GB2425074A - Selective cellular adhesion on polymer microarrays - Google Patents

Abstract

Agarose is used to provide a coating having a cytophobic effect on a support used for screening an array of polymer samples for cell-binding properties. Polymers for cell-binding properties are screened by forming a polymer library, such as a library of polyurethanes, by parallel synthesis, at least partially characterising the library members, forming an array of the library members on a support such as a glass slide, incubating the microarray with a cell suspension and observing the microarray to determine the binding activity of library members with specific cells. The method is used to identify polymers that bind to ostoprogenitor cells, so that the polymers can be used as a substrate for contact with unselected human bone marrow mononuclear cell preparations for selective immobilisation of ostoprogenitor cells. Suitable polyurethanes are made from polyols such as poly(propylene glycol), poly(tetramethylene glycol, poly [1, 6- hexanediol/neopentyl glycol/diethylene glycol-alt-(adipic acid)] diol, poly[1, 6- hexanediol/neopentyl glycol-alt- (adipic acid)] diol, isocyanate such as 1 ,6-diisocyanohexane, 4, 4' methylenebis(phenylisocyanate), 4-methyl- 1,3, phenylene diisocyanate, 1, 4 diisocyanobenzene, 4, 4' methylenebis(cyclohexylisocyanate) and 1, 3 bis(isocyanatomethyl)cyclohexane and chain extenders such as 1, 4 butanediol, ethylene glycol, propylene glycol, ethylene diamine, 3-dimethylamino-1,2-propanediol, 2-nitro-2- methyl-1, 3 propanediol, 2,2,3, 3, 4, 4, 5, 5 -octofluoro-1,6, hexanediol, and diethyl bis(hydroxymethyl)malonate.

Description

Selective Cellular Adhesion on Polymer Microarrays

Field of the Invention

This invention is concerned with procedures for the selective adhesion of cells on polymer microarrays, which allows selective enrichment and immobilization of specific cell types.

Background of the Invention

Arrays are having a major influence in the biological arena, with the major driving force being the huge potential multiplexing ability an array can offer to the specific application under consideration. The profound impact of arrays in the biological arena cannot be overlooked, taking into consideration the tremendous multiplexing ability an array can offer to a specific application. The most common examples are DNA arrays" or chips, which are widely used for mRNA profiling, proposed for diagnostic applications and used for SNP analysis, and which potentially have a role to play in DNA sequencing. The multiplexing power of arrays has been exploited in an increasing number of arenas such as the high throughput characterization of gene function with cell-based screens developed in a microarray format. This method is potentially very powerful but only applicable to cells which adhere to the matrix and promote cell spreading.

Polymers are essential in the area of biomaterials and have been used in a myriad of applications - see Ranter, B.D., Hoffman, A.S., Schoen, F.J., Lemons, J.E., Biomaterja/s science, an introduction to materials in medicine, edn. 2 (Elsevier, San Diego, 2005). The mechanism of cell immobilization onto polymer surfaces in cell culture has been extensively studied. It is broadly accepted that the first steps in this process are the adsorption of extracellular matrix proteins onto the surface of the polymer. Cells then indirectly interact with the polymer through the adsorbed proteins which control a variety of cellular processes such as adhesion, growth and differentiation - see Greiger, B., Bershasky, A., Pankov, R., Yamada, K.M., Transmembrane extracellular matrix-cytoskeleton crosstalk. Nat. Rev. Mo!. Cell Bio. 2, 793-805 (2001). As a result of such complex and imperfectly understood interactions, it is still impossible to predict, from the chemical structures of a polymer, how such materials will perform when in contact with cells, blood or body fluids. As a consequence, the use of a high- throughput approach to allow the rapid

I

synthesis of chemically diverse polymers offers an important tool to be able to find correlations between the design and performance of such materials - see Smith, J.R., et al., Integration of combinatorial synthesis, rapid screening, and computational modelling in biomaterials development. Macromol. Rapid Commun.

25, 127-140 (2004). Traditional methods of synthesis, identification and testing of new polymers are slow and thus over recent years, the field of automated and parallel synthesis of polymers has grown enormously, but, as is usually the case in any HT process, the development of high- throughput characterization and screening methods are often the rate limiting steps. The use of polymer arrays for cellular screening was recently reported where human embryonic stem cells were successfully differentiated following attachment and growth onto a poly(acrylate) array - see Anderson, D.G., Levenberg, S., Langer, R., Nanoliter-scale synthesis of arrayed biomaterials and application to human embryonic stem cells. Nat.

Biotechnol. 22, 863-866 (2004). However, in this case the polymers were not fully characterised due to the nanoliter-scale synthetic approach.

Summary of the Invention

One object of this invention is to provide a new approach for the preparation of polymer microarrays that allows the selective enrichment and immobilization of specific cell types, and so the conversion of "suspension" cells to adherent cells, and allows a range of biological studies to be undertaken which were previously impossible, and thus an improved understanding of cell/polymer interactions.

According to the present invention there is provided a method of screening polymers for cell-binding properties comprising: forming a polymer library by parallel synthesis; at least partially characterizing the library members; forming a microarray of the library members; incubating the microarray with a cell suspension; observing the microarray to determine the binding activity of library members with specific cells.

Preferably each polymer is characterized by at least its molecular weight or Tg or Tm independently of forming test samples on the microarray.

Suitable polymers for the library are poly(urethanes), which may be prepared under parallel synthesis conditions using different polyol, polyisocyanate and chain extender components. Poly(acrylates) are also suitable candidates for formation of varied libraries by parallel synthesis using different monomers such as methylacrylate and acrylamide.

Cell binding to samples of library members in the microarray may be detected, qualitatively and quantitatively by cell staining especially with fluorescent dyes, or by other labeling techniques that are selective for the cells rather than the polymer samples. In addition cell binding and cell type may be detected using a bright field microscope.

The procedure may be used, for example, to detect the capacity for immobilisation of epithelial cells or stromal cells and bone marrow cells. In the latter case, by identifying polymers that bind to osteoprogenitor cells, the polymers can than be used to isolate osteoprogenitor cells from unselected human bone marrow mononuclear cell preparations.

Accordingly, the invention also provides a method of isolating desired cell lines from a mixture of unselected cell preparations screening polymers comprising identifying a polymer or polymers that bind to desired cell line using procedure mentioned above, and using the identified polymer or polymers as a substrate for contact with unselected cell preparations for selective immobilisation of the desired cell line.

Alternatively the method can be used in reverse - removing undesired cells from a mixed population.

Another object of this invention is to provide a coating for microarray supports that prevents non-specific cellular binding, and so offers great potential in multiple areas of microarray screening.

Accordingly the present invention provides the use of agarose as a cytophobic coating on a microarray support.

The invention also provides a microarray support for assessing cellbinding properties of samples of interest, in which the support is coated with agarose as a substrate for a microarray of the samples.

Suitably the agarose is applied to the support as an aqueous solution which is dried to form a coating. The support is typically a glass slide that is preferably provided with an aminoalkylsilane finish before the agarose coating is applied.

Detailed Description of the Invention

In this invention, polymer libraries are prepared by parallel synthesis to give purified and well characterized polymers. For example, molecular weights may be determined by GPC, and transition temperatures Tg and Tm by DSC. Contact angles may also be measured for characterisation - see Thaburet, J.F., Mizomoto, H., Bradley, M., High-throughput evaluation of the wettability of polymer libraries.

Macromol. Rapid Commun. 25, 366-370 (2004). The library members are set out in microarray format and incubated with cell suspensions to determine the binding activity of library members with specific cells. With this information, the library can be incubated with unknown cell mixtures to identify cells within the mixture or to provide profile of the mixture.

The process of the invention has been used to detect polymers capable of binding to cell lines such as Jurkat: T cell line; 721.221: B lymphoblastoid cell line; K562: erythroleukemia (myeloid) cell line; JY: EBV transformed B cell Line; RMA-S: murine equivalent of T2 cell line; and LP: B cell line; and also to B16F1O: mouse melanoma cell line and ND7: mouse neuroblastoma cell line.

Library members found to have high binding activity with specific cells can be used to extract such cells from mixtures. For example a library member of interest may be synthesized on a larger scale and formed into membranes or fibers for the manufacture of filters or adsorbents for the cells of interest.

This procedure has also been used to detect polymers with the ability to bind to Stro-1+ osteoprogenitor cells, and to use the detected polymers as a means of extracting osteoprogenitor cells from unselected human bone marrow mononuclear cell preparations.

In forming a microarray, the library members may be used in solution, preferably dissolved in a common solvent, so that they can be printed on a support in a microarray format.

As an illustrative embodiment of the invention a library of poly(urethanes) was prepared in a parallel synthesis procedure which allowed variation of polyol and isocyanate components and also inclusion and variation of a chain extender. An exemplary library was created from the following: a polyol was selected from: poly(ethylene glycol) poly(propylene glycol) poly(tetramethylene glycol) poly[1,6hexanediol/neopentyl glycol/diethylene glycol-a/t-(adiptic acid)]diol poly[l,6-hexanediol/neopentyl glycol-a/t-(adiptic acid)]diol; a diisocyanate was selected from: 1,6-diisocyanohexane 44'-methylenebis(phenylisocyanate) 4-methyl-i,3-phenylene diisocyanate 1,4-diisocyanobenzene 4,4'-methylenebis(cyclohexylisocyanate) 1,3-bis(isocyanatomethyl)cyclohexane; and a chain extender was selected from: 1,4-butanedjol ethylene glycol propylene glycol ethylene diamine 3-dimethylamino-1,2-propanediol 2-nitro-2-methyl-i, 3-propanediol 2,2,3,3,4,4,5,5-octafluoro-1,6-hexanediol diethyl bis(hydroxymethyl) malonate The invention is illustrated by way of example only, in the following test procedures and experimental methods

Brief Description of the Figures

Figure 1 is a scatter plot representing the inter-slide reproducibility. Each data point represents the background corrected mean fluorescence intensity of each polymer arising from the binding of ND7 cells stained with CellTracker Orange in an experiment run twice on two identical arrays.

Figure 2 shows the effect of the chain extender on the adhesion of ND7 cells. The molecular weight of the polyol (PTMG) is plotted against the background corrected mean fluorescence intensities (arbitrary units) arising from the binding of ND7 cells.

Figure 3 shows polymer specificity with 2 non-adherent cell lineages. Scans (non- processed) were obtained with cell lines stained with CeilTracker Green over an array of 8x8 polymer spots containing a total of 16 polymers. Microarray screening with (a) JY cells, (b) RMA-S cells; (c) schematic representation of the array; (d) cell binding expressed as background corrected mean fluorescent intensity on this array of 16 polymers.

Figure 4 shows enrichment of Stro-1 + osteoprogenitor cells on Pu-i 6 coated- coverslip performed using unselected human bone marrow mononuclear cell preparations. (a), (b) and (c) using FITC-immunolabelled and DAPI-stained Stro-1+ cells; (d), (e) and (f) control (non-immunolabelled, DAPI-stained bone marrow mononuclear cells); (g) Overall proportion of Stro-1 + cells immobilized on coated- coverslips.

Supplementary Figure 1 is a scatter plot representing the inter-slide reproducibility.

Each data point represent the background corrected mean fluorescence intensity of each polymers arising from the binding of B16F1O cells stained with CellTracker Green from an experiment run twice on two identical arrays.

Supplementary Figure 2 (a) is a scatter plot representing the effect of the stain.

Each data point represent the background corrected mean fluorescence intensity of each polymers arising from the binding of ND7 cells stained with CellTracker Orange (b) and CellTracker Green (c) run twice on two identical arrays (non- processed scans).

Supplementary Figure 3 is a scatter plot representing the validity of the duplex screen with ND7 cells. Each data point represent the background corrected mean fluorescence intensity of each polymers arising from the binding of ND7 cells stained with CeliTracker Orange run twice in the presence of ND7 only and a mixture of ND7 and B16F1O cells.

Supplementary Figure 4 is a scatter plot representing the validity of the duplex screen with B16FIO cells. Each data point represent the background corrected mean fluorescence intensity of each polymers arising from the binding of B16F1O cells stained with Celllracker Green run twice in the presence of B16F1O only and a mixture of ND7 and B16F1O cells.

Test Procedures Polymers A library of poly(urethanes) was prepared by parallel synthesis from the following components: Polyol: PEG: poly(ethylene glycol) PPG: poly(propylene glycol) PTMG: poly(tetramethylene glycol) PHNGAD: poly[1,6-hexanediol/neopentyl glycol/diethylene glycol-a/t-(adiptic acid)]diol PHNAD: poly[i,6hexanediol/neopentyl glycol-a/t-(adiptic acid)]diol Dilsocyanate: HDI: 1, 6-diisocyanohexane MDI: 44'-methylenebis(phenylisocyanate) TDI: 4-methyli,3-phenylene diisocyanate PDI: 1,4-diisocyanobenzene HMDI: 44'methylenebis(cyciohexyIisocyanate) BICH: 1,3-bis(isocyanatomethyl) cyclohexane Chain Extender: BD: 1,4-butanedjol EG: ethylene glycol PG: propylene glycol ED: ethylene diamine DMAPD: 3-dimethylamino-1,2propanediol NMPD: 2-nitro-2-methyl-1,3-propanediol OFHD: 2,2,3,3,4,4,5,5octafluoro-1,6-hexanediol DHM: d iethyl bis(hydroxymethyl)malonate The composition of each member of the library which was screened is shown below

in Tablet

Table I

Pu- Polyol Dis Ext x (Polyol) x (Dis) x (Ext) 3 PEG 400 HDI none 0485 0515 0 4 - PPG 2000 HDJ none 0485 0515 0 - - PEG 400 BICH none 0485 0.515 0 - PPG 2000 BICH none 0485 0515 0 P1MG 2000 BICH none 0485 0.515 fl 12 PEG 900 TDI none 0485 0515 ______ 13 PEG 400 TDI none 0485 0515 ______ 14 PPG 2000 TDI none 0485 0.515 ______ PTMG 2000 TDI none 0485 0.515 0 ______ 16 PEG 2000 MDI none 0485 0515 0 ______ 17 PEG 900 MDI none 0485 0515 0 ______ 18 PEG 400 MDI none 0485 0515 0 ______ 19 PPG 2000 MDI none 0485 0.515 0 ______ P1MG 2000 MDI none 0485 0.515 ______ 22 PEG 900 PDI none 0.485 515 0 _______ 23 PEG 400 P01 none 0485 515 0 ______ 24 PPG 2000 PDI none 0485 515 0 ______ PTMG 2000 PDI none 0485 0515 0 _____ PEG 400 HMDI none 0485 0515 0 _____ 29 PPG 2000 HMDI none 0485 0515 0 ______ PTMG 2000 HMDI none 0485 0515 ______ 31 PEG 2000 HDI BD 0250 0.520 230 33 PEG 900 HDI BD 0250 0520 0230 PEG 400 HDI BD 0.250 520 0230 37 PPG 2000 HDI BD 0250 0520 0230 PPG 2000 HDI ED 0.250 0520 0230 P1MG 2000 HDI ED 0250 0 520 0230 41 PEG 2000 BICH BD 0250 0.520 0.230 43 PEG 900 BICH BD 0.250 0.520 0.230 PEG 400 BICH BD 0250 0 520 0230 PEG 400 BICH ED 0250 0520 0230 47 PPG 2000 BICH 80 0250 0520 0230 48 PPG 2000 BICH ED 0250 0520 0230 49 P1MG 2000 BICH BD 0250 0 520 230 P1MG 2000 BICH ED 0250 0520 230 53 PEG 900 TDI BD 0250 0520 230 PEG 400 101 BD 250 0.520 0.230 57 PPG 2000 TDI BD 250 0.520 0230 59 PTMG 2000 TDI BD 250 0 520 0 230 61 PEG 2000 MDI BD 250 0.520 0.230 63 PEG 900 MDI BD U 250 520 0230 PEG 400 MDI BD 0250 520 0230 67 PPG 2000 MDI BD 0.250 520 0230 69 P1MG 2000 MDI BD 0250 520 0 230 71 PEG 2000 P01 BO 0250 0520 0230 73 PEG 900 P01 BD 0250 0.520 0.230 77 PPG 2000 PDI BO 0250 fl 520 0230 79 PTMG 2000 P01 BD 0250 520 0230 81 PEG 2000 HMDI BO 0250 520 0230 83 PEG 900 HMDI BD 0.250 520 0 230 PEG 400 HMDI BO 250 520 n 230 87 PPG 2000 HMDI BD 250 0 520 230 PTMG 2000 HMDI BO 250 0 520 230 91 PTMG 650 HOP BD 485 0515 ______ 92 PTMG 1000 HDI BD u485 0515 ______ 93 P1MG 650 BICH BO 0485 0515 0 ______ 94 PTMG 1000 BICH BD 0485 515 0 ______ PTMG 650 MDI 60 0485 0 515 0 _______ 96 PTMG 1000 MDI BO 0485 0515 0 PHNGAO 1800 BICH DMAPD 0250 0520 230 PPG 2000 MDI DMAPD 0.250 0 520 230 159 PTMG 250 MDI BO 0250 0520 230 P1MG 250 MDI EG 0250 0520 u 230 PTMG 650 MDI EG 0250 0 520 0230 162 PTMG 1000 MDI EG 0.250 0520 0230 PTMG 2000 MDI EG 250 0 520 0 230 164 PTMG 250 MDI PG 250 0520 0230 Th PTMG 650 MDI PG 250 0.520 0230 9ë P1MG 1000 MDI PG v250 0520 0230 PTMG 2000 MDI PG 0250 n 520 0230 P1MG 250 BICH none 0485 515 0 169 P1MG 650 BICH none 0485 515 0 P1MG 1000 BICH none 0485 515 0 171 PTMG 250 HOI none 0485 0515 0 172 P1MG 650 HOP none 0485 0.515 0 174 PTMG 250 MDI none 0485 0515 0 PTMG 650 MDI none 0485 0 515 0 176 P1MG 1000 MDI none 0485 515 0 177 P1MG 250 HDI NMPO 0250 520 0230 P1MG 1000 HDI NMPD 0.250 520 0 230 T P1MG 2000 HDI NMPD 0250 0520 0230 PTMG 1000 BICH NMPD 0250 0 520 230 TE P1MG 2000 BICH NMPD 0250 0 520 230 P1MG 650 MDI NMPD 0.250 0 520 230 183 PTMG 1000 MDI NMPD 0250 520 0230 P1MG 2000 MDI NMPD 250.52fl 0230 PHNAD 900 MDI OFHD 170 520 0330 iä PTMG 650 BICH OFHD 250 v.520 0230 1T P1MG 1000 BICH OFHO v250 0520 230 1ä PTMG 2000 BICH OFHD 0250 0520 230 189 PPG 1000 BICH OFF-ID 0 170 0 520 330 P1MG 650 HDI OFHD 0250 0.520 0. 230 191 PTMG 1000 HDI OFHD 0250 0520 0230 P1MG 2000 HOl OFHD 0.250 0 520 0230 PPG 1000 MDI OMAPO 0.170 0520 0.330 PTMG 650 MDI OFHO 0.250 0.520 0230 i P1MG 1000 MDI OFHO 0250 0 520 230 PTMG 2000 MDI OFHD 0250 0 520 230 197 P1MG 650 BICH OHM 0.250 0520.230 198 P1MG 1000 BICH OHM 250 0.520 0230 199 PTMG 2000 BICH OHM.250 0520 0.230 P1MG 650 HDI OHM 250 0 520 0230 - r P1MG 1000 HDI OHM 0250 0520 0230 202 PTMG 2000 HDI OHM 0250 0 520 230 203 PTMG 650 MDI OHM 0250 0 520 230 P1MG 1000 MDI OHM 0250 0 520 230 PTMG 2000 MDI OHM 0.250 520 v 230 W PPG 1000 HOP OFHO 0250 520 0230 PPG 1000 BICH OFHD 0250 520 0230 ä PPG 1000 MDI OFHD 0250 0520 0230 PPG 1000 HDI PG 0250 0520 0230 iö PPG 1000 BICH PG 0250 0 520 0230 211 PPG I 1000 MDI PG 0250 0520 0230 212 I PHNAD I 900 HOP PG 0250 I 0520 0230 L213 PHNAD 900 BICH PG 0250 0520 0230 214 PHNAD 900 MDI PG 0250 0520 0230 [ 215 PHNAD 900 HDI BD 0250 0.520 0230 [16 PHNAD 900 BICH BD 0.250 0.520 0230 217 PHNAD 900 MDI BD 0250 0520 0230 Each member of the library was dissolved in a common solvent and transferred into a 384 well plate prior to contact printing of a microarray. For effective formation of a microarray, a number of parameters, such as the nature of solvent and substrate, inking and printing time need to be optimised to ensure uniformity of the polymer spots within the array.

Selection of Solvent To obtain uniform printing, the polymer library is printed from a common, non- volatile solvent. l-methyl-2-pyrrolidinone (NMP) was finally selected as majority (>95%) of the polymer library members obtained using the components given above were found to be soluble in that solvent, allowing good, uniform, spots to be printed.

The formation of so called "rings" (see Deegan, R.D., Capillary flow as the cause of ring stains from dried liquid drops. Nature 389, 827-829, 1997) during solvent evaporation was minimized by a combination of highly viscous solvent and successive layering of polymer solutions (5 stamps per spot).

Selection of Substrate In order to develop a cell compatible assay in a microarray format, the substrate needs to comply with the following requirements. Firstly, the substrate had to be unaltered by the contact printing of polymer solution in NMP which ruled out the use of polymer coatings such as poly(hydroxyethyl methacrylate) p(HEMA), which would be dissolved locally and give rise to polymer mixtures. Secondly, a substrate with low levels of background cell binding had to be developed to facilitate data analysis and thirdly, the substrate had to be stable under UV-irradiation to allow sterilization prior to the plating of the cells. The best results were obtained by dip-coating aminoalkylsilane slides (Silane-PrepTM; Sigma) with a thin film of agarose. Although agarose gels have been used to amplify loading on DNA arrays (Afanassiev, V., Hanemann, V., Wolfl, S., Preparation of DNA and protein micro arrays on glass slides coated with an agarose film. Nucleic Acids Res. 28, e66, 2000) and are known to inhibit cellular adhesion in a number of different formats (Folch, A., Toner, M., Microengineering of cellular interactions. Annu. Rev. Biomed. Eng. 2, 227-256, 2000; Roth, E.A., et al., Inkjet printing for high-throughput cell patterning.

Biomaterials 25, 3707-371 5, 2004), agarose has not been used as coating material for cell based microarray assays.

Preparation of Arrays The polymer arrays were fabricated by contact printing using polymer solutions in NMP. Once printed, the slides were dried overnight under vacuum at 45 C and were sterilized by exposure to UV irradiation for 15 minutes prior to cell plating.

Screening Cells The polymer arrays were screened with both adherent and suspension mammalian cell lines (provided by Salim Khakoo, Cancer Research, Southampton, UK) stained with membrane-permeant fluorescent dyes. Cell binding was evaluated (FIPS, LaVision Biolech GmbH, D) by integrating the fluorescent intensity across the whole area of the spot. The mean fluorescence intensity and coefficient of variation for each set of 4 identical polymer spots was then calculated. Background correction was carried out by subtracting the mean fluorescent intensity of the 32 "empty" spots (one array of 4 x 4 spots missing in each field) from the mean fluorescence intensity calculated for each library member.

Initial sets of experiments were carried out to evaluate the reproducibility of the method, both intra and inter-slide. The intraslide reproducibility was assessed by calculating the coefficient of variation among the 4 identical spots of each polymer; whereas the interslide reproducibility was evaluated by calculating the correlation coefficient when plotting the background corrected fluorescence intensities for each of the 120 polymers resulting from the same experiments run on two identical arrays.

The initial screening was performed using B16FIO and ND7 mouse cell lines stained with CeilTracker Green and Orange (Molecular Probes), respectively. The average intra-slide variation was minimal for ND7 cell stained with CellTracker Orange (average CV of 13%), while in the case of B16F1O cells stained with CellTracker Green, the average CV was 30%. The inter-slide reproducibility, (Figure I and Supplementary Figure 1) evaluated using two identical slides, gave correlation coefficients of r2=0.85 and 0.79 with ND7 and B16FIO cells respectively, showing good chip to chip reproducibility and showing the robustness of the method.

Subsequent experiments assessed the possible effect of the stain on cellular adhesion by comparing ND7 adhesion (using CellTracker Orange) to the adhesion of ND7 cells stained with CellTracker Green. When comparing the cellular adhesion of these two experiments, the correlation coefficient was r2=0.73, which showed that the nature of the stain had minimal effect on the cellular binding (Supplementary Figure 2).

The feasibility of simultaneous screening of two different cell lines was carried out by screening a mixture of ND7 cells labelled with CeliTracker Orange and B16F1O cells labelled with CeliTracker Green (1.5x106 cells of each lineage) plated onto the array of polymers. The slide was scanned using both Cy3 and FITC filters and the cellular adhesion for each cell line was evaluated as described previously. In this experiment, the intraslide reproducibility was similar to the single cell experiments (average CV's calculated from the 4 identical polymer spots were 13% and 27% for ND7 and BI6F1O respectively). Comparison of the double and single cell experiments gave correlation coefficient for ND7 and BI6F1O cells of r2=O.82 and r=0.68 respectively (Supplementary Figures 3 and 4) showing the validity of the duplex screen.

Whilst studying the relationship between cellular adhesion and composition of different polymers, it was possible to elucidate the effects of both the molecular weight (MW) of the polyol and the nature of the chain extender on cellular binding (Figure 2). Thus an increase in binding of ND7 cells was observed when the MW of the polyol decreased in the series of 14 poly(urethanes) which were prepared from poly(tetramethylene glycol) (PTMG) and 4,4'- methylenebis(phenylisocyanate) (MDI).

Additionally, the presence of a chain extender in the backbone of the polymers increased cell adhesion, with 2,2,3,3,4,4,5,5-octafluoro-1,6hexanediol (OFHD) giving better cell adhesion than both 2-nitro-2-methyl1,3-propandiol (NMPD) and I,4-butanediol (BD).

The immobilization of non-adherent cell lines (JURKAT, RMA-S and JY, all labelled with CellTracker Green) was investigated using the polymer microarrays. With JURKAT cells, none of the 120 polymers showed cellular adhesion and growth, however both RMA-S and JY cells showed significant polymer specific immobilization. The immobilization of these non-adherent cells was shown to be highly dependent on both the structure and properties of the polymer, and the type of the cells investigated, as shown in Figure 3, where cell binding is shown with an array of 16 polymers. The selectivity toward these two cell lines is remarkable with PU-I 98, 199 and 202 binding RMA-S only, whereas PU-i 94, 195, 207 and 210 bind predominantly JY cells. PU-21i and 214 were the only two polymers giving moderate cellular immob ilisat ion with both cell lines.

One of the current approaches for enrichment of osteoprogenitors from human bone marrow involves immunoselection of these progenitor populations by monoclonal antibodies such as Stro-1. To determine the efficacy of the polymers in facilitating a potential enrichment of osteoprogenitors from human bone marrow by virtue of their cell binding specificity, the full PU array was incubated with unselected human bone marrow mononuclear cell preparations. Polymer arrays incubated with Stro- 1 + cells, isolated from bone marrow cell preparations by magnetically activated cell sorting (MACS - Stewart, K., et al., Further characterization of cells expressing STRO-i in cultures of adult human bone marrow stromal cells. J. Bone Miner. Res. 14, 1345-1356, 1999), served as positive controls for confirmation of the results.

Detection was facilitated by selectively immunolabelling Stro-1 + cells using STRO-1 mouse monoclonal primary antibody (Simmons, P.J., TorokStorb, B., Identification of stromal cell precursors in human bone marrow by a novel monoclonal antibody, STRO-i. Blood 78, 55-62, 1991), followed by the (FITC)-conjugated AffiniPure F(ab')2 fragment Goat anti-mouse 1gM and the nuclear stain Hoechst 33342.

Several polymers showed high affinity for the Stro-1+ cells isolated by MACS. Using the results of this first screen, the polymers showing the highest level of Stro-1+ immobilization were reprinted on a single array containing 16 identical spots for each of the 31 polymers.

These focused arrays were subsequently incubated with unselected human bone marrow mononuclear cell preparations. Several of the poly(urethanes) identified in the first step were shown to selectively bind fluorescently immunolabelled Stro-i+ osteoprogenitor cells from human bone marrow (fluorescent intensity associated with both FITC and DAPI). In order to confirm these results and study the possible scale-up of the method as a means of cell enrichment, 3 polymers, PU-16, 17 and 61, which were found to exhibit high selectivity for Stro-1+ cell binding, were spin- coated onto glass coverslips. The polymer-coated coverslips were then incubated with unselected bone marrow mononuclear cell preparations (from 3 individuals) in which the Stro-1+ cells were immunolabelled. Control experiments were also carried out by incubating polymer-coated coverslips with non-immunolabelled unselected human bone marrow mononuclear cell preparations in order to obtain the background cell intensity for each poly(urethane) substrates. Quantitative analysis involved measurement of the background corrected FITC intensity of each immobilized Stro-1+ cell over 5 randomly selected areas (1230 by 940 microns). For a cell to be deemed as Stro-1+, its background corrected FITC intensity had to be over the mean plus standard deviation of the control cells. Although the results from patient to patient showed variation in terms of overall immobilized cell density and Stro-1+ proportion, the overall resultsshowed significant level of enrichment of the Stro-1 population (Figure 4).

The development of high throughput screening of polymer libraries is essential for a better understanding of the factors that dictate the biocompatibility of such materials. In this HT screening method, we have demonstrated the use of an easy to prepare agarose gel substrate that allows the development of low background cell microarrays. The study shows the robustness of the method with excellent inter- and intra-slide reproducibility which is essential when developing any screening method. Screens with non-adherent cells showed that several poly(urethanes) from the library were able to immobilize such lineages. However the true power of the method is demonstrated by the cell specificity of certain polymers, with several polymers selectively immobilizing human mesenchymal progenitor cells (Stro-1+ fraction) from bone marrow.

This process has major advantages over MACS, for example there is no need for specific antibodies like Stro-1 (which are not always available), the MACS technique is specialized, cumbersome and time-consuming; on the other hand, polymers are cheap and readily prepared en masse and can be coated and used in a variety formats. This platform allows the isolation and immobilization of specific cells from heterogeneous mixtures and will provide an important tool for the medical community where selective enrichment/purification of specific cell types is desired.

Using this approach, a whole library of bio-compatible polymers presenting a wide range of properties can be screened in a single experiment. Furthermore, each library member has been physically characterized to facilitate the successful synthesis of the identified polymers on a larger scale.

Polymer microarray cell immobilization using agarose substrates is a reliable technology, with the potential to be developed for use with a large variety of cell- specific applications.

Detailed Procedures Substrate Preparation Coating with agarose was achieved by dip-coating the slide in a 1% w/v solution of agarose Type IB (Sigma) at 65 C followed by removal of the coating on the bottom side. After drying overnight at room temperature, the coated slides could be stored or used immediately for printing.

Array Printing The polymer arrays were fabricated by contact printing (Qarray mini, Genetix) with 16 aQu solid pins (K2785; Genetix; UK) using 1% polymer solutions (w/v) in 1- methyl-2-pyrrolidinone placed into polypropylene 384-wells microplates. The following printing conditions were used, 5 stamping per spot, 200 ms inking time and 10 ms stamping time. The typical spot sizes was 300-320 pm in diameter with spot to spot distances of approximately 1125 pm allowing up to 480 polymers to be printed on a standard 25 x 75 mm slide. In order to assess the reproducibility of the screen, only 120 polymers (from our library of 278 pure poly(urethanes)) were printed in quadruplicate within 2 fields of 16 x 16 spots, while within each field a pattern of 4 x 4 spots was left empty and was utilized for background correction.

Once printed, the slides were dried overnight under vacuum at 45 C and were sterilized by exposure to UV irradiation for 15 minutes.

Coverslip Coating Poly(urethanes) were dissolved at 2% w/v in tetrahydrofuran (THF). These solutions were spin-coated onto 22 mm diameter glass coverslip using a P6708 Spincoater (Speedline Technologies, US). Following coating, the coverslips were dried under vacuum overnight at 45 C and sterilized by exposure to UV irradiation for 15 minutes.

Cell Culture Adherent and non-adherent cell lines were grown in DMEM and RPMI growth medium, respectively. Both media were supplemented with 10% v/v heat inactivated fetal calf serum, penicillin (100 units/mL), streptomycin (100 mg/mL) and L- glutamine (4 mM) at 37 C with 5% CO2. Cells were stained with Celllracker Green or Orange (C2925 or 02927, Molecular Probes) according to the manufacturer's protocols. Prior to seeding onto the polymer array, 3 X 10 cells were suspended in mL of medium. The slides were subsequently incubated at 37 C with 5% CO2 for 24 h. After a controlled washing in 30 mL PBS (2 mm shaking on a microplate shaker at 600 rpm), the cells were fixed with 3.7% w/v p- formaldehyde and 4.0% w/v sucrose in H20 for 15 mm, rinsed and stored in PBS at 4 C. Image capture was carried out with a Bioanalyser 4F/4S white light scanner (LaVision BioTech GmbH, D) using an FITC or Cy3 filter.

Stro-1+ cells were isolated using the magnetically activated cell sorting (MACS) technique described by Stewart and co-workers (Stewart, K., et al., Further characterization of cells expressing STRO-1 in cultures of adult human bone marrow stromal cells. J. Bone Miner. Res. 14, 1345-1356, 1999) In brief; red blood cells were removed by centrifugation using lymphoprep solution (Robins scientific, Solihull, UK). The cells from the buffy coat (bone marrow mononuclear cells) were resuspended at 1x108 cells per 10 ml blocking solution (HBSS, 10mM HEPES with 5% (vlv) FCS, 10% (v/v) human normal AB serum, and 1% (wlv) bovine serum albumin/ BSA), followed by incubation with the Stro-1 mouse monoclonal antibody (undiluted culture supernatant from the Stro-1 hybridoma; provided by Dr. J. Beresford, University of Bath). Cells were then incubated either with MACS anti-lgM beads (1:5 dilution, Miltenyi Biotech, Bisley, UK) or with the fluorescein (FITC)- conjugated AffiniPure F(ab')2 fragment Goat anti-mouse 1gM, p chain specific (1:20 dilution, Jackson lmmunoResearch Laboratories, Inc., Baltimore, USA), after washing the excess STRO-1 antibody with MACS buffer (HBSS, 10mM HEPES containing 1% BSA). Cell suspension incubated with the MACS anti-lgM beads was added to a column within the magnet and the Stro-1 negative fraction was eluted using MACS buffer, since the magnetically labelled Stro-1 positive cells were held in the column under the influence of the magnetic field. The column was washed to remove traces of the Stro-1 negative fraction and the Stro-1 positive fraction was then eluted in 1 ml MACS buffer in absence of the magnet.

Stro-1+ cells, isolated using MACS, and preparations of unselected bone marrow mononuclear cells containing FITC-labelled Stro-1 cells were resuspended in 1.5 ml a-MEM (with 10% FCS) at densities of 5 x and 1 x lO7cells respectively. The cell suspensions were directly plated onto polymer arrays and incubated at 37 C, 5%CO2for 19h.

Each polymer-coated coverslip was incubated at 37 C, 5% CO2 for 19 h. with 1 x unselected bone marrow mononuclear cells containing FITC-labelled Stro-1+ cells per well of a 6-well plate.

Following incubation and thorough washing with PBS, cells were fixed with 4% wlv p-formaldehyde for 30 mm, rinsed in PBS and nuclei were stained using Hoechst 33342 (Sigma; 5.ig/mL for 15 mm). MACS-isolated Stro-1 cells bound to the polymers on the slide were fluorescently immunolabelled using Stro-1 mouse monoclonal primary antibody, followed by the (FITC)-conjugated AffiniPure F(ab')2 fragment Goat anti-mouse 1gM. Slides were stored in PBS at 4 C. Coverslips were mounted in aqueous mounting medium and stored at 4 C. Automated image capture and analysis was carried out using the High Content Screening (HCS) platform on PathfinderlM (IMSTAR. S.A., www. imstar.fr).

Claims (1)

1. Use of agarose to provide a coating having a cytophobic effect.

2. Use according to claim 1 in which the coating is formed on a support used for screening an array of polymer samples.

3. A method of screening polymers for cell-binding properties comprising: forming a polymer library; forming an array of samples of the library members on a support that has been coated with agarose to provide a cytophobic surface; incubating the array with a cell suspension; observing the array to determine the binding activity of the library samples with specific cells.

4. A method according to claim 3 in which the array support is a glass slide.

5. A method according to claim 4 in which the glass slide is pre-treated with an aminoalkylsilane 6. A method according to any one of claims 3 to 5 in which the array of samples is a microarray.

7. A microarray support for assessing cell-binding properties of samples of interest, in which the support is coated with agarose as a cytophobic substrate for a microarray of the samples.

8. A support according to claim 7 in which is a glass slide.

9. A support according to claim 8 in which the glass slide is pre-treated with an aminoalkylsilane.

10. A method of screening polymers for cell-binding properties comprising: forming a polymer library by parallel synthesis; at least partially characterizing the library members; forming a microarray of the library members; incubating the microarray with a cell suspension; observing the microarray to determine the binding activity of library members with specific cells.

11. A method according to claim 10 in which the polymer fibrary is a library of poly(urethanes) formed from variable polyol, polyisocyanate and chain extender components.

14. A method according to claim 10 or 11 in which the microarray support is coated with agarose before forming the microarray.

15. A method according to claim 14 in which the microarray support is a glass slide.

16. A method according to claim 15 in which the glass slide is pretreated with an aminoalkylsilane 17. A method of isolating desired cell lines from a mixture of unselected cell preparations comprising identifying a polymer or polymers that bind to the desired cell line using the method claimed in any one of claims 10 to 16, and using the identified polymer or polymers as a substrate for contact with unselected cell preparations for selective immobilisation of the desired cell line.

18. A method according to claim 17 comprising identifying a polymer or polymers that bind to osteoprogenitor cells, and using the identified polymer or polymers as a substrate for contact with unselected human bone marrow mononuclear cell preparations for selective immobilisation of osteoprogenitor cells.

19. A polyurethane formed from polyol, polyisocyanate and optional chanikn extender components in which the polyol is selected from one or more of: poly(propylene glycol) poly(tetramethylene glycol) poly[1,6hexanediol/neopentyl glycol/diethylene glycol-a!t-(adiptic acid)]diol poly[1,6-hexanediol/neopentyl glycol-a/t-(adiptic acid)Jdiol.

20. A polyurethane according to claim 19 in which the isocyanate is selected from one or more of: I,6-diisocyanohexane 44'-methylenebis(phenyIisocyanate) 4-methyl-I,3-phenylene diisocyanate 1,4-diisocyanobenzene 4,4'-methylenebis(cyclohexyljsocyanate) 1,3-bis(isocyanatomethyl)cyclohexane 21. A polyurethane according to claim 19 or 20 in which the chain extender is selected from one or more of: 1,4-butanediol ethylene glycol propylene glycol ethylene diamine 3-dimethylamino-1,2-propanediol 2-nitro-2-methyl-1,3-propanediol 2,2,3,3,4,4,5,5-octafluoro-1,6-hexanediol.

diethyl bis(hydroxymethyl)malonate.