SEARCH

SEARCH BY CITATION

Keywords:

  • cell microarrays;
  • germ cells;
  • Sertoli cells;
  • cell isolation;
  • cell surface markers

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. CONCLUSIONS
  6. Acknowledgements
  7. LITERATURE CITED
  8. Supporting Information

Cell microarrays can serve as high-throughput platforms for the screening of a diverse range of biologically active factors and biomaterials that can induce desired cellular responses such as attachment, proliferation, or differentiation. Here, we demonstrate that surface-engineered microarrays can be used for the screening and identification of factors that allow the enrichment and isolation of rare cells from tissue-derived heterogeneous cell populations. In particular, we have focused on the enrichment of bovine testicular cells including type A spermatogonia and Sertoli cells. Microarray slides were coated with a copolymer synthesized from poly(ethylene glycol) methacrylate and glycidyl methacrylate to enable both the prevention of cell attachment between printed spots and the covalent anchoring of various factors such as antibodies, lectins, growth factors, extracellular matrix proteins, and synthetic macromolecules on printed spots. Microarrays were incubated with mixed cell populations from freshly isolated bovine testicular tissue. Overall, cell attachment was evaluated using CellTracker™ staining, whereas differential attachment of testicular cells was determined by immunohistochemistry staining with Plzf and vimentin antibodies as markers for type A spermatogonia and Sertoli cells, respectively. The results indicate that various surface immobilized factors, but in particular Dolichos biflorus lectin, allowed the enrichment of Plzf positive cells. Furthermore, Pisum sativum lectin, concanavalin A, collagen type IV, and vitronectin were identified as suitable negative selection factors. To our best knowledge, this work is the first to demonstrate the utility of surface engineered cell-based microarrays for the identification of factors that allow the selective capture of rare cells from tissue isolated heterogeneous mixtures. © 2010 International Society for Advancement of Cytometry

The development of smart cell-culture substrate materials that can selectively capture rare cells from a tissue-isolated mixture of cells and encourage self-renewal of these cells would be greatly beneficial for many stem cell applications. Common methods for the screening of surface-bound signals suitable for selective cell isolation include assays that routinely use coated multiwell plates and culture flasks. These methods generally require large amounts of cells and reagents, and the throughput of materials is limited (1–3). More recently, cell microarray technologies have proved increasingly useful for addressing these limitations by providing miniaturized microenvironments for the screening of stem cell-surface interactions (4, 5). Current robotic microarray spotters are capable of dispensing nanoliter volumes and creating spots in a size range that is suitable for cell microarrays (6–9). Using such robotic tools, it is possible to produce microscale spots featuring individual surface-bound signals or, alternatively, mixtures of signals in the desired ratios (6, 7). Cell microarrays have been implemented for many different biomedical applications including antibody-based microarrays for immunophenotyping (10–12) and high-throughput diagnostic tools for cancer profiling (13–15), whereas cell microarrays have also been used successfully in drug discovery (16). It has even been shown feasible to micro-engineer extracellular matrix (ECM) microarrays for studying biological conditions that induce differentiation of mouse embryonic stem cells down a hepatic lineage (6). Likewise, Soen et al. captured primary neural progenitor cells on a cell microarray composed of mixtures of ECM proteins, morphogens, and other signaling proteins and identified combinations of signals promoting neurogenesis, others promoting gliogenesis as well as signals maintaining the cells in an undifferentiated-like, proliferative state (9). Others have fabricated intricate, synthetic polymeric microarrays displaying libraries of polyacrylates for high-throughput combinatorial screening of human embryonic stem cells (7).

A relatively recent area that has garnered interest in the fields of genetic engineering and reproductive technologies are methods for the isolation of spermatogonial stem cells (SSCs), a subpopulation of type A spermatogonia and the stimulation of their self renewal (17). SSCs have recently been demonstrated to be pluripotent (18–21), and successful germ cell transfer has been established in mice, rats, sheep, cattle, monkeys, and humans (22–24). Although suitable methods for transplantation have been demonstrated, it has proved difficult to replicate the stem cell niche necessary for in vitro expansion of SSCs. Species-specific SSC surface markers have been described in humans and rodents (25, 26), but few have been established for livestock species such as cattle and sheep (27, 28). The lack of methods for isolation and subsequent culture system for spermatogonia in vitro has severely hindered the application of germ cell transfer technology in livestock species. Instead of harnessing in vitro technologies, the SSC niche is currently replicated by allo- or xenografting of the SSC of interest, an expensive, and cumbersome practice (22–24). Therefore, there is an imminent need to identify the surface markers for spermatogonia that can be used for cell isolation and expansion.

Using bovine testicular cells as a model, the work presented herein demonstrates the utility of surface-engineered microarray platforms for the screening and identification of biological and synthetic surface-bound factors that support selective cell adhesion from bovine testis tissue isolates. This study was predicated on the fact that only a limited amount of cell surface markers have been discovered for germ cells, particularly SSCs, and many of these are species dependent (27, 29, 30). Information on the bovine testis system is particularly scarce, yet any insights would be of significant value to cattle breed engineering (31). The scarcity of information also applies to immunohistochemical markers. For microarray analysis, we hence draw on immunohistochemistry developed for other species, realizing that some interspecies variation in these markers is expected. Although the focus of this study is on the bovine testicular cell model, we believe that this cell-screening microarray platform holds great promise for a range of applications, including those in human regenerative medicine.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. CONCLUSIONS
  6. Acknowledgements
  7. LITERATURE CITED
  8. Supporting Information

Fabrication of Microarray Substrates

Partially frosted glass slides (Biolab, USA) were surface modified as described elsewhere (32, 33). Briefly, glass slides were first coated with an allylamine (Aldrich, 98 % purity) plasma polymer (ALAPP) using radio frequency glow discharge deposition in a custom-built reactor defined by a height of 35 cm and a diameter of 17 cm. The lower and upper electrodes were both circular, with a diameter of 10.5 cm and separation of 16 cm. Samples were placed on the lower electrode. Plasma polymer coatings were deposited using a frequency of 200 kHz, a power of 20 W, an initial monomer pressure of 0.200 mbar, and a treatment time of 25 s.

Freshly ALAPP-coated glass slides were subsequently spin coated with a 2.5 (w/v) copolymer solution in dioxane containing a crosslinker. The copolymer was freshly synthesized from poly(ethylene glycol) methyl ether methacrylate (PEGMA, Aldrich, MW 475 g/mol) and glycidyl methacrylate (GMA, Fluka) in dioxane (BDH Chemicals) in a 1:1 (w/w) ratio using 2,2′-azobisisobutyronitrile (Aldrich) as an initiator. ω,ω′-Bis(2-aminoethyl)polyethylene glycol (PEG diamine, MW 3400 g/mol, Fluka) was used as a crosslinker in the spin-coating solution at a concentration of 0.2% (w/v). Spin coating was performed at 5,000 rpm for 30 s after placing ALAPP-coated glass slides on the spin coater (WS-400B-6NPP/Lite, Laurell Technologies Corporation, USA) and covering the entire surface with the solution containing the copolymer and the crosslinker. After spin coating, the remaining solvent was removed in a vacuum.

Printing of Microarrays

Before printing, the PEGMA/GMA copolymer-coated glass slides were soaked in sterile Dulbecco's phosphate-buffered saline solution (PBS) for 2 h at 37°C, rinsed with sterile MilliQ water, and dried under a stream of nitrogen. The lectin factors printed on the arrays, Pisum sativum (PSA), Dolichos biflorus (DBA), Lycopersicon esculentum, wheat germ agglutinin, concanavalin A (ConA), peanut agglutinin (PNA), and Datura stramonium (DSA) were purchased from Sigma. ECM proteins fibronectin (FBR), fibronectin-like engineered protein (FEP), laminin (LMN), collagen type I (Coll I), collagen type IV (Coll IV), vitronectin (VTR), and other biological factors including leptin (LP), chondroitin (COD), heparin (HP), and bovine serum albumin (BSA) were also purchased from Sigma. Synthetic factors printed on the arrays including amine-terminated PAMAM dendrimers generation 2 (D2) and generation 3 (D3) were received from Aldrich, whereas polyallylamine (PAL) and polyethyleneimine (PEI) were purchased from Sigma. Growth factors glial-derived neurotrophic factor (GDNF) and stem-cell factor (SCF) as well as antibodies against GFR alpha 1 (GFR-α1), CD9 and CD49f (integrin α6β1) were received from Chemicon. Serial dilutions of all printing factors (200 μg/ml, 100 μg/ml, and 25 μg/ml) were carried out in sterile PBS with the exception of collagen type I and IV, which were diluted in 1 M acetic acid to prevent gelling and precipitation of collagen. A Cy3-labeled antibody (anti-goat IgG, Sigma) was used as a fluorescent locator dye on the microarrays. Thirty microliters of aliquots of all factor dilutions were placed in 384-well (Nunc, Denmark) source plates. A BioOdyssey Calligrapher (BioRad, USA) was used for the contact printing of the microarrays. A solid pin print head (ArrayIt, SSP015, tip diameter 375 μm) was cleaned by sonication in MilliQ water for 10 min before printing. To avoid cross-contamination, pins were cleaned in PBS buffer solution containing 0.1% Tween-20 (Sigma), followed by rinsing in MilliQ water and drying (cycled thrice) between printing of each factor and each dilution. During printing, the chamber was held at 11°C and at 65% humidity. After printing, the microarrays were stored at 4°C for 48 h to allow for sufficient binding of the factors to the reactive surface of the glass substrates.

Acquisition of Tissue Isolate from Bovine Testis

Animals were handled and treated according to the guidelines of the Animal Ethics Committee at CSIRO, Armidale. Calves were castrated under general anesthesia (xylazine at 0.1 mg/kg followed by ketamine at 3 mg/kg). Antibiotics (long-acting penicillin) and an analgesic/anti-inflammatory (flunixen at 2 mg/kg) were administered postoperatively. After castration, the testes were washed in Dulbecco's PBS. The tunica albuginea, testes, epididymides, and excess connective tissue were removed. A two-step enzymatic isolation procedure described elsewhere (22) was adapted to isolate individual tubular cells. In brief, the tunica vaginalis was removed, and the testes were weighed and then washed in sterile PBS. A segment of 6–10 g tissue (from 50 g testes) was used as an isolation unit. Testis tissue was dissected free of the rete testis and connective tissue and transferred to a Petri dish that contained 5 ml Dulbecco's modified Eagle's medium/F12 (DMEM/F12) with 100 U/ml penicillin and 100 μg/ml streptomycin (Gibco). After rinsing, the tissues were transferred to a second Petri dish containing 5 ml DMEM/F12 and chopped finely. The tissues were placed in a tea strainer and ground with a 5 ml syringe plunger. The remaining material was transferred into a 50 ml Falcon tube and incubated with collagenase (1 mg/ml; type IV, Sigma) in a shaking water bath at 37°C. During this period, the tissue samples were frequently checked by microscopy, and the reaction was stopped when individual tubules appeared under microscopic observation. The supernatant was removed, and the tissue was rinsed five times in PBS at room temperature. The fragments were then treated with trypsin (2.5 mg/ml; Gibco BRL) in PBS for 5–10 min at 37°C. DNAse I (7 mg/ml; Sigma) in DMEM was added 1 min after trypsin treatment. An equal volume of heat-inactivated fetal bovine serum (FBS, Gibco) was used to inactivate trypsin digestion. The resultant cell suspension was then filtered though a cell strainer with two layers of nylon mesh (upper layer of 96 μm and lower layer of 55 μm pore size) and centrifuged at 400g for 5 min at room temperature. The pellets were resuspended in 10 ml DMEM containing 5% FBS at a density of 4–8 × 107 cells/ml. Cell viability was assessed by Trypan blue exclusion.

Incubation of Tissue Isolate on the Microarrays

Immediately before incubation with cells, printed microarrays were soaked in sterile PBS for 1 h at 37°C and gently rinsed with MilliQ water to remove any noncovalently bound factors. The microarrays were then dried in ambient conditions, and the perimeters of the arrayed substrates were encased with a wax barrier pen (Dako) to minimize the cell-suspension volumes required for incubation. The microarrays were placed in polystyrene Petri dishes (Nunc) converted into humidity chambers. The cell seeding density on the arrays was adjusted to 1.5 × 105 cells/cm2. Cell incubation occurred for 2 h at 37°C and 5% CO2 in DMEM. After incubation, the microarrays were gently rinsed with PBS three times and dried in ambient conditions.

Immunohistochemistry and Fluorescent Labeling of Cells on Microarrays

Putative Type A spermatogonia were identified using a Plzf antibody (anti-Plzf, Santa Cruz), whereas a vimentin antibody (anti-vimentin, Zymed) was used to identify Sertoli and myoid-cell populations (34, 35). The location and percentages of Plzf+ and vimentin+ cells were determined by immunohistochemical staining of the cells on the microarrays. The smears and microarrays were stained for Plzf and vimentin according to standard immunohistochemistry protocols and counterstained with hematoxylin (Merck) to determine total cell attachment. Initially, the microarrays and smears were fixed in 50 ml 37% formaldehyde, 75 ml ethanol, 25 ml glacial acetic acid, and 250 ml deionized H2O (MDF) for 2 min. Slides were incubated with anti-Plzf 1:100 (v:v) in 0.5% BSA in Tris-buffered saline (TBS, containing 6.06 g Tris–HCl, 0.93 g Tris–Base, 8.7 g NaCl in 1 l deionized H2O, pH 7.5), and separate slides were incubated with antivimentin (in stock solutions supplied by Zymed) overnight at 4°C. The samples were rinsed with TBS, and Envision Dual Link (Dako) was applied and incubated for 30 min. The samples were counterstained with hematoxylin for 1 min and dehydrated by briefly washing in a series of graded alcohols (50, 70, 90, and twice in 100% ethanol in water) and a final wash with xylene. The microarrays were mounted with DePex mounting media and cover-slipped. Mouse and rabbit immunoglobin (Sigma) at 1 μg/ml in PBS were used as negative controls.

Complementary staining to assess overall cell attachment was performed with a fluorescent CellTracker probe (CellTracker™ Orange, Invitrogen) and was carried out according to the standard protocol supplied by Invitrogen immediately following cell incubation on the microarrays.

Analysis and Scoring Methods

Immunohistochemical staining of immobilized cells on the arrays was evaluated with a Nikon Eclipse E600 microscope with 20×, 40×, and 100× objectives. Scoring was carried out by scanning across the microarray counting Plzf+ cells, vimentin+ cells, and total cell attachment for each arrayed spot. Scoring was not carried out on the 25 μg/ml concentration of printed factors on the microarrays due to the very low cell attachment observed for all factors screened. For the 100 and 200 μg/ml concentration portions of the arrays, all spots that featured >10 adhered cells were considered greater than background and scored accordingly. A Nikon Eclipse E600 upright microscope was also used to observe cell attachment and cell morphology in fluorescence mode on fluorescently labeled cell microarrays. CellTracker™ Orange labeled samples were also scanned with a Typhoon 9400 scanner (Amersham Biosciences Corp.) using a 532 nm excitation filter at 25 μm resolution.

Statistical Methods

Five replicate spots per printed factor at each concentration were scored, and statistical analysis was determined by a one-way ANOVA with a Dunnett post hoc test (Kaleidagraph software). Error bars in all figures correspond to standard error measure (n = 5).

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. CONCLUSIONS
  6. Acknowledgements
  7. LITERATURE CITED
  8. Supporting Information

Surface Chemistry of the Microarray Substrates

The microarray substrates needed for the experiments carried out in this study had two main requirements: first, the prevention of cell attachment between printed spots and second, the covalent immobilization of printed factors. The surface chemistry used for the preparation of the microarray substrates used in this study has been previously reported (32, 33). As described in the experimental part, the glass slides consisted of a thin interlayer coating of ALAPP (∼30-nm thickness) for the purpose of introducing amine functional groups on the glass surface. Subsequent spin coating with a solution containing a freshly synthesized copolymer of PEGMA and GMA in dioxane as well as bis(2-aminoethyl) polyethylene glycol (as a crosslinker) enabled the formation of a coating (PEGMA/GMA) that was cross-linked as well as covalently anchored to the underlying ALAPP coating. The epoxy functional groups in the GMA polymer component served as reactive anchors for the subsequent immobilization of the printed factors on the microarray surface, whereas the function of the PEGMA polymer component in the copolymer was to reduce cell attachment between spots (Fig. 1). Diol functionalities generated upon epoxy ring opening due to the reaction with water also contributed to a low cell attachment background. These coatings proved to be robust for the duration of the experiments and to be compatible with the tissue culture, immunohistochemistry, and microscopy experiments carried out in this study.

thumbnail image

Figure 1. Schematic representation of surface modification of the microarray substrates. A robotic microarrayer was used to dispense nanoliter volumes of biological and synthetic factors. The GMA/PEGMA copolymer coatings allowed the covalent attachment of biomolecules (through epoxy functional groups) while also providing a nonfouling background to inhibit nonspecific cell attachment (through PEG modification). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Selection of Microarray Printing Factors

To determine appropriate factors for efficient differential binding, the factors must respond in two ways: (1) bind cells in high numbers and (2) discriminate between various subpopulations within a mixture. We printed a range of biological factors with potential germ cell-surface recognition properties. These include a range ECM proteins, cell-surface active growth factors, lectins, and antibodies against cell surface proteins, some of which have been shown to surface-label cells in the testis of mammalian species (26, 27, 36–38). We further evaluated some synthetic macromolecules with amine functional groups, which have been shown previously to influence cell-surface interactions (39, 40). The microarrays were engineered to screen over 26 different factors at three different concentrations (25, 100, and 200 μg/ml). Preliminary work demonstrates very low cell binding below 50 μg/ml printing concentration of most selected factors and greater cell binding at higher concentrations. Optimum cell binding occurs around 200 μg/ml printing concentration and can vary depending on the specific factor. At higher printing concentrations (>200 μg/ml), we did not observe significant increases in cell binding, which we attribute to saturation of immobilized factor coverage. The print layout used in this study is summarized in Figure 2.

thumbnail image

Figure 2. Microarray printing layout. (a) Legend and arrangement of factors presented on the surface of the array. (b) A standard print layout representing three different factor concentrations. (c) Fluorescence microscope image of Cy3 locator dye observed on the surface of the array. (d) Cell Tracker™ Orange stained cells observed on a Coll IV spot (200 μg/ml) of the microarray. (e) Key to the printed factors. Scale bars are 100 μm.

Download figure to PowerPoint

Factors Supporting High-Cell Attachment

We first screened for factors that support a high degree of cell attachment from an initial testis tissue isolate (50 g testis). Gonocytes were not expected to be present at that stage of testis development. Very low and inconsistent cell attachment was observed at the 25 μg/ml concentration of printed factors on the microarrays and consequently spots printed at this concentration were not included for scoring. At higher concentrations, we generally observed more consistent cell attachment. More specifically, at 200 μg/ml concentration of printed factors, we observed the highest cell attachment (∼100+ cells per arrayed spot) for PSA, DSA, and ConA on the arrays (Fig. 3a). Other factors that demonstrated moderate attachment (∼50–80 cells per arrayed spot) included DBA, Coll I, and CD49f. The remaining factors that showed a response greater than background include PNA, FBR, LMN, Coll IV, VTR, and CD9. Many of these factors, when printed at lower concentrations (100 μg/ml and 25 μg/ml), showed no cell attachment above background. We observed some cell attachment for PSA, DBA, ConA, LMN, Coll I, Coll IV, and DSA (Fig. 3b) printed at 100 μg/ml.

thumbnail image

Figure 3. Total cell attachment (averaged per spot) determined by a cell count after hematoxylin staining of the arrays at two different printing concentrations. With (a) spots printed at 200 μg/ml and (b) spots printed at 100 μg/ml after incubation with the bovine testis tissue isolate. Error bars correspond to standard error measurement (n = 5). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Differential Cell Attachment on the Microarrays

Although a general CellTracker stain and hematoxylin was beneficial to observe and quantify overall cell binding on the arrayed substrates (Fig. 1e), immunohistochemistry allowed us to discriminate between the different cell types and the factors they preferentially bind to. A Plzf antibody was used to determine the fraction of Plzf+ cells on the microarrays (Figs. 4a–4f). Plzf is a nuclear transcription factor present in type A spermatogonia in several mammalian species including mouse (41), rat (42), human, monkeys (43), pig (38), and bovine (Supporting Information Fig. 1). A vimentin antibody was used to identify Sertoli and myoid cell populations on the arrays (Figs. 4g–4l). Vimentin is a commonly used perinuclear marker for bovine Sertoli cells and is also used to identify myoid and fibroblastic cells (22, 34, 35, 41, 44). Interactions of the cells with the various printed factors were evaluated on the basis of (1) total cell attachment (hematoxylin staining), (2) vimentin staining (vimentin+), and (3) Plzf staining (Plzf+). We observed that the highest proportion of Plzf+ cells was attached to arrayed DBA (Fig. 5a). DBA lectin is a surface-specific marker for type A spermatogonia in bovine. It has been used previously for immunohistochemistry (24, 44–46) and for spermatogonia enrichment with MACS and FACS but resulted in poor cell recovery (28). Plzf+ cells on the arrayed DBA microarrays were ∼20% of total adhered cells. In comparison with a standard smear of the initial isolate (which stained only 3% Plzf+), arrayed DBA lectin yielded an enrichment by a factor of 6.7. No enrichment for Plzf+ cells was observed with any of the other printed factors, although some attachment of Plzf+ cells (<5%) was observed for arrayed DSA and CD49f. CD49f has been used to purify murine SSC's with FACS analysis (26, 45).

thumbnail image

Figure 4. Representative light microscopy images for Plzf and vimentin staining on various printed factors after 2-h incubation on the arrays. The images show Plzf staining (in brown) on (a) DBA, (b) PSA, (c) VTR, (d) CD49f, (e) DSA, and (f) Coll IV. The cells were also counterstained with hematoxylin (in blue). The images of (g) DBA, (h) PSA, (i) VTR, (j) CD49f, (k) DSA, and (l) Coll IV were observed after staining with vimentin. The scale bar represents 200 μm.

Download figure to PowerPoint

thumbnail image

Figure 5. Ratio of Plzf+ and vimentin+ cells to total cells (as determined by hematoxylin staining) on spots printed at a concentration of 200 μg/ml after different stages of enrichment. Arrays were stained with vimentin to quantify the fraction of attached Sertoli and myoid cells, whereas Plzf was used to identify attached type A spermatogonia. With (a) fraction of Plzf+ cells attached after incubation with the tissue isolate, (b) combined vimentin+ (Sertoli and myoid cell) attachment after incubation with the tissue isolate. Error bars represent standard error measurement (n = 5), whereas all factors in (a, b) represent significant difference in cell binding in comparison to DBA (P < 0.001). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Identifying factors that support selective attachment of either Sertoli or myoid cells would be beneficial for development of tissue-culture materials that may be conducive to a negative selection strategy, removing unwanted cells from a heterogeneous cell population. Harvesting Sertoli cells is also of interest in terms of the production of laminins and other basement membrane components (47). We compared the number of vimentin+ cells to the total number of adherent cells per spot (Fig. 5b). Staining of the smears of the starting-cell population showed 69% of cells were vimentin+. On the microarrays, DBA showed <25% of vimentin+ cells, the lowest percentage observed. The factors we identified as selective and efficient for cell attachment for Sertoli and myoid cells were PSA and ConA. These factors showed high overall cell attachment (at the 200 μg/ml printing concentration) and <15% of cells that stained negative for vimentin (Fig. 5). Moreover, <5% of the total cells was Plzf+. Of the antibodies, we observed CD9 to be highly selective for vimentin+ cells though overall cell binding efficiency was low (<40 cells/spot), and CD49f showed no differential attachment of vimentin+ cells in comparison with the smears. In terms of isolating Plzf+ cells from tissue, it is within this vimentin negative population that these cells would be located, and so we propose that factors with high vimentin+ cell attachment would yield ideal materials for a negative selection strategy for purifying Plzf+ cells. As for the synthetic materials screened, the PAMAM dendrimers (D2 and D3), and polymers PAL and PEI, no significant attachment was observed.

Last, we also screened for differential cell attachment with an enriched fraction of bovine testis isolated cells to demonstrate that the identified factors can also be used for the capture and enrichment of cultured cells. Differential plating methods are commonly used techniques for enrichment and isolation of various germ cells (27, 28). The fraction used here had been enriched in spermatogonia by plating of the initial isolate on BSA-coated flasks (2 h) and subsequent plating onto gelatin coated flasks (16 h), thereby removing a large population of Sertoli and myoid cells from the heterogeneous population (Supporting Information). Immunohistochemistry of smears from cells obtained by this procedure yielded 12% Plzf+ cells. Upon incubation of the enriched fraction on the microarrays, we observed an increase to ∼37% Plzf+ on the DBA spots (Supporting Information Fig. 2).

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. CONCLUSIONS
  6. Acknowledgements
  7. LITERATURE CITED
  8. Supporting Information

We have designed and fabricated cell microarrays on which covalently anchored biological factors including antibodies, lectins, growth factors, ECM proteins, and synthetic dendrimers, and polymers were screened for differential selective cell attachment. This platform can potentially be used for screening cell surface markers or isolation platforms for virtually any heterogeneous population. In this study, we use cells derived from bovine testes as a model. The microarray approach confirmed that amongst the biological factors screened for Plzf+ cell enrichment, DBA achieved the greatest degree of differential cell attachment. Here, 20% of total cell attachment was attributed to Plzf+ cells after incubation on the microarray, whereas other lectin factors such as PSA and ConA were identified as possible candidates for Plzf+ cell purification by negative selection, because Sertoli and myoid cells attached to these factors in high numbers (>85% of total cell attachment). Sertoli and myoid cells also bound to these factors in high numbers with <5% of the total cells attached being Plzf+. For the antibodies as well as the synthetic factors screened, overall cell attachment efficiency or selectivity was not sufficient for these factors to be considered as candidates for positive or negative selection factors. In this study, we have shown utility of a microarray platform as a screening tool for the identification of factors that allow the isolation of rare and/or specified cells from heterogeneous cell populations. The surface chemistry based on the covalent immobilization of factors used on the microarray substrates facilitates the translation to other substrate materials such as tissue culture ware including multiwell plates, flasks, and microparticles. Based on this, new methods of preparative purification may also be established that provide enhancement over already existing enrichment and isolation methods.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. CONCLUSIONS
  6. Acknowledgements
  7. LITERATURE CITED
  8. Supporting Information

The authors gratefully acknowledge Peter Macardle, Sigrid Lehnert, and Frances Harding for helpful discussions. We also thank Brendan Hatton and Andrew Eichorn for assistance with animal management.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. CONCLUSIONS
  6. Acknowledgements
  7. LITERATURE CITED
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS AND DISCUSSION
  5. CONCLUSIONS
  6. Acknowledgements
  7. LITERATURE CITED
  8. Supporting Information

Additional supporting information may be found in the online version of this article.

FilenameFormatSizeDescription
CYTO_20913_sm_suppinfo.doc27KSupporting Information
CYTO_20913_sm_suppinfofigures.doc3344KHistological sections of ∼ 60g bovine testis at prepubertal stage. The sections have been stained with an antibody to Plzf and counterstained with hematoxylin. Plzf positive cells are located on the tubular basal lamina and morphologically resemble type A spermatogonia. Scale bar = 50 μm. Supplementary Figure 2. Ratio of Plzf+ cells to total cells on arrayed spots at a concentration of 200 μg/mL after overnight enrichment plating procedure. Arrays were stained with Plzf and hematoxylin. Error bars represent standard error measurement (n=5) while all factors represent significant difference in cell binding in comparison to DBA (P <0.001).

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.