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.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

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

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

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.

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.

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.

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.

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.

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.

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.

Currently, there are many commercial sources for 6-well, 24-well or 96-well Boyden chamber-derived chemotaxis kits, including the ChemoTx™ system sold by Neuroprobe Inc. (Gaithersburg, Md.), the Transwell™ system sold by Corning Life Science (Acton, Mass.) and the HTS Fluoroblo^k™ 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.

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.

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.

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].

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.

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.

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

A. Current Boyden Chamber Devices are not Compatible with Phase-Contrast Imaging

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.

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.

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.

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.

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.

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).

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

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.

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.

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:

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.

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:

• • 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
  • 2.) Fluorescent or optical dye staining of the cells can be detrimental to the physiology of the cells
  • 3.) Current Boyden chamber products requires many cells per measurement well (typically 50,000 to 100,000)
  • 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

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.

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.

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.

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.

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

FIG. 1 depicts a prior art single well of an existing Boyden chemotaxis chamber geometry;

FIG. 2 shows a prior-art bright field image of a track-etch membrane surface;

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;

FIG. 4 illustrates a prior art artifact associated with a time, concentration dependent gradient degradation found in existing Boyden chamber chemotaxis assays;

FIG. 5 illustrates a preferred embodiment of a Boyden chamber using a photo-machined membrane of the present invention;

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;

FIGS. 7A-7D show a time lapse phase contrast images demonstrating chemotactic directed migration of HT1080 cells on a photomachined membrane;

FIG. 8 shows an embodiment of the present invention, a rectangular array of micro-holes forming the porous membrane of a Boyden chamber geometry;

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

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

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.

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.

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.

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.

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.

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 3×3 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.

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.

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.

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.

FIG. 8 is a top view of a single well 802 depicting a preferred embodiment using a 16×16 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.

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.

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.

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.

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.

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.

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.

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

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).

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.

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:

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).

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.

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.

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

• 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).
• 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.
• 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)
• 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.
• 5.) Corning Life Sciences Inc., Corning N.Y., Cell Migration, Chemotaxis and Invasion Assay Protocol—CLS-AN-061
• 6.) Suparna Sanyal, Susan Qian, Jeff Partridge and Marhsall Kosovsky, “Optimized Chemotaxis Conditions for Primary Blood Monocytes or THP-1 Cells using BD Falcon™ FluoroBlok™ 96-Multiwell Insert Plates, Technical Bulletin #457, BD Biosciences, BD Biosciences-Discovery Labware, Bedford, Mass. 01730

Claims

1. Biological measurement apparatus, comprising: wherein:

a bottom reservoir;

a top reservoir;

a thin porous membrane separating the top and bottom reservoirs; and

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.