomplished using trichlorophenyl
active esters in the presence of hexamethylphosphoramide (HMPA) as
described in Drey et al., J. Chem. Soc., Perkin Trans. (1982)
1:1587-1592 which is incorporated herein by reference for all
purposes. Other syntheses well-known in the art may be employed in
the present invention to produce a variety of peptide
oligomers.
[0592] Three dihedral angles (shown below) may be controlled in
.beta.-peptides. 32
[0593] The angle depicted as .psi. above has no analogy in a
.alpha.-peptides and is an angle over which there is appreciable
control via the .alpha. and .beta. substituents. Considering the
erythro and threo isomers of
.alpha.,.beta.-dimethyl-.beta.-alanine, a prototype disubstituted
.beta.-peptide, empirical force field calculations suggest that, in
both cases, anti orientation of the methyl groups is favored. This
results in the threo isomer introducing a turn in the chain, while
the anti isomer would tend to maintain an extended backbone.
Certainly, other conformations are close in energy, and will also
be populated. However, 33
[0594] alkyl groups may be linked to a ring to control the rotation
of the .psi. bond. This allows variation in the orientation of the
main chain in a systematic way. A family of cyclic .beta.-amino
acids has been designed for this purpose. The size of the ring onto
which the .beta.-amino acid unit is fused and the fusion geometry
limit the possibilities for the dihedral angle between the carboxy
carbon and the amino group (angle .psi.). For example, the
cis-cyclopropyl compound is constrained to eclipse these bonds
(.psi.=0.degree.), while the trans-cyclopropyl locks them at a
144.degree. angle. Energy minimization of other members of this
homologous series showed an orderly progression of dihedral angles
for several low energy conformers.
[0595] The generic structure for the cyclic .beta.-amino acid
monomers is shown below. 34
[0596] where X is carbon, silicon or the like, and Y is one or more
carbon, nitrogen, oxygen, sulfur, silicon or no atom.
[0597] Depending on the enantiomeric series to be employed, the
turn or extension introduced into the backbone can be either of
positive or negative helicity. 35
[0598] Some of the compounds (e.g., the diaxial cyclohexane system)
do not have as a global minimum in the conformation shown, and
further constraints may be utilized to enforce the desired
stereochemical relationship. These constraints will be well-known
in the art and include, by way of example, polar effects, other
rings, or allylic strain. 36
[0599] One advantage of this homologous series of compounds is that
a unified synthetic approach can be employed. For example, the
cycloaddition of chlorosulfonylisocyanate with cycloalkenes may be
utilized. This is most applicable to cis isomers of cyclopentyl and
cyclohexyl systems. Enantiomer selective enzymatic hydrolysis of
the .beta.-lactam then gives the desired .beta.-amino acids ready
for derivatization. A second route is more general, and longer. It
begins with 1,2-cycloalkanedicarboxylic esters (or their derived
diol acetates), which can be prepared via Diels-Alder reactions of
maleates or from dianion alkylation of succinate as decribed in
Garratt et al., Tetrahedron Lett. (1987) 28:351-352 which is
incorporated herein by reference for all purposes. Enzymatic
transformation then introduces optical activity and differentiates
the carboxyl groups, which permits selective conversion of one into
an amino group as described in Sabbioni et al., J. Org. Chem.
(1987) 52:4565-4570 which is incorporated by reference herein for
all purposes. One advantage of this stategy is that it directly
provides the amino protected building block. 37
[0600] Other compounds related to the above .beta.-amino acid
analogues may be employed in peptide syntheses according to the
present invention. For example, .alpha.-aminoxy acids in which the
.beta.-carbon of a .beta.-amino acid is replaced with an oxygen may
be used. These compounds are easily prepared in optically active
form from the amino acids via the bromo acid. Several peptide
analogues incorporating this unit have been synthesized in solution
and are known to be resistant to mammalian proteases. See e.g.
Briggs et al., J. Chem. Soc. Perkin Trans. (1979) 1:2138-2143 which
is incorporated by reference herein for all purposes.
[0601] F. Reduced Amide Bonds
[0602] Reduced amide peptide isosteres have been incorporated into
peptides by reductive alkylation to produce antagonists, enzyme
inhibitors, and resistance to biodegradation. For example, in the
case of a protease, one can render the scissile bond non-cleavable
by introduction of the reduced amide at the cleavage site. In the
case of Renin, a reduced peptide analog of the native Renin
substrate (H-142) has been shown to reduce blood pressure in
clinical tests.
[0603] According to some embodiments, the present invention
provides for the introduction of a reduced amide bond into a
growing peptide chain or other polymer chain on a substrate in
situ, without the preformation of a dipeptide containing a reduced
amide bond. FIG. 41 illustrates the introduction of a reduced amide
bond into a growing peptide or other polymer. According to a first
step of the process, a substrate is formed having an optional chain
of amino acids (indicated by AA.sub.n) and an exposed amino
terminus. The substrate is reacted with an aldehyde derivative of a
peptide in the presence of, for example, HCL or AcOH in DMF,
providing a substrate with the chain of amino acids and a terminal
imine group. The substrate is the reacted with, for example,
NaBH.sub.4 or NaCNBH.sub.3 forming a sequence with a reduced amide
bond in the growing polymer chain. The substrate is then optionally
processed according to the methods described above to provide
additional amino acids on the growing chain.
VI. CONCLUSION
[0604] The inventions herein provide a new approach for the
simultaneous synthesis of a large number of compounds. The method
can be applied whenever one has chemical building blocks that can
be coupled in a solid-phase format, and when light can be used to
generate a reactive group.
[0605] The above description is illustrative and not restrictive.
Many variations of the invention will become apparent to those of
skill in the art upon review of this disclosure. Merely by way of
example, while the invention is illustrated primarily with regard
to peptide, oligosaccharide and nucleotide synthesis, the invention
is not so limited. By way of another example, while the detection
apparatus has been illustrated primarily herein with regard to the
detection of marked receptors, the invention will find application
in other areas. For example, the detection apparatus disclosed
herein could be used in the fields of catalysis, DNA or protein gel
scanning, and the like. The scope of the invention should,
therefore, be determined not with reference to the above
description, but instead should be determined with reference to the
appended claims along with their full scope of equivalents.
Sequence CWU 1
1
* * * * *