U.S. patent application number 14/125249 was filed with the patent office on 2014-05-15 for methods and apparatus for improving in vitro measurements using boyden chambers.
This patent application is currently assigned to ESSEN BIOSCIENCE, INC.. The applicant listed for this patent is Bradley D. Neagle, Kirk Schroeder. Invention is credited to Bradley D. Neagle, Kirk Schroeder.
Application Number | 20140134666 14/125249 |
Document ID | / |
Family ID | 47293508 |
Filed Date | 2014-05-15 |
United States Patent
Application |
20140134666 |
Kind Code |
A1 |
Schroeder; Kirk ; et
al. |
May 15, 2014 |
METHODS AND APPARATUS FOR IMPROVING IN VITRO MEASUREMENTS USING
BOYDEN CHAMBERS
Abstract
Apparatus and methods to improve the Boyden chamber used in
cellular biological measurements, allowing quantitative optical
microscopy of biological cells in situ without using fluorescent
probes or optical staining. A thin, porous membrane separating top
and bottom reservoirs includes an array of precisely positioned
micropores pores manufactured using a laser-based photo-machining
(ablation) process. The membrane may be composed of polyethylene
terephthalate (PET), polycarbonate, polyimide, polyether ether
ketone (PEEK), polystyrene, or other appropriate material. The
pores formed in the membrane may have diameters in the range of 1
to 15 microns and spaced apart at a distance ranging from 10 to 500
microns. A plurality of upper and lower reservoirs may be provided
to form a multi-well plate. Potential biological applications where
Boyden chamber geometries are currently used include co-culture
studies, tissue remodeling studies, cell polarity determinations,
endocrine signaling, cell transport, cell permeability, cell
invasion and chemotaxis assays.
Inventors: |
Schroeder; Kirk; (Ann Arbor,
MI) ; Neagle; Bradley D.; (Ann Arbor, MI) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Schroeder; Kirk
Neagle; Bradley D. |
Ann Arbor
Ann Arbor |
MI
MI |
US
US |
|
|
Assignee: |
ESSEN BIOSCIENCE, INC.
Ann Arbor
MI
|
Family ID: |
47293508 |
Appl. No.: |
14/125249 |
Filed: |
June 8, 2012 |
PCT Filed: |
June 8, 2012 |
PCT NO: |
PCT/US2012/041652 |
371 Date: |
December 10, 2013 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
13157873 |
Jun 10, 2011 |
8673628 |
|
|
14125249 |
|
|
|
|
Current U.S.
Class: |
435/29 ;
435/288.7 |
Current CPC
Class: |
B01L 2200/0647 20130101;
G01N 21/6452 20130101; B01L 2300/163 20130101; G01N 21/6458
20130101; G01N 33/5008 20130101; B01L 3/50255 20130101; B01L
2300/0654 20130101; G01N 33/5029 20130101; G01N 33/58 20130101;
C12Q 1/02 20130101; B01L 2300/0681 20130101; B01L 3/5085 20130101;
B01L 2300/0829 20130101; B01L 2200/142 20130101; G01N 33/54366
20130101 |
Class at
Publication: |
435/29 ;
435/288.7 |
International
Class: |
C12Q 1/02 20060101
C12Q001/02 |
Claims
1. Biological measurement apparatus, comprising: a bottom
reservoir; a top reservoir; a thin porous membrane separating the
top and bottom reservoirs; and wherein: the pores of the membrane
are formed using a laser-based photo-machining (ablation) process;
and the porous membrane enables quantitative optical imaging of
biological cells on the membrane without the use of fluorescent
probes or optical staining.
2. The apparatus of claim 1, wherein the membrane is composed of
polyethylene terephthalate (PET), biaxially-oriented polyethylene
terephthalate (boPET), polycarbonate, polyimide, polyether ether
ketone (PEEK), or polystyrene.
3. The apparatus of claim 1, where the porous membrane is in the
range of 10 to 125 microns thick.
4. The apparatus of claim 1, where the pores of the member are
arranged in a predetermined rectangular or other geometric
array.
5. The apparatus of claim 1, where the pores of the member have a
uniform and consistent spacing, density and diameter.
6. The apparatus of claim 1, wherein the pores of the membrane have
diameters in range of 1 to 15 microns.
7. The apparatus of claim 1, wherein the pores of the membrane are
spaced apart at a distance ranging from 10 to 500 microns.
8. The apparatus of claim 1, further including a plurality of upper
and lower reservoirs forming a multi-well plate.
9. The apparatus of claim 1, wherein the membrane is coated with
collagen 1, fibronectin, laminin or other extracellular matrix.
10. The apparatus of claim 1, wherein the reservoirs are
manufactured with an injection molded plastic.
11. The apparatus of claim 1, wherein the reservoirs are
manufactured with injection molded polystyrene, polycarbonate,
polyethylene terephthalate (PET) or biaxially-oriented polyethylene
terephthalate (boPET).
12. The apparatus of claim 1, wherein the membrane is attached to
either the top or bottom reservoir using an ultrasonic welding
process or chemical bonding agent.
13. The apparatus of claim 1, wherein the membrane is attached to
either the top or bottom reservoir using laser welding, or laser
mask welding.
14. The apparatus of claim 1, wherein the size of the reservoirs,
the size of the pores, the number of the pores, and the location of
the pores are optimized to reduce the number of biological cells
needed for a given assay precision.
15. The apparatus of claim 1, wherein the membrane is attached to
the bottom surface of the top reservoir thereby forming an insert
that fits inside the bottom reservoir.
16. The apparatus of claim 1, where a plurality of upper reservoirs
are attached to a porous membrane, thereby forming a removable
insert that fits inside a plurality of co-aligned bottom reservoirs
forming a microplate.
17. The apparatus of claim 1, wherein the porous membrane is
substantially smooth and optically transparent to enable
quantitative phase-contrast imaging of the biological cells on the
membrane without the use of fluorescent probes or optical
staining.
18. The apparatus of claim 1, further including a phase contrast
imaging system for performing the quantitative optical imaging of
biological cells on the membrane without the use of fluorescent
probes or optical staining.
19. The apparatus of claim 18, wherein the phase contrast imaging
system utilizes one of Zernike phase contrast, differential
interference contrast (DIC) or Hoffman modulation contrast.
20. A biological measurement method, comprising the steps of:
providing a substantially smooth, optically transparent, thin film
membrane having an upper surface and a lower surface; forming a
plurality of micropores through the membrane using a laser-based
photo-machining (ablation) process; using the membrane to separate
upper and lower fluid-containing reservoirs; and performing
quantitative optical imaging of biological cells on the upper
surface of the membrane without the use of fluorescent probes or
optical stains.
21. The method of claim 20, including the step of using a
phase-contrast technique to perform the quantitative optical
imaging.
22. The method of claim 20, including the step of using Zernike
phase contrast, differential interference contrast (DIG), or
Hoffman modulation contrast to perform the quantitative optical
imaging.
23. The method of claim 20, further including the use of
epifluorescence microscopy.
24. The method of claim 20, wherein the quantitative optical
imaging is used for the measurement of cell migration (chemotaxis),
cell invasion, cell permeability, tissue remodeling, cell polarity
endocrine signaling or cell transport.
25. The method of claim 20, wherein the step of quantitative
imaging involves a morphological assessment of shape, and/or the
counting of the cells on the surface of the membrane and/or the
identification of a particular cell type within a mixed cell
population.
26. The method of claim 20, including the step of counting the
number of cells remaining on the upper surface of the membrane over
time to quantify cell chemotaxis or cell invasion.
27. The method of claim 26, including the use of kinetic,
multi-time-point quantitative optical microscopic measurements to
reduce artifacts associated with transient chemical gradients in
chemotaxis or chemo-invasion assays.
28. The method of claim 20, wherein the step of forming a plurality
of micropores in the membrane includes forming pores with diameters
in range of 1 to 15 microns and spaced apart at a distance ranging
from 10 to 500 microns.
29. The method of claim 20, including the step of providing a
polyethylene terephthalate (PET), biaxially-oriented polyethylene
terephthalate (boPET), polycarbonate, polyimide, polyether ether
ketone (PEEK) or polystyrene thin film membrane.
30. The method of claim 20, including the step of fabricating a
plurality of porous membranes, each separating a respective upper
and lower reservoir, thereby forming a multi-well plate.
31. The method of claim 20, including the step of coating the
membrane with collagen 1, fibronectin, laminin or other
extracellular matrix.
32. The method of claim 20, including the step of forming the upper
and lower reservoirs using injection-molded polystyrene,
polycarbonate, polyethylene terephthalate (PET) or
biaxially-oriented polyethylene terephthalate (boPET).
33. The method of claim 20, including the step of ultrasonically
welding or chemically bonding the porous membrane to the upper or
lower reservoir.
34. The method of claim 20, including the step of laser welding or
laser mask welding the porous membrane to the upper or lower
reservoir.
35. The method of claim 20, including the step of attaching the
membrane to the bottom surface of the top reservoir, thereby
forming a removable insert that fits inside the bottom
reservoir.
36. The method of claim 20, including the step of optimizing the
pore diameter, pore locations and timing of data acquisition in
order to reduce artifacts associated with a transient diffusion
gradient in chemotaxis or chemo-invasion assays.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This is a U.S. national stage patent application of
PCT/US2012/041652, filed Jun. 8, 2012, which claims priority of
U.S. patent application Ser. No. 13/157,873 filed Jun. 10, 2011,
the content of both of which is incorporated herein by
reference.
FIELD OF THE INVENTION
[0002] This invention relates generally to cellular biological
measurement including cell migration (chemotaxis), cell invasion,
cell permeability, tissue remodeling, cell polarity endocrine
signaling and cell transport and, in particular, to apparatus and
methods for improving the Boyden chamber apparatus used in such
measurements.
BACKGROUND OF THE INVENTION
[0003] Cell migration is critical in many physiological processes.
Chemotaxis, for example, is the study of cell motion in response to
a soluble chemo-attractant stimulus. Similar and related mechanisms
include haptotaxis and chemoinvasion, which rely on cell motility
on a substrate-bound stimulus and movement through an extracellular
matrix (ECM) boundary layer, respectively. These processes play a
vital role in the basic physiology of many diseases in a large
number of therapeutic areas and the study of cell migration is a
widely adopted research tool for both in vitro and in vivo
biological research.
[0004] The most common technique for measuring cell migration in
vitro is via a measurement geometry known as the Boyden chamber [1]
first described in 1962. This geometry, developed by Dr. Stephen
Boyden, consists of two chambers separated by a porous membrane.
FIG. 1 illustrates a prior-art Boyden Chamber single-well geometry
for measuring cell migration utilizing a porous track-etch membrane
filter 104. Migrating cells respond to a chemical gradient that
forms by diffusion of the chemoattractant from the lower chamber
106 to the upper chamber 102 via the porous membrane 104. Cells
respond to the chemical gradient by directional cues, migrating
through the holes of the porous membrane. Cells which migrate
through the holes of the membrane can either adhere to the lower
side of the membrane, or fall through the membrane to a lower
reservoir for detection.
[0005] For biological applications, the membrane pores are
typically in the range of 3 to 8 uM in diameter, slightly smaller
than the diameter of most cells, but large enough that cells can
squeeze through. When using a Boyden chamber as a chemotaxis
measuring device, a chemoattractant is typically added to the lower
chamber, cells are added to the upper chamber and the porous
membrane serves as a means to establish a diffusion-based,
time-dependent chemical gradient between the upper and lower
chambers [2]. The cells on the top side of the membrane detect the
chemical gradient, migrate to the individual pores in the membrane,
and then crawl through the holes to the lower chamber. After
migrating through the small pores, the cells ultimately either fall
through the membrane to a lower reservoir, or remain attached on
the bottom side of the membrane.
[0006] As an in vitro assay for chemotactic cell migration, test
compounds (drugs) are either added to the upper chamber at the
start of the experiment, or the cells are pre-incubated with the
compound prior to loading the Boyden chamber device.
Pharmacological modulation is then measured by comparing the
migration response in cells exposed to test compound vs. cells with
no test compound. Quite often in using these devices there is also
a correction made for cell migration which may occur in the
measurement chambers which contain no chemoattractant in the lower
reservoir. This "random migration" component can then be subtracted
off the response in order to deduce the number of cells responding
to the chemotactic gradient alone.
[0007] Determination of the chemotactic response relies on
quantifying the number of cells that have migrated through the
porous membrane. Historically, this is done by fluorescently
labeling the cells either prior to adding to the Boyden chamber
(pre-labeling), or after the cells have migrated (post labeling).
The fluorescence detection is then performed using an imaging
system (fluorescence microscopy) and counting the individual
fluorescently labeled cells which have migrated, or by making a
bulk fluorescence measurement with a fluorescent plate reader and
subsequently calibrating this bulk fluorescent signal to total cell
number.
[0008] There have been many improvements to the standard Boyden
chamber geometry since its inception, but the basic device geometry
and membrane components have remained fairly consistent. Goodwin,
in a series of U.S. Pat. Nos. 5,210,021; 5,284,753 and 5,302,515
describes improvements wherein a device is constructed of multiple
sites, often using hydrophobic coatings to make the individual well
compartments. While extending the art at the time of invention,
these devices all rely on labeling the cells with a fluorescent dye
for detection. Another disadvantage of these inventions is the
requirement to scrape away cells on the top side of the membrane in
order to differentiate them from those which have migrated to the
bottom side of the membrane.
[0009] In U.S. Pat. No. 5,601,997, Tchao describes a device to help
eliminate this problem by using an organic absorption dye in the
porous membrane to block fluorescent excitation light from getting
to the top side of the membrane. In this way, fluorescent
pre-labeled cells which had migrated to the bottom side of the
membrane, or migrated through the membrane and fallen into a lower
reservoir could be counted using a fluorescent reader without
fluorescent signal interference from the cells remaining on the top
side of the membrane. This invention extended the state-of-the art
by not requiring the tedious and error-prone cell "scraping" steps.
It also enabled the possibility of performing kinetic, time-lapse
experiments using fluorescent pre-labeled cells.
[0010] Currently, there are many commercial sources for 6-well,
24-well or 96-well Boyden chamber-derived chemotaxis kits,
including the ChemoTx.TM. system sold by Neuroprobe Inc.
(Gaithersburg, Md.), the Transwell.TM. system sold by Corning Life
Science (Acton, Mass.) and the HTS Fluoroblo.sup.k.TM. system sold
by Becton-Dickinson (Franklin Lakes, N.J.). All of these devices
are basically rectangular arrays of Boyden chambers using standard
microplate formats and injection mold fabrication techniques for
creating upper and lower reservoirs separated by a porous filter
membrane. All of these commercial devices also use the same basic
material for the porous membrane which is known as a "track-etch"
membrane.
[0011] Track-etch membranes are manufactured by exposing 10 to 20
micron thick, polymer films (e.g. polyester, or polycarbonate) to
radioactive particle bombardment, followed by chemical etching [3].
The results of this manufacturing process are a porous film with a
random pattern of precisely sized micro-holes as shown in the
high-resolution microscopic brightfield image of FIG. 2. The
cumulative density of micro-holes using this fabrication technique
is controlled by the exposure time to, as well as the physical
geometry between, the membrane and the radioactive source. The size
of the micro-holes in a given track-etch membrane is governed by a
combination of time, temperature and the chemical concentration
used during the etch step. Typical etch solutions include highly
concentrated NaOH, or HF. For a given process, the micro-hole size
is very uniform, and in general the holes are orthogonal to the
surface of the membrane. Pore size typically ranges from 0.2
microns, up to 10 microns in diameter. As shown in FIG. 2, the
pores are in random locations with some certain local areas on the
membrane having much higher pore densities than other areas on the
membrane. The random nature of the radioactive bombardment does not
allow for precise control of the local pore spacing or density.
[0012] The predominant application for track-etched membranes is
fine particle and contaminant filtering of fluids as well a method
of capturing and detecting microorganisms. The filtering
applications take advantage of the very uniform and defined pore
size of the membrane. This characteristic makes these types of
membranes ideal for precisely filtering particles or
micro-organisms of a given size. For biological applications such
as cell migration, track-etch membranes of 2, 3, 5 and 8 micron
diameter pore sizes are most commonly used with pore densities
ranging from 1000 pores per square millimeter to 20,000 pores per
square millimeter.
[0013] For in vitro chemotactic cell migration applications, pore
size is often matched to size of the cells being studied, bigger
cells use bigger pore sizes. Typically, the pore size is chosen to
be slightly larger than the nucleus of the cells under
investigation. Most of the commercial manufacturers of Boyden-style
chemotaxis chambers offer a variety of products incorporating
different pore sizes. They also offer membranes in different
materials, most commonly polycarbonate and polyester, and often
supply biological substrate coatings (e.g. collagen, fibronectin or
laminin) or surface coating protocols. These biological coatings
are sometimes useful so as to more accurately mimic the in vivo
surface/adhesion biology of migrating cells and often the surface
coatings are necessary to illicit the proper surface receptor
activation (e.g. integrin signaling) necessary for the cells to
migrate efficiently [4].
[0014] The quantitative read-out from a Boyden chamber chemotaxis
assay is based quantifying the number of cells which migrate
through the individual holes of the track-etch membrane. The cells
which crawl through the pores, either adhere onto the bottom size
of the membrane, or alternatively fall into the lower reservoir. In
such test systems it is often necessary to include "control wells"
which do not contain chemoattractant in the bottom reservoir to
correct for the occurrence of random (non-directed) migration which
can occur. In existing commercial Boyden chamber technologies,
quantification of the number of cells is accomplished by using
fluorescent dye labeling of the cells.
[0015] Labeling of the cells is necessary as the cells cannot be
visualized on the surface of the track-etch membranes directly
without using a labeling dye. Today, there are a host of live-cell
dye markers which can be used, such as Calcein-AM (Sigma Aldrich,
St. Louis Mo.). Once labeled, the cells can be counted directly
using cell counting microscopy; or if a proportionality
relationship can be established between fluorescence and cell
number, a bulk fluorescent measurement can be made as a surrogate
for cell number. Cells are typically labeled at the beginning of
the experiment, i.e. before cell migration occurs. Cells can also
be "post-labeled," i.e. after the cell migration occurs. This
latter method is often preferred when working with cell types which
are adversely affected by the fluorescent dyes.
[0016] Boyden chamber technology is the current "gold standard" for
in vitro chemotaxis and chemoinvasion type assays and has been
around for almost fifty years. The modern incarnations of the
technique have the advantage of being amenable to multi-well
microplate formats and the precision of plastic injection molding
techniques; as such they are reasonably high throughput assays.
While being the current gold standard, and clearly dominating the
research market, there are several disadvantages to the current
Boyden chamber systems. These disadvantages will be discussed in
the following paragraphs.
Disadvantages of Prior Art Boyden Chamber Geometries
[0017] A. Current Boyden Chamber Devices are not Compatible with
Phase-Contrast Imaging
[0018] One limitation of the current Boyden chamber technologies is
the inability to image the cells on the surface of the membrane
in-situ, without using fluorescent labels. The reason for this has
to do with the optical quality of the track-etch membranes used in
commercial Boyden chamber products. Biological cells are
essentially optically transparent. Imaging the cells without the
use of an optical dye or stain requires a special kind of optical
imaging. The most common method to do this is using imaging
techniques which encode refractive optical phase changes introduced
by the cells into detectable intensity changes. Imaging techniques
which work on this premise include Zernike phase contrast,
differential interference contrast (DIC) and Hoffman modulation
contrast. For the purposes of this discussion we will refer to all
of these as phase contrast imaging.
[0019] Phase contrast imaging techniques rely on placing the cells
on very high optical quality, optically clear substrate, mostly
typically glass slides or thin plastic. The substrate must be
optically transparent. In addition, the substrate must also be very
flat, smooth and free of composition characteristics (i.e. air
bubbles, material inhomogeneity from molding stresses, etc.) which,
in-turn can cause refractive phase changes. Such phase changes
result in imaging artifacts which can easily occlude the subtle
phase variations introduced by the biological cells. Note, phase
contrast micrscopy is very much distinct from "brightfield"
microscopy. Brightfield imaging relies on light absorption by the
specimen to encode intensity variations which are visible to the
user or detection apparatus. Phase contrast imaging relies optical
phase changes created by the specimen, and, as such can produce an
image even if the specimen is mostly transparent as is the case
with unlabeled biological cells.
[0020] Phase contrast imaging is not amenable to the current
commercially-available Boyden chamber consumables (prior art) which
rely on porous membranes manufactured by the track-etch
manufacturing process. Membranes produced by the track-etch process
are not smooth, and introduce a variety of optical phase
perturbations when imaged with a phase contrast microscope, making
detection of the cells on the surface using these techniques
impossible. FIG. 3A, similar to FIG. 2, is a brightfield image of a
typical track-etch membrane used for biological studies. In this
case the membrane has randomly spaced 8 micron diameter pores which
are clearly visible under brightfield imaging.
[0021] FIG. 3A is a brightfield image of a track-etch membrane with
cells. Pores are visible and exhibit a random pattern under
brightfield imaging. The particular membrane shown in FIG. 3A also
had migrating cancer cells (HT1080) on the surface, although under
brightfield imaging the cells are not visible as previously
described. By comparison, FIG. 3B is an image of the same HT1080
cells on a clear plastic substrate taken at the same resolution,
showing what "unlabeled" cells look like using phase contrast
imaging on a high quality imaging substrate. Lastly, FIG. 3C is a
phase contrast image of the same track-etch membrane with HT1080
cells as shown in FIG. 3A. As demonstrated by these series of
images, the surface irregularities on the track-etch membrane
surface result in an incomprehensible phase contrast image. Indeed,
due to the optical perturbations caused by the membrane fabrication
process, cells on the surface of the membrane cannot be
identified.
[0022] The imaging artifacts introduced by these surface and
material irregularities overwhelm the variations introduced by the
cells making identification and enumeration of the cell number on
these membranes using phase contrast imaging techniques impossible.
As such, imaging of the cells on the surface of the existing
track-etch membranes necessarily requires a fluorescent or optical
dye to label (identify) the cells.
[0023] The practical impact of requiring an optical dye or
fluorescent label to effectively "count" the cells is a big
disadvantage in running these types of assays. First and foremost,
many cell types are adversely affected by the presence of
fluorescent or optical labeling dyes. It is well known that the
presence of optical dyes can change cell viability, growth, and
function. This is especially problematic when using "primary"
hematopoetic blood cells (e.g. T-lymphocytes, neutrophils).
[0024] In addition, fluorescent or optical dye labeling techniques
makes the assays more cumbersome and costly to perform. Lastly, the
use of fluorescent or optical dyes often precludes the ability to
take multiple time point recordings due to inherent phototoxicity
and the generation of free radicals during the imaging process. One
of the advantages of the techniques presented by the invention
disclosed here is the ability to make these types of measurements
on a substrate which is compatible with phase contrast imaging,
thereby eliminating the requirement of labeling cells.
B. Current Boyden Chamber Devices are Manufactured with Porous
Membranes with No Precise Control of Hole Spacing or Hole
Density
[0025] Boyden chambers rely on passive chemical diffusion in order
for a gradient to be formed between the upper and lower fluid
reservoirs. The chemical gradient formed is a complicated function
of the pore geometry (hole size and spacing), time, concentration
of the chemoattractant and the molecular weight of the
chemoattractant. The gradient is, by definition, time varying and
eventually, given enough time, the top chamber and bottom chamber
concentrations equalize and the diffusion-based chemical gradient
decays. While the geometry of track-etch membranes have been
optimized for particle filtration, pore density/spacing using these
membranes has not been optimized for measuring cell migration.
[0026] This situation is made worse by the non-homogeneous, random
pore spacing of the track etch membranes (prior art) which causes
time-dependent and local varying chemotactic gradients. In regions
with high pore density, the chemical gradient can decay much
quicker than in regions with low pore density. As the chemical
gradient decays, the cells lose the ability to find the pores.
These in turn can cause artifacts where, depending on the time of
the experiment and concentration of the chemoattractant, the cells
get confused and stop their directional migration. When this
occurs, it becomes impossible to determine if a reduction in cell
migration is due to degradation of the chemical gradient, or due to
the inhibitory effect of a pharmacological test agent.
[0027] FIG. 4 demonstrates an example of this effect using
commercially available prior art Boyden chamber devices; in
particular, the migration of human neutrophils in response to
increasing concentrations of chemotactic agent in commercially
available Boyden chamber measurements. Shown is the number of cells
which have migrated through the track-etch membrane (vertical axis)
vs. increasing concentrations of chemoattractant, IL-S in the top
chart, and compound C5a in the bottom chart of FIG. 4,
respectively. As shown, at higher concentrations of
chemoattractant, the cells undergo a diminished migratory response
(dashed circle). This effect is inseparable from a diminished
migratory response produced from an inhibitory compound and is
generally the result of a time-concentration dependent decay of the
chemotactic gradient. The practical consequence of this effect is
that absolute comparisons between wells becomes a concentration and
time dependent phenomenon requiring extreme diligence in
experimental technique, adding imprecision to the resulting
measurements and typically requiring the use of many replicate
measurement wells per test compound to achieve assay precision.
C. Number of Cells Required to Quantify the Response:
[0028] Another disadvantage of currently available Boyden chamber
derived chemotaxis kits is that they require large number of cells
to characterize the response, typically 50,000 to 100,000 cells per
well [5] [6]. Large cell numbers can be a major cost/adoption
disadvantage. This is especially the case when using rare, and
perhaps difficult to isolate primary hematopoietic cells from the
blood.
[0029] In summary, the Boyden chamber geometry remains the
gold-standard measurement technique for measuring in-vitro
chemotaxis. However, the commercial solutions (prior art) for the
Boyden chamber geometry all suffer from the following
disadvantages: [0030] 1.) Current Boyden chamber products
manufactured with track-etch membranes are not amenable to phase
contrast imaging, and therefore require that the cells be labeled
for detection [0031] 2.) Fluorescent or optical dye staining of the
cells can be detrimental to the physiology of the cells [0032] 3.)
Current Boyden chamber products requires many cells per measurement
well (typically 50,000 to 100,000) [0033] 4.) Current Boyden
chamber products incorporating track-etch membranes have very
non-uniform, irregular hole patterns which have not been optimized
for this process. This results in non-homogeneous local chemical
gradients at the cell surface, and local hole densities which are
either too high (often) or too low. Combined, these characteristics
tend to enhance the time-dependence of the gradient process, often
making it difficult to distinguish a real pharmacological
inhibition of migration from a time dependent decay in
gradient.
SUMMARY OF THE INVENTION
[0034] The invention described here relates to an apparatus and
methods for improving the Boyden chamber used in cellular
biological measurements. In accordance with the invention, cells
can be directly imaged and analyzed in situ, using phase contrast
imaging techniques and without using fluorescent labels or optical
dye staining.
[0035] Apparatus constructed in accordance with the invention
includes a bottom reservoir, a top reservoir, and a thin porous
membrane separating the top and bottom reservoirs. In the preferred
embodiment, the pores of the membrane are manufactured using a
laser-based photo-machining (ablation) process. The hole diameter,
spacing and density of the pores in the porous membrane can be
precisely controlled, enhancing the applicability of these devices
for chemotaxis and chemo-invasion type assays. For example, the
pores formed in the membrane may have diameters in the range of 1
to 30 microns and spaced apart at a distance ranging from 5 to 500
microns. The pores may be arranged in a predetermined array having
a rectangular or other geometry. The combined attributes of this
invention result in greater assay precision using fewer cells.
[0036] The membrane may be composed of polyethylene terephthalate
(PET), biaxially-oriented polyethylene terephthalate (boPET),
polycarbonate, polyimide, polyether ether ketone (PEEK),
polystyrene or other appropriate material with optical
characteristics which do not introduce significant phase
perturbation to an incoming light wavefront in relation to those
introduced by biological cells on the surface of the membrane. In
particular, the membrane may be substantially smooth and optically
transparent even following pore formation. A plurality of upper and
lower reservoirs may be provided to form a multi-well plate.
[0037] The invention may be used for the measurement of cell
migration (chemotaxis), cell invasion, cell permeability, tissue
remodeling, cell polarity endocrine signaling or cell transport. In
a typical application, the porous membrane is used to separate
upper and lower fluid-containing reservoirs and coated with
collagen 1, fibronectin, laminin or other extracellular matrix
material. An inverted phase contrast, DIC, or Hoffman
Modulation-type microscope may be used to assess the morphology
and/or number of biological cells on the membrane. An advantage of
the invention is that chemotaxis, cell migration, cell invasion and
other processes may be carried out by counting the cells on the
surface of the porous membrane directly and without using
fluorescent labels or optical dye staining.
[0038] The improvements made possible by the invention impact a
range of potential biological applications where Boyden chamber
geometries are currently used including co-culture studies, tissue
remodeling studies, cell polarity determinations, endocrine
signaling, cell transport, cell permeability, invasion and
chemotaxis assays. While the description presented here is intended
to describe the advantages of the invention specifically as it
relates to the measurement of in vitro chemotaxis and in vitro cell
invasion assays, many of the attributes associated with the
invention are extendable to other common uses of Boyden chamber
systems.
BRIEF DESCRIPTION OF THE DRAWINGS
[0039] FIG. 1 depicts a prior art single well of an existing Boyden
chemotaxis chamber geometry;
[0040] FIG. 2 shows a prior-art bright field image of a track-etch
membrane surface;
[0041] FIGS. 3A-3C illustrate prior art brightfield images of a
track-etch membrane with cells, a phase contrast image of cells on
high quality plastic, and a phase contrast image of a track-etch
membrane with cells;
[0042] FIG. 4 illustrates a prior art artifact associated with a
time, concentration dependent gradient degradation found in
existing Boyden chamber chemotaxis assays;
[0043] FIG. 5 illustrates a preferred embodiment of a Boyden
chamber using a photo-machined membrane of the present
invention;
[0044] FIG. 6 shows a phase contrast image of a photomachined
membrane demonstrating precise pore size and pore location. Also
shown are HT1080 cells on the surface of the membrane;
[0045] FIGS. 7A-7D show a time lapse phase contrast images
demonstrating chemotactic directed migration of HT1080 cells on a
photomachined membrane;
[0046] FIG. 8 shows an embodiment of the present invention, a
rectangular array of micro-holes forming the porous membrane of a
Boyden chamber geometry;
[0047] FIGS. 9A-9B depict a preferred embodiment of the invention
including a 96-well consumable comprised of 96 individual Boyden
chambers using a laser photo-machined membrane; and
[0048] FIG. 10 shows how an automated phase contrast imaging system
can be used to image the cells on the surface of the membrane.
DETAILED DESCRIPTION OF THE INVENTION
[0049] This invention is broadly directed to a process of
manufacturing the porous membrane used in Boyden chamber design.
Traditional track-etch membranes used in the production of
commercially available Boyden chamber devices use thin polymer
films such as polycarbonate or polyethylene terephthalate (PET).
The process of making pores in track-etch membranes involves
bombarding the surface with radiation, typically alpha particles
followed by a chemical etching step using highly concentrated NaOH
or HF acid. The pore density (number of pores and density) is
grossly controlled by the physical proximity and exposure levels to
the radiation source. The pore size is determined by the
concentration and exposure time in the etch step. The spatial
location of the resulting etched pores is random as determined by
the random particle generation of the radioactive source. The
result of this process is a material that has significant optical
surface blemishes. These blemishes prohibit the use of common
non-labeled, non-invasive imaging techniques such as phase
contrast, or differential interference contrast (DIC) to view cells
in situ on the surface of the membranes.
[0050] The preferred embodiment of this invention resides in a
porous membrane manufactured by a different manufacturing process,
one based on laser-based photo-machining. Laser-based
photo-machining generally uses a pulsed laser of the proper
wavelength and pulse energy such that the optical energy is
absorbed by the target material (in this case a thin polymer film)
resulting is small pieces being ablated from the surface with each
pulse. The laser beam can be precisely positioned in X-Y lateral
dimensions and the beam dimension, optical power and pulse
repetition frequency can be precisely controlled so as to allow for
very fine micro-hole machining.
[0051] An objective of the invention is to provide a regular-spaced
grid of micro-holes in a thin film whereby the pore density and
pore size are carefully controlled and the resulting porous
material has much improved optical surface characteristics in
comparison to track-etch membranes (prior art). It should also be
noted that there are many techniques for making such photomachined
holes, including a range of laser excitation wavelengths, optical
scanning systems and one, or two-dimensional laser mask approaches.
All such techniques would be well known to those skilled in the art
of laser photomachining processes and the specific implementation
should not limit the invention presented here.
[0052] FIG. 5 illustrates a specific embodiment of the invention,
that being a Boyden Chamber single-well geometry having an upper
chamber 502, a lower chamber 506, and a laser-photo machined
membrane 504. In this preferred embodiment the substantially
smooth, optically clear, laser photomachined porous membrane 504
replaces the track-etch membrane used in current commercial
devices. The membrane in this case is of a thin transparent
material which is sufficiently smooth to facilitate phase-contrast
imaging techniques.
[0053] It is envisaged that much like current track-etch membranes,
a variety of micro-hole sizes and holes spacings could be utilized
to optimize measurement geometry for a particular cell type and
assay. One of the advantages of this approach is being able to
control the precise location and density of the photo-machined
micro-holes. This is in contrast to the random location of the
microholes as provided with prior-art track-etch membranes.
[0054] FIG. 6 is a phase-contrast image of a photomachined membrane
used in a Boyden chamber device constructed in our laboratory. In
this example, the laser machined holes (9 total) were each drilled
8 microns in diameter and 180 microns apart in a regular 3.times.3
pattern. Also shown are HTI080 cells on the surface of the
membrane. As shown by this example the use of a photomachined
membrane allows for a.) imaging of the cells on the surface of the
membrane using phase contrast imaging i.e. no labels and b.)
precise hole size and hole locations to be constructed. FIG. 6 very
clearly shows the ability to detect the unlabeled biological cells
on the surface of the membrane using phase contrast imaging. This
image should be viewed in comparison with that of FIG. 3C
(prior-art) where the cells were not discernble using phase
contrast imaging techniques on traditional track-etch
membranes.
[0055] FIGS. 7A-7D depict a time lapse history of demonstrated cell
chemotaxis of HT1080 cells on a photomachined membrane reduced to
practice in our laboratory. In this example, the HT1080 cells were
serum starved for 24 hours, and then loaded on the top side of the
membrane (upper reservoir of the Boyden chamber) in serum free
media. After the cells settled onto the membrane, the bottom
chamber was then loaded with media and 10% serum. Once the device
is fully primed, the serum diffuses (from high concentration to
low) up through the individual microholes to the upper reservoir,
thereby creating a local chemical gradient around each pore for the
cells to follow. Once the cells reach the pore, they typically
crawl through the hole, either to the bottom side of the membrane,
or falling to the lower reservoir.
[0056] Using photo-machining to make the individual micro-holes
provides, for the first time, the ability to control the spacing of
the microholes. This helps alleviate a several of the disadvantages
of the current Boyden chamber systems previously described. First,
by controlling the microhole spacing one can insure that the
chemical gradient is spatially uniform and consistent around the
microholes. This is in comparison to the non-uniform local
gradients resulting from random microhole patterns in the prior-art
track-etch membranes.
[0057] Second, the holes can be precisely and regularly spaced such
that all of the cells in the local vicinity of the pores will
migrate (under optimum chemotactic conditions) to the holes. This,
effectively, optimizes the measurement geometry by minimizing the
number of holes required for a given cell density. Third, using a
uniform pore spacing and minimizing the number of holes required
extends and optimizes the time before the gradient starts to decay.
This helps to eliminate the "roll-over" effect shown in FIG. 4 by
allowing all cells in the vicinity of the microholes to migrate
before the gradient effectively decays.
[0058] FIG. 8 is a top view of a single well 802 depicting a
preferred embodiment using a 16.times.16 array of photomachined
microholes 804, 8 uM in diameter and 180 microns apart. The
specific dimensions shown here are only examples, demonstrating the
fact that, unlike the track-etch membranes, the pore spacing can be
carefully controlled. Using this approach, one could design and
implement many defined variations of micro-hole size and spatial
density depending on the biological system under study. Variables
would include optimizing the pore size for a given cell type, as
well as the pore spacing for given experimental paradigm.
[0059] The use of laser photo-machining (ablation) to fabricate the
porous membrane has additional advantages for biological
applications. First, the photo-machining process is not detrimental
to the optical quality of the thin film polymer material. Unlike
the track-etch membrane process which requires a chemical etch of
the entire surface to form the pores, the laser machining process
has minimal detrimental impact on the optical quality of the film
in the regions between the micro-holes as shown in FIGS. 6 and 7.
Maintaining a high optical quality substrate enables quantitative,
high contrast morphological analysis and/or cell counting of
individual living cells on the membrane without using fluorescent
labels or optical dyes. This is a big advantage over current Boyden
chamber methodology which requires fluorescent labeling to image
and count the cells due to the poor optical characteristics of the
track-etch membranes.
[0060] A non-labeled approach is also more amenable to a kinetic,
multiple time point read-out. Pre-labeling the cells before the
experiment can lead to phototoxicity effects during the experiment.
To avoid this, researchers often rely on "post labeling" the cells
after the experiment has been performed. This latter approach,
however, is by definition a single, end-point determination.
[0061] Having the ability to directly view the cells during the
chemotaxis process allows researchers to study morphological
changes, and or associate the response to other imaging parameters
for example the "shape change" associated with a migrating or
invasive morphological phenotype. One can not underestimate the
value of being able to image the chemotactic process in real time
when validating and/or interpreting assay data. It should also be
understood that while the present invention does not require the
use of fluorescent labels for detection, it does not preclude their
use either. Aside from the improved compatibility with phase
contrast imaging, better optical quality of the membrane will also
improve the image quality when fluorescence detection is desired,
such as may be necessary for analyzing mixed cell populations.
[0062] It should be noted, that the single-well geometry described
in FIGS. 5 (preferred embodiment) can easily be extended to a
rectangular array comprised of a plurality of wells and common
formats used the biological sciences for example 6-well, 24-well,
96-well, 384-well and 1536-well geometries. An example of a 96-well
format is depicted in FIGS. 9A (side view) and 9B (top view). Each
of the individual wells would have its only two-dimensional grid of
laser machined micro-pores. The commercial solutions previously
described are generally found in 6-well, 24-well and 96-well
formats.
[0063] One of the anticipated benefits of this invention is to be
able utilize the optimized geometry and enhanced precision in or to
reduce the number of cells required per measurement chamber. We
estimate being able to obtain improved data precision to existing
Boyden chamber products using 1,000 to 5,000 cells per well, as
opposed to the 50,000 to 100,000 cells required for existing
commercially available Boyden chamber products. FIG. 10 illustrates
the automated imaging of a single-well of a Boyden chamber geometry
using an optically clear, photo-machined membrane 1004. A light
source supported above the membrane is shown at 1002, microscope
objective disposed below the membrane is shown at 1006, and a
detector (i.e., CCD camera) is depicted at 1008.
[0064] Precise photo-machining of the thin file polymer membrane
1004 leaves an optically smooth surface free from aberrations and
enables the use of phase contrast, or other non-labeled, imaging
techniques for enumerating the cells on the membrane in situ using
manual or automated microscopy. This simple example of an automated
phase contrast imaging geometry could be used to analyze all of the
wells of a microplate-based consumable. Such systems are
commercially available (e.g. Essen Biosciences' IncuCyte) and could
be used to automatically focus on the membrane and quantify cell
migration parameters in real time.
[0065] Although the invention described in this document is
directed to a test chamber for measuring the migration of cells to
chemical stimuli, e.g., chemotaxis or chemokinesis, the invention
has applications beyond cell migration for example in the
measurement of cell permeability, cell transport, cell invasion
(others) where direct imaging of the surface would allow for
non-invasive, quantitative assessment of the cells on the membrane
in situ.
ALTERNATIVE EMBODIMENTS
[0066] Different types of polymers and polymer thicknesses,
amenable to the laser ablation photo-machining may be used in
accordance with the invention. Variations in pore size, pore
density and pore location are also anticipated. It is anticipated
that pore geometry and spacing will be used to enhance the gradient
homogeneity at the top surface of the membrane for various
experimental paradigms. Various imaging systems could be used to
image the membrane in situ, including epifluorescence (when
fluorescent labeling is desired for other reasons), Zernike phase
contrast, differential interference contrast (DIC), Hoffman
modulation contrast and others. Different data processing schemes
could also be utilized including measuring just the cells on the
top side of the membrane, or alternatively measuring cells at three
different planes being the a) top side of the membrane, b) bottom
side of the membrane and at the bottom of the collection reservoir
(lower chamber).
[0067] In order to reduce cell usage, many different potential
reservoir configurations are possible, including moving to smaller
reservoir formats, or smaller microplate formats such as a
half-area 96-well format, 384-well format, or 1536-well format.
Another alternative embodiment would be to apply biological
coatings to the surface of the membrane such as collagen 1,
fibronectin, or laminin. This type of coating would be a very thin
molecular surface coating so as not to plug the microholes. It is
also possible to apply a thicker extracellular matrix (ECM) coating
to the membrane, where the Boyden chamber can be used to measure
the ability of cells to invade the ECM, an assay known as a cell
invasion assay. This is a natural extension and common use of
existing Boyden chamber consumables.
[0068] In summary, the invention described here involves replacing
the traditional track-etch membrane used in current Boyden chamber
devices with a porous membrane manufactured via laser
photomachining. There are several advantages of this invention over
existing prior-art Boyden chamber devices. While much of this
document has described the advantages for measurements of cell
migration, many of these advantages are extendable to other
applications for Boyden chambers with numerous practical
benefits:
[0069] a) Laser-based photo-machining of the membrane allows
precise control of both the pore size and pore spacing, enabling,
for the first time, the ability to use these parameters to optimize
the chemical gradient formation process when used as a chemotaxis
or chemo-invasion measurement device. Minimizing the number of
holes required, and providing the holes at defined locations
provides a more uniform, temporally stable diffusional gradient in
comparison to existing commercially available devices. It is
anticipated that Boyden chamber devices constructed using this
invention will provide more stable pharmacological data and reduce
time dependent artifacts associated with random, non-optimized hole
patterns found in existing devices (see FIG. 4).
[0070] b) Replacing the track-etch membrane of the Boyden chamber
with one manufactured by laser ablation allows, for the first time,
imaging of the cells on the surface of the membrane in situ using
non-labeled phase contrast imaging techniques. Consequently, cells
no longer have to be exposed to potentially invasive, optical or
fluorescent labeling dyes or protocols. This is very important for
primary cells which can be time-sensitive, or label sensitive. It
also save operator time and reduces reagent cost.
[0071] c) Direct imaging of the cells on the surface of the
membrane, without using fluorescent probes or optical stains
greatly enhances the extension of these assays to a homogeneous,
automated, multi-time point data collection and analysis.
Morphological changes to the cells, such as shape change can be
monitored in situ. Direct non-labeled imaging of the cells during
the migration processes can be used to help validate and interpret
data.
[0072] d) Existing commercially available Boyden chamber solutions
dictate that at least 50,000 to 100,000 cells are required per
measurement well to achieve reasonable measurement precision. Due
to the combined benefits of this invention (optimized pore spacing,
in situ surface imaging), we anticipate needing only 1,000 to 5,000
cells per well to achieve comparable data precision to existing
devices.
REFERENCES
[0073] 1.) Stephen Boyden, Ph.D., "The Chemotactic Effect of
Mixtures of Antibody and Antigen on Polymorphonuclear Leucocytes"
J. Exp. Med. 115: pp. 453-466, (1962). [0074] 2.) C. W. Frevert, V.
A. Wong, R. B. Goodman, R. Goodwin, T. R. Martin, "Rapid
fluorescence-based measurement of neutrophil migration in vitro,
Journal of Immunological Methods 213 (1998) 41-52. [0075] 3.) J. A.
Quinn, J. L. Ajnderson, W. S. Ho, and W. J. Petzny, J., Model Pores
of Molecular Dimension, the Preparation and Characterization of
Track-Etched Membranes, Biophysical Journal Volume 12, (1972)
[0076] 4.) B. Heit, P. Colarusso, P. Kubes, "Fundamentally
different roles for LFA-1, Mac-1 and alpha4-integrin in neutrophil
chemotaxis, Journal of Cell Science 118 (22), (2005) 5205-5220.
[0077] 5.) Corning Life Sciences Inc., Corning N.Y., Cell
Migration, Chemotaxis and Invasion Assay Protocol--CLS-AN-061
[0078] 6.) Suparna Sanyal, Susan Qian, Jeff Partridge and Marhsall
Kosovsky, "Optimized Chemotaxis Conditions for Primary Blood
Monocytes or THP-1 Cells using BD Falcon.TM. FluoroBlok.TM.
96-Multiwell Insert Plates, Technical Bulletin #457, BD
Biosciences, BD Biosciences-Discovery Labware, Bedford, Mass.
01730
* * * * *