SEARCH

SEARCH BY CITATION

Keywords:

  • Laminins;
  • Mouse embryonic stem cells;
  • Self-renewal;
  • Extracellular matrix;
  • Adhesion;
  • Proliferation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References
  11. Supporting Information

We tested specific laminin (LN) isoforms for their ability to serve as substrata for maintaining mouse embryonic stem (ES) cells pluripotent in vitro in the absence of leukemia inhibitory factor or any other differentiation inhibitors or feeder cells. Recombinant human LN-511 alone was sufficient to enable self-renewal of mouse ES cells for up to 169 days (31 passages). Cells cultured on LN-511 maintained expression of pluripotency markers, such as Oct4, Sox2, Tert, UTF1, and Nanog, during the entire period, and cells cultured for 95 days (17 passages) were used to generate chimeric mice. LN-332 enabled ES cells proliferation but not pluripotency. In contrast, under the same conditions LN-111, Matrigel, and gelatin caused rapid differentiation, whereas LN-411 and poly-d-lysine did not support survival. ES cells formed a thin monolayer on LN-511 that differed strikingly from typical dense cluster ES cell morphology. However, expression of pluripotency markers was not affected by morphological changes. The effect was achieved at low ES cell density (<200 cell/mm2). The ability of LN-511 and LN-332 to support ES cell proliferation correlated with increased cell contact area with those adhesive substrata. ES cells interacted with LN-511 via β1-integrins, mostly α6β1 and αVβ1. This is the first demonstration that certain extracellular matrix molecules can support ES cell self-renewal in the absence of differentiation inhibitors and at low cell density. The results suggest that recombinant laminin isoforms can provide a basis for defined surface coating systems for feeder-free maintenance of undifferentiated mammalian ES cells in vitro.

Disclosure of potential conflicts of interest is found at the end of this article.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References
  11. Supporting Information

Author contributions: A.D. and S.R.: conception and design, provision of material, collection and/or assembly of data, data analysis and interpretation, manuscript writing; A.D. and S.R. contributed equally to this work. A.B.: provision of material; K.T.: conception and design, financial support, administrative support, data analysis and interpretation, manuscript writing, final approval of manuscript

There is a need to develop defined feeder cell-free in vitro culture systems for establishment and maintenance of undifferentiated mammalian embryonic stem cells. A major problem of embryonic stem (ES) cell cultures is the lack of appropriate surface coatings, particularly regarding human ES cells. Although defined xeno-free ES cell culture media have successfully been developed for human ES cells [1], researchers are still looking for defined xeno-free nonimmunogenic culture plate coatings that do not induce cellular differentiation [2]. Adhesive surface coatings are usually based on various combinations of extracellular matrix (ECM) proteins, such as laminin-111, collagen IV, gelatin, fibronectin, or Matrigel (BD Biosciences, San Diego, http://www.bdbiosciences.com), most of which are undefined or not human [3]. For instance, coatings used successfully, such as Matrigel [1, 4, 5] prepared from the mouse Engelbreth-Holm-Swarm (EHS) sarcoma and ECM derived from mouse embryonic fibroblasts [6], are of undefined nonreproducible composition and are derived from animal sources. Therefore, they are not applicable for clinical purposes.

Laminins (LN) that are central components of basement membranes (BM) are the first extracellular matrix molecules to contact cells in the early embryo [7, [8], [9], [10]11]. Expression of laminin chains has been shown to occur as early as the two- to four-cell stage in mouse embryos [7, 8]. One of the laminins, laminin-511, contacts the inner cell mass of blastocysts, which is the natural origin of ES cells [11]. Laminins have been shown to influence cellular differentiation and migration, in addition to promoting adhesion and proliferation [10, 12, 13], whereas some other ECM molecules, such as collagens, primarily provide adhesion and structural support functions [14]. Thus, laminins may be useful for culturing ES cells in vitro, as they are a natural part of the niche for their origin in vivo.

Laminins are a large family of heterotrimeric macromolecules consisting of α, β, and γ chains [13]. There exist five α chains, three β chains, and three γ chains in mice and humans. Thus far, 15 different combinations have been identified in mammalian tissues [15]. The different isoforms are developmentally regulated and have tissue-specific locations and functions. LN-111 (previously named laminin-1 or laminin) is present in the early embryo and later in certain epithelial cells and murine EHS sarcoma, but otherwise it is a rare isoform in vivo [10]. LN-511 (LN-10) is the most common form found in BMs of the early embryo and most adult tissues [9, 16]. Importantly, it is found in the extracellular matrix between cells of the inner cell mass of blastocysts [11]. Laminins are cell type-specific mediators regulating cell adhesion, proliferation, migration, and resistance to apoptosis [17]. Mutations in most laminin chains result in severe pathologies and mortality [17].

Most laminin isoforms, except for laminin-111, are difficult to extract and purify in native forms from tissues because of extensive cross-linking with other laminins or other macromolecules. Only recently, human/mouse hybrid LN-111 (LN-1) [18] and human LN-211 (LN-2) [19], LN-332 (LN-5) (specific for epithelial BMs) [20], LN-411 (LN-8) [21] (common in vascular BMs), and the ubiquitous LN-511 (LN-10) [22] have been successfully produced as recombinant proteins. It is also possible to isolate some laminin isoforms, such as LN-332 (LN-5), from cultured cells, but only in low quantities [23, 24].

The aim of this study was to determine whether certain laminin isoforms can either act as differentiation inductors or sustain ES cell self-renewal. The results revealed that various laminin isoforms cause diverse effects on ES cells: LN-511 enabled mouse ES cell self-renewal for more than 5 months in the absence of any differentiation inhibitors; LN-332 enabled ES cell proliferation but not pluripotency, unlike LN-111, which caused ES cell differentiation within 2 weeks; and LN-411 was not capable of supporting ES cell survival.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References
  11. Supporting Information

Laminin Nomenclature

We used the nomenclature for laminins described by Aumailley et al. [15].

Cell Culture

Mouse embryonic stem cells (line GSI-1, derived from 129SvJ mice, and line RW4) were cultured on ECM coatings in medium containing 80% Dulbecco's modified Eagle's medium, supplemented with GlutaMax I and 4.5 g/l glucose, 20% ES-qualified fetal serum, 0.5% penicillin, 0.5% streptomycin, 10 mM Hepes buffer, 1 mM sodium pyruvate, 1% nonessential amino acids (all provided by Invitrogen, Carlsbad, CA, http://www.invitrogen.com), 0.1 mM β-mercaptoethanol (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com), and 10 ng/ml basic-fibroblast-growth factor (Millipore, Billerica, MA, http://www.millipore.com) at 37°C, 5% CO2. ES cells were plated upon ECM coatings at an initial density of 300 cells per mm2. Cells were split once in 4–6 days with 0.05% trypsin-EDTA solution and were plated at cell density of 180 cells per mm2. ES cells were cultured as two independent controls on each coating. Cells were counted after each passage using a hematocytometer.

Ninety-six-well tissue cell culture plates were coated overnight at 4°C with sterile solutions of the ECM proteins, such as mouse LN-111 (Invitrogen), human recombinant LN-332 [20], human recombinant LN-411 [21], and human recombinant LN-511 [22], all at a concentration of 30 μg/ml (5 μg/cm2); growth factor-reduced Matrigel, diluted 1:30 (BD Biosciences); 1 mg/ml bovine gelatin (Sigma-Aldrich); and 0.1 mg/ml poly-d-lysine (Sigma-Aldrich).

Antibody

To study pluripotency or differentiation status of ES cells, primary antibodies against the following antigens were used: Sox2, Nanog, undifferentiated embryonic cell transcription factor (UTF1), collagen IV (all from Millipore), and Oct4 (Oct3/4; BD Biosciences). To study integrin receptor expression and involvement in interaction with LN-511 in mouse ES cells, primary antibodies against the following subunits were used: α2, α2β1, α3, α4, α5β1, α6, αV, αVβ6, β1, β2, and β4 from Chemicon; β1 from R&D Systems Inc. (Minneapolis, http://www.rndsystems.com); α3, β1, β1 (function-blocking), and β3 (function-blocking) from BD Biosciences; and αV, β1, and β3 (all function-blocking) from BioLegend (San Diego, http://www.biolegend.com). Immunoglobulins derived from the respective species (Chemicon; BioLegend) were used as negative control. Secondary antibodies for immunofluorescence staining (anti-mouse, anti-rat, anti-goat, and anti-rabbit; Alexa-350-, Alexa-488-, and Alexa-546-labeled) were from Molecular Probes (Eugene, OR, http://probes.invitrogen.com). Horseradish peroxidase (HRP)-conjugated anti-mouse and anti-rabbit secondary antibodies used for Western blots were from GE Healthcare (Little Chalfont, U.K., http://www.gehealthcare.com).

Immunofluorescence

For immunofluorescence, ES cells were fixed in the wells of a 96-well plate with 4% paraformaldehyde, permeabilized with 0.1% Triton X, and blocked with 10% bovine fetal serum (Invitrogen) in phosphate-buffered saline (PBS) containing 0.1% Tween-20 (Sigma-Aldrich) for 1 hour. Incubation with primary antibody was performed for 1.5 hours at room temperature. Incubation with secondary antibody and 4,6-diamidino-2-phenylindole (Molecular Probes) was performed for 40 minutes. Between incubations, specimens were washed with 0.1% Tween-20 in PBS buffer three to five times, preserved in fluorescence mounting medium (Dako, Glostrup, Denmark, http://www.dako.com), and observed under a fluorescence microscope (Leica, Heerbrugg, Switzerland, http://www.leica.com).

Reverse Transcription-Polymerase Chain Reaction

Total RNA was isolated using Absolutely RNA Microprep Kit (Stratagene, La Jolla, CA, http://www.stratagene.com) according to the manufacturer's instructions. cDNA was synthesized using 0.2 μg of total RNA in 20 μl of reaction mixture containing oligo(dT)12–18 primers and Superscript II reverse transcriptase (Invitrogen), according to the manufacturer's instructions. To compensate for variable cDNA yields, the amount of cDNA for each polymerase chain reaction (PCR) was calibrated by using expression level of the housekeeping gene of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a standard. Amounts of cDNA yielding an equivalent amount of GAPDH PCR product (at 20 cycles; data not shown) were used for subsequent PCRs. cDNAs were amplified using the primers shown in supplemental online Table 1. All PCRs were run for 30 cycles (including those GAPDH PCRs that are shown in Fig. 1D) and were performed in 20 μl under standard conditions using 1 U of Taq DNA Polymerase Recombinant (Invitrogen). The PCR products were analyzed on a 1.5% agarose gel containing ethidium bromide. For each RNA sample, reverse transcription (RT)-PCR without reverse transcriptase was performed to confirm that no genomic DNA were isolated.

Western Blot and Densitometry Analysis

After being cultured on different ECM coatings, as described above, mouse ES cells were collected, counted and pelleted by centrifugation, mixed with nonreduced SDS-polyacrylamide gel electrophoresis sample buffer to equal concentrations of 2,000 cells per microliter, and sonicated five times for 15 seconds each time. Gradient (4%–12%) gels were used for SDS electrophoresis, and the proteins were transferred to polyvinylidene difluoride membranes. Membranes were blocked with 5% milk solution in PBS-0.1% Tween buffer for 2 hours. Primary antibodies against Oct-3/4 (1:300) and Sox2 (1:1,000) in 5% milk solution in PBS-0.1% Tween buffer were incubated with the membranes overnight at +4°C. After being washed four times, HRP-conjugated secondary antibodies in a 5% milk solution in PBS-0.1% Tween buffer (dilution, 1:1,000) were incubated with the membranes for 40 minutes at room temperature and washed five times with PBS. Chemiluminescent HRP substrate from Amersham Biosciences (Piscataway, NJ, http://www.amersham.com) was used for visualization. Films were scanned at 600 dpi and analyzed with the ChemiImager5500 program (1D-Multi Line densitometry mode, Alpha Innotech, San Leandro, CA, http://www.alphainnotech.com). Mouse ES cells cultured in the presence of leukemia inhibitory factor (LIF) were used as a positive control. Error bars show Figures 1E, 1F.

Chimeric Mice

After being cultured for 95 days (17 passages) on laminin-332 and laminin-511, ES cells of GSI-1 line were injected into C57Bl mice blastocysts (procedure was performed in the Karolinska Center for Transgene Technologies, Karolinska Institute, Stockholm, Sweden). ES cells of the RW4 line were injected after 45–50 days (11–15 passages) in same way. Ethical permission was obtained from the local ethics committee for experimental animal research.

Cell Adhesion Assay

Attachment assay was performed as described in [25, 26]. Briefly, MaxiSorp 96-well plates (Nunc, Rochester, NY, http://www.nuncbrand.com) were coated with ECM proteins as described above and blocked with 1% heat-denatured bovine serum albumin (BSA) solution. Undifferentiated ES cells were plated at a cell density of 800 cells per mm2 on ECM-coated plates and were left to adhere for 1 hour at 37°C. Nonadherent cells were washed away, and adherent cells were fixed for 20 minutes with 5% glutaraldehyde, washed, and stained with 0.1% Crystal Violet (Kebo Lab, Spanga, Sweden, http://www.kebolab.se). After 1 hour Crystal Violet was extracted from cells with 10% acetic acid and quantified by measuring optical density at 570 nm. Error bars show standard deviation (SD) (n = 3).

Cell Contact Area Measurement for Adhesion/Blocking Experiments

For measuring cell contact area, undifferentiated ES cells were plated (150 cells per mm2) and then treated, fixed, and stained as described above. Photos of 10–20 random fields were taken, and the cell contact area of 40–150 cells was measured using the Volocity imaging software (Improvision, Waltham, MA, http://www.improvision.com). Error bars show standard error of the mean (SEM).

Adhesion-Blocking Assay Using Anti-Integrin Antibody

Adhesion-blocking assays were performed as described previously [27]. Briefly, plates were coated with LN-511 and blocked with 1% heat-denatured BSA solution. ES cell suspension was incubated with function-blocking anti-integrin antibodies (concentration as recommended by supplier) for 30 minutes, plated on LN-511-coated plates, and allowed to adhere for 1 hour at 37°C. Nonattached cells were removed, and the remaining cells were fixed, stained, and quantified as described above. Error bars show SEM.

Cell Attachment to Surface Coated by Anti-Integrin Antibody Assay

The assay was designed to identify integrin receptors that are expressed in sufficient amounts to retain cells attached to the surface coated with anti-integrin-specific antibody. MaxiSorp 96-well plates (Nunc) were coated with purified anti-integrin antibodies at a concentration of 10 μg/ml at +4°C overnight and later washed and blocked with 1% heat-denatured BSA solution. ES cells were plated on antibody-coated plates and allowed to adhere for 1 hour at 37°C. Nonattached cells were removed, and the remaining cells were fixed, stained, and quantified as described above. Error bars show SEM.

Affymetrix Array

RNA was extracted from undifferentiated ES cells as described above and frozen to −80°C. Affymetrix Mouse Expression Array MOE 430 2.0 (Affymetrix, Santa Clara, CA, http://www.affymetrix.com) was performed by the core facility of the Karolinska Institute.

Sequence Similarity Analysis

Comparison of mouse and human laminin chain sequence similarity was performed using the BLAST tool (http://blast.ncbi.nlm.nih.gov/Blast.cgi).

Statistics

Statistical significance was determined by Student's two-tailed t test for unequal variances.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References
  11. Supporting Information

To explore whether specific laminin isoforms can either trigger differentiation or sustain self renewal, we cultured mouse ES cells in vitro without any feeder cells, differentiation inhibitors, or differentiation inductors. We produced and purified recombinant human LN-332, LN-411, and LN-511 from culture media of HEK293 cells as described previously [20, [21]22], and we obtained commercially available mouse LN-111 (Invitrogen) isolated from the EHS sarcoma [28, 29]. In addition, Matrigel, gelatin, and poly-d-lysine were used as substrata.

As a positive control for undifferentiated mouse ES cells in our experiments, we used ES cells cultured in the presence of the differentiation inhibitor LIF [5, 30, 31] on gelatin. This is the best-documented conventional system for feeder-free culture of mouse ES cells, and it has been used successfully for decades [5, 30, 31].

LN-511 Enables Mouse ES Cell Self-Renewal in the Absence of Differentiation Inhibitors for at Least 169 Days

As shown in Figure 1A and 1B, mouse ES cells cultured on LN-511 proliferated at a stable rate in the absence of feeder cells or LIF for at least 169 days (31 passages), the average doubling time being 1.2 days. Importantly, immunofluorescence, RT-PCR, and quantitative Western blot analyses revealed that the ES cells expressed pluripotency markers [32], such as Sox2, Oct4, Tert, UTF1, and Nanog (Fig. 1C–1F; supplemental online Fig. 1, 2; the list of primers used for RT-PCR is shown in supplemental online Table 1), during the entire experiment to the same extent as control pluripotent ES cells cultured in the presence of LIF. Expression of differentiation markers, such as α-fetoprotein, brachyury, nestin, and vimentin, was low or nonexistent throughout the experiment (Fig. 1D). Levels of pluripotency markers Oct4 and Sox2 were measured using Western blot and quantified by densitometry (Fig. 1E, 1F). ES cells cultured on LN-511 for more than 3 months maintained levels of those markers at the same level as the control (105% and 94%, respectively, compared with positive control ES cells cultured in the presence of LIF). To verify that the ES cells cultured on LN-511 were pluripotent, cells maintained for 95 days (17 passages) were injected into mouse blastocysts that were subsequently implanted into pseudopregnant mice. This led to the generation of chimeric mice (Fig. 2A, 2B) demonstrating that the cells were indeed pluripotent. Two different mouse ES cell lines (GSI-1 and RW4) were used to generate chimeric mice. The percentage of chimeras among the progeny and degree of chimerism for both lines did not differ from those of positive control ES cells cultured in the presence of LIF (Table 1).

thumbnail image

Figure Figure 1.. LN-511 supports mouse embryonic stem (ES) cell self-renewal for more than 5 months in the absence of differentiation inhibitors or feeder cells. (A): Proliferation of mouse ES cells on LN, MG, GEL, and PL. Mouse ES cells expanded on LN-511 and LN-332 for up to at least 169 days, during which approximately 150 doublings in cell number occurred. In contrast, the cells did not proliferate on LN-111, LN-411, MG, GEL, or PL. (B): Magnification of the first 20 days in (A), showing that the cells proliferated to a certain limit on LN-111, LN-411, MG, and GEL, but proliferation ceased within 1–2 weeks. The cells attached poorly to PL. (C): ES cells cultured on LN-511 for 169 days in the absence of feeder cells or LIF continuously expressed Oct4 (green) and Sox2 (red) at same level as Ctrl pluripotent ES cells cultured in the presence of LIF, as determined by immunofluorescent staining. Cells cultured on LN-322 expressed Oct4 and Sox2 to a lesser extent. However, expression of those pluripotency markers was distinctly reduced in ES cells cultured on LN-111 or MG for 11 days. Magnification, ×40. Scale bars = 27 μm. (D): Reverse transcription-polymerase chain reaction analyses revealed that pluripotent mouse ES cells grown on LN-332 or LN-511 for 14 or 145 days maintained expression of pluripotency markers Sox2 and Oct4 but exhibited only low expression of differentiation markers, such as brachyury and α-fetoprotein. Ctrl pluripotent ES cells grown in the presence of LIF exhibited a similar expression profile. In contrast, cells grown for 14 days on LN-111, MG, or GEL started to express brachyury and α-fetoprotein, signs of differentiation. Cells grown on LN-411 did not differentiate, but they did not properly adhere or proliferate. The cells survived only a few days on PL, but they had already differentiated during that time. (E): Expression of pluripotency marker Oct4 in mouse ES cells cultured on LN-511 (more than 5 months), LN-332 (more than 5 months), LN-111 (less than 20 days), and MG (less than 20 days), as measured by Western blot and quantified by densitometry. Expression was compared with that in positive Ctrl ES cells cultured with LIF. Error bars represent range. (F): Expression of pluripotency marker Sox2 in mouse ES cells cultured on LN-511 (more than 5 months), LN-332 (more than 5 months), LN-111 (less than 20 days), and MG (less than 20 days), as measured by Western blot and quantified by densitometry. Expression was compared with that of positive Ctrl ES cells cultured with LIF. Error bars represent range. Abbreviations: Ctrl, control; DAPI, 4,6-diamidino-2-phenylindole; GEL, gelatin; LIF, leukemia inhibitory factor; LN, laminin; MG, Matrigel; PL, poly-d-lysine; T, time.

Download figure to PowerPoint

thumbnail image

Figure Figure 2.. Chimeric mice generated from ES cells cultured on laminin (LN)-511 in the absence of differentiation inhibitors. (A): Chimeric mice generated from mouse embryonic stem (ES) cell line RW4 cultured for 45–50 days (11–15 passages) on LN-511. (B): Chimeric mice generated from mouse ES cell line GSI-1 cultured for 95 days (17 passages) on LN-511.

Download figure to PowerPoint

Table Table 1.. Efficiency of chimera formation of mouse embryonic stem cell lines cultured on different coatings
  1. Abbreviations: LIF, leukemia inhibitory factor; LN, laminin; NA, data not available.

Thumbnail image of

Mouse ES Cells Cultured on LN-332 Proliferate but Do Not Maintain Pluripotency

Mouse ES cells cultured on LN-332 proliferated for 169 days with a high rate similar to that on LN-511 (Fig. 1A, 1B) and expressed pluripotency markers (Fig. 1C–1E). However, quantification of Western blot revealed that Oct4 and Sox2 levels declined to 23% and 46%, respectively, in comparison with the positive control. Notably, mouse ES cells cultured on LN-332 either formed weak chimeras (line GSI-1) or could not form any chimeras at all (line RW4). The results revealed a distinct difference between the effects of LN-511 and LN-332 on mouse ES cells. Apparently LN-332 could support cell proliferation but not mouse ES cell self-renewal.

Mouse ES Cells Cultured on LN-111 or Matrigel Undergo Differentiation Within 2 Weeks, Whereas LN-411 Does Not Support ES Cell Survival

In contrast to LN-511 and LN-332, cells cultured on LN-111 or Matrigel did not proliferate or self-renew in the absence of differentiation inhibitors. Within 4 days proliferation ceased in the absence of LIF (Fig. 1B), and after 11 days the cells formed cobblestone-like structures and started to express differentiation markers, such as collagen IV (Fig. 3D) and brachyury (Fig. 1D), but decreased expression of pluripotency markers Sox2, Oct4, Nanog, and UTF1 (Figs. 1C–1F, 3D; supplemental online Figs. 1, 2). Although proliferation had ceased, the differentiated cells remained viable for at least 25–30, days retaining cobblestone-like morphology. The ES cells did not survive on LN-411 or poly-d-lysine, because of low adhesion. When cultured on gelatin, the cells underwent spontaneous differentiation (Fig. 1D). Differentiation markers, such as α-fetoprotein and brachyury, were strongly expressed, whereas nestin and vimentin exhibited lower expression levels. All ECM coatings considered above, including laminin-111 and Matrigel, were not able to sustain ES cell self-renewal or proliferation in the absence of differentiation inhibitors.

thumbnail image

Figure Figure 3.. Contact with LN-511 enables embryonic stem (ES) cell pluripotency even at low cell density and in the absence of differentiation inhibitors, and it results in monolayer morphology different from conventional cluster morphology. (A): Expression of pluripotency markers in mouse ES cells plated at a low density of 180 cells per mm2 on LN-511 at time points of 1, 3, 24, 48, 72, and 96 h after passaging. The cells adhered and proliferated, and they were studied for expression of Sox2 (red) and Oct4 (green) and the presence of DNA (DAPI; blue) at different time points. All cells expressed pluripotency markers, and in most cases the cells did coexpress Oct4 and Sox2 (yellow), as shown by merged images. Magnification, ×40. Scale bars = 27 μm. (B): Morphology of ES cells on LN-511: immunofluorescent staining for Oct4 (blue), integrin β1 (green), and f-actin (red) versus phase contrast (Nomarsky) (low magnification). The spread monolayer morphology of ES cells cultured on LN-511 differed strikingly from the conventional dense cluster morphology of ES cells cultured in the presence of LIF on gelatin. The unusual morphology did not affect Oct4 expression in ES cells. Magnification, ×10. Scale bars = 500 μm. (C): Morphology of ES cells on LN-511: immunofluorescent staining and phase contrast (Nomarsky) (high magnification). ES cells spread as monolayer on LN-511, so they are hardly visible in phase contrast, unlike the clearly visible ES cell clusters grown in the presence of LIF. Magnification, ×40. Scale bars = 100 μm. (D): Immunofluorescent analysis of expression of collagen IV (red) and Oct4 (green) and the presence of DNA (DAPI, blue) in ES cells cultured for on LN-511 for 169 days and on LN-111 for 11 days in the absence of differentiation inhibitors. ES cells cultured in the presence of LIF on gelatin served as a positive control. Undifferentiated ES cells positive for Oct4 (LN-511 and control) did not express collagen IV, but cells differentiated on LN-111 were negative for Oct4 and strongly positive for collagen IV. Magnification, ×40. Scale bars = 27 μm. Abbreviations: Col, collagen; DAPI, 4,6-diamidino-2-phenylindole; h, hours; LIF, leukemia inhibitory factor; LN, laminin.

Download figure to PowerPoint

Contact with LN-511 Is Sufficient to Maintain Pluripotency of Mouse ES Cells

To study whether contact with LN-511 is indeed the only factor sufficient for ES cell self-renewal, we monitored ES cell status for several days after replating them, as usual, at low cell density. LN-511 supported pluripotency of undifferentiated mouse ES cells even under low cell density conditions, where the cells lacked contact with other cells, soluble differentiation inhibitors, or any other ECM proteins, for at least 2 days (Fig. 3A). Thus, expression of the pluripotency markers Sox2 and Oct4 remained stable, and the cells proliferated rapidly and spread upon LN-511. Therefore, the results indicate that LN-511 is indeed a factor that can transmit signals necessary and sufficient for sustaining self-renewal and promotion of rapid proliferation. Notably, ES cells, when cultured on other coatings, need not only soluble differentiation inhibitors, such as LIF, but also constant, multiple-cell contact with neighbor cells for pluripotency maintenance [33].

Strong Adhesion to LN-511 Results in Spread Monolayer Morphology of Undifferentiated ES Cells

Interestingly, proliferation of the mouse ES cells promoted by certain laminin isoforms correlated with the adhesion to those isoforms (LN-511 and LN-332) (Fig. 4A, 4B). The average contact area of an adherent mouse ES cell grown on LN-511 was approximately 2.5 times higher than that of cells plated on nonspecific coating, such as poly-d-lysine (Fig. 4A, 4C). ES cell spreading on all other coatings, including LN-111, Matrigel, and gelatin, was significantly less than that on LN-511 or LN-332 (statistical significance, p < .001; Student's two-tailed t test).

thumbnail image

Figure Figure 4.. Adhesion of mouse embryonic stem (ES) cells to different coatings. (A): Crystal Violet staining of ES cells adherent to different LN (LN-511, -322, -411, and -111). Magnification, ×5 (insets, ×20). (B): Adhesion of ES cells to different coating surfaces: LN, MG, GEL, and PL. Values are shown as average percentages of cells that attached. Error bars show SD (n = 3). Statistical significance calculated by the Student t test is shown: *, p < .05. (C): Contact area of ES cells with different adhesive substrata. Values are shown as average relative cell contact area (compared with nonspread ES cells; percentage). Error bars show SEM; number inside each bar shows number of independent measurements (n = 40–179). Statistical significance calculated by the Student t test is shown: ***, p < .001. Abbreviations: GEL, gelatin; LN, laminin; MG, Matrigel; PL, poly-d-lysine.

Download figure to PowerPoint

The morphology of mouse ES cells cultured on LN-511 differed significantly from that of cells cultured in a conventional system (i.e., in the presence of LIF or on mouse fibroblast feeders or gelatin) (Fig. 3B, 3C). In the latter case, ES cells form typical dense clusters with sharp, defined borders. In contrast, mouse ES cells cultured on LN-511 first formed a monolayer and started to form multilayers only at high cell density. Apparently, affinity of ES cells to neighbor cells is lower than that for LN-511 but higher than affinity to conventional ECM coatings.

Integrin Receptor Expression and Role in Mouse ES Cell Interaction with LN-511

To analyze the complete integrin expression profile in our ES cells, we used Affymetrix Arrays. The results are summarized in Table 2 (raw array data on integrin receptor subunits expression are given in supplemental online Table 2). However, the Affymetrix Array could not reveal which integrin receptors were expressed in sufficient amounts to attach cell to the surface and which were expressed in minor concentrations. To address that question, we immobilized anti-integrin antibody on plastic surface and identified specific antibodies that could bind and retain ES cells attached to plastic surface (detailed description is given in Materials and Methods). We found out that antibodies against β1-integrin retained ES cells strongly attached to the surface (Fig. 5A, 5B), whereas antibodies against α6, α5β1, and αV could provide only partial adhesion (60%, 17%, and 13%, respectively). Antibodies against α2β1, α3, α4, β2, β3, and αVβ6 could not retain ES cells on the surface (Fig. 5A, 5B).

Table Table 2.. Integrin expression in mouse embryonic stem cells
  1. Abbreviations: ES, embryonic stem; LN, laminin; NA, data not available.

Thumbnail image of
thumbnail image

Figure Figure 5.. Role of integrin receptors in mouse embryonic stem (ES) cell adhesion to LN-511. (A): Adhesion of ES cells to surfaces coated by different anti-integrin antibodies (experiment I). Bars represent (from left to right) ctrl for 100% adhesion, β1 (BD), α6, α5β1, αV, α2β1, α4, αVβ6, β2, and IgG. Positive ctrl (100%) relates to the total number of cells introduced to the surface. IgG were used as negative ctrl. Error bars show SEM (n = 4). Statistical significance calculated by the Student t test is shown: ***, p < .001. (B): Adhesion of ES cells to surface coated by different anti-integrin antibodies (experiment II). Bars represent (from left to right) ctrl for 100% adhesion, β1 (R&D), β1 + β3, β1 (BL), β1 + αV, α2β1, α3, α4, β2, β3 (BD), β3 (BL), and IgG. Error bars show SEM (n = 4). Statistical significance calculated by the Student t test is shown: **, p < .01; ***, p < .001. (C): Adhesion-blocking experiment: inhibition of ES cell adhesion to LN-511 by different anti-integrin antibodies (experiment I). Bars represent (from left to right) ctrl for 100% inhibition, β1 (BD, from mouse), β1 (BD; from Armenian hamster), α6, α6 + αV, α3, α4, αV (BL), αV (Ch), β2, β3 (BD), β3 (BL), β4, and IgG. Ctrl relates to the absence of adherent cells. IgG were used as a ctrl for uninhibited cell adhesion. Error bars show SEM (n = 4). Statistical significance calculated by the Student t test is shown: **, p < .01; ***, p < .001. (D): Adhesion-blocking experiment: inhibition of ES cell adhesion to LN-511 by different anti-integrin antibodies (experiment II). Bars represent (from left to right) ctrl for 100% inhibition, α6, αV, α2β1, α3, α3 (BD), α5β1, β4, and IgG. Ctrl relates to the absence of adherent cells. IgG were used as a ctrl for uninhibited cell adhesion. Error bars show SEM (n = 3). Statistical significance calculated by the Student t test is shown: *, p < .05. (E): ES cell contact area with LN-511-coated surface was reduced after blocking of α6 integrin receptor by antibody. Error bars show SEM; number inside each bar shows number of independent measurements. Statistical significance calculated by the Student t test is shown: ***, p < .001. (F): Immunofluorescence: integrin α6, αV, and α5β1 coexpression with the β1 integrin subunit in pluripotent (Sox2-positive) ES cells cultured on LN-511 (left) or in the presence of LIF on gelatin (right). Magnification, ×40. Scale bars = 100 μm. Abbreviations: BD, BD Biosciences; BL, BioLegend; Ch, Chemicon; ctrl, control; LIF, leukemia inhibitory factor; LN, laminin; R&D, R&D Systems.

Download figure to PowerPoint

To identify integrin receptors involved in ES cell interaction with LN-511, we used function-blocking antibodies against specific integrin receptors in the cell adhesion assay. Blocking of β1-integrins completely inhibited mouse ES cell adhesion to LN-511, as confirmed by two different antibody (Fig. 5C). Blocking of α6-integrin only partially inhibited mouse ES cell adhesion to LN-511, and blocking both α6- and αV-integrin had the same effect (Fig. 5C). Blocking of α2β1, α3, α4, α5β1, αV, β2, β3, and β4-integrins did not affect ES cell adhesion to LN-511 (Fig. 5C, 5D). Interestingly, blocking α6-integrin significantly decreased the cell contact area with LN-511-coated surface (p < .001) (Fig. 5E).

Immunofluorescence staining confirmed expression of integrins α6β1, αVβ1, and α5β1 in pluripotent (Sox2-positive) ES cells cultured on LN-511 in the absence of inhibitors, as well as ES cells cultured on gelatin in the presence of LIF (Fig. 5F). Immunostaining also confirmed colocalization of the α6 and β1 integrin subunits in both culture systems (Fig. 5F).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References
  11. Supporting Information

The present results showed that a specific laminin isoform, namely LN-511, can support self-renewal of mouse ES cells for at least 169 days (31 passages) in the absence of feeder cells, LIF, or other differentiation inhibitors. The effect occurs even at low cell density (<200 cells per mm2). In previous studies, pluripotency of ES cells has been facilitated by soluble differentiation inhibitors [3, 30]. Moreover, in those studies ES cells grew in dense clusters that provided each cell multiple contacts with neighbor cells, which, according to [33], is essential to maintain pluripotent status in the presence of differentiation inhibitors. However, several previous studies have shown that different combinations of matrix proteins can support pluripotency and self-renewal of ES cells, in combination with soluble differentiation inhibitors [3]. For example, Klimanskaya et al. [6] have shown that mouse embryonic fibroblast-derived ECM, together with LIF, is sufficient for human ES cell self-renewal. Our results suggest a novel principle, namely, that contact with certain extracellular matrix molecules can sustain ES cell pluripotency and proliferation in the absence of differentiation inhibitors and in the lack of contacts with neighboring ES cells or feeder cells. We also demonstrated that the effect was laminin isoform-specific.

Notably, there was a striking difference in the effects of various laminin isoforms on ES cells. Although LN-511 enabled self-renewal, LN-111 triggered differentiation and inhibited proliferation, LN-332 enabled proliferation but not self-renewal, and LN-411 did not support adhesion or survival of ES cells at all. LN-511 is the most ancient laminin isoform and one of the first laminin isoforms to appear during embryonic development [9, 16, 11]. The laminin α5 chain (LN-511/LN-521), unlike laminin α1 (LN-111/LN-121), has been found in the matrix between cells of the inner cell mass of blastocysts (the in vivo origin of ES cells) [11]. The laminin α1 chain appears in early development and has another ancient progenitor (α1,2 chain in Drosophila) [9]. It is expressed in the Reichert membrane and some early embryonic basement membranes [9]. Differentiation of nonpolar primitive ectoderm into columnar epithelium of the epiblast is induced by LN-111, which is synthesized by the primitive endoderm [34, 35]. In mice lacking the laminin α1 chain LG4–5 domain, presumptive epiblast cells failed to polarize and did not survive past day 6.5, which demonstrates that this domain provides vital signals for the conversion of stem cells to polarized epithelium [36].

Notably, the ability of LN-511 and LN-332 to support ES cell proliferation correlated with strong adhesion of the ES cells to those two laminins. This observation agrees with previously published data, proving that cell proliferation is strongly dependent on cell contact area with adhesive substratum [37]. From an evolutionary standpoint, the laminin α3 and α5 chains have the same progenitor (α3,5 chain in Drosophila) [9], which may explain partial similarity of their effect on ES cells.

We suggest that mouse ES cells interact with LN-511 via β1-integrins, as we proved that blocking of β1-integrins completely inhibits ES cell adhesion to LN-511. Of all β1-integrins, α6β1 seems to be the main integrin receptor involved in interaction with LN-511 but not the only one. αVβ1 integrin may also contribute to adhesion; however, blocking of both α6 and αV still did not provide complete inhibition or any synergetic effect. We suggest that α8β1, α9β1, or α11β1 may contribute to adhesion, but function-blocking antibodies for those α-subunits were not available.

In our study we used human laminins (LN-332, LN-411, and LN-511) with mouse ES cells. Sequence analysis shows that the sequence similarity between human and mouse species for α-chains of those laminins is high (77%, 88%, and 78%, respectively). Notably, sequence similarity between α-chains of those human laminins is significantly lower (below 35%).

A major problem concerning the use of ECM proteins for technological purposes is the lack of availability of pure native isoforms for such purposes. At present, mouse EHS sarcoma-derived LN-111 is the only laminin isoform commercially available in pure native form for cell culture use. Protease-solubilized laminins from human placenta have been shown to contain a mixture of more or less degraded LN-211, LN-411, or LN-511 that can yield variable and irreproducible results [38]. The present study has demonstrated, however, that human recombinant laminins can be used to develop defined cell culturing systems for mouse ES cells. It was demonstrated that LN-511 possesses differentiation-inhibitory activity that allows the cells to preserve pluripotency even if they lack contacts with neighboring cells and are deprived of soluble differentiation inhibitors. Recombinant human laminins may become a useful tool for stem cell research, as they can be used to generate defined coating substrata. This applies particularly to the establishment and expansion of human ES cells that need to be cultured in xeno-free defined environments if they are to be used for the purpose of human cell therapy.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References
  11. Supporting Information

The present results showed that a specific laminin isoform, namely LN-511, can support self-renewal of mouse ES cells in the absence of feeder cells, LIF, or other differentiation inhibitors, at low cell density. Our results suggest a novel principle, namely, that contact with certain extracellular matrix molecules can sustain ES cell pluripotency. Recombinant human laminins may become a useful tool for stem cell research, as they can be used to generate defined coating substrata.

Disclosure of Potential Conflicts of Interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References
  11. Supporting Information

K.T. owns stock in, has acted as a consultant for, and has performed contract work for BioStratum, Inc., and owns stock in, has acted as a consultant for NephroGenex, Inc.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References
  11. Supporting Information

This study was supported in part by grants from the Swedish Research Council, the Swedish Foundation for Strategic Research, the Knut and Alice Wallenberg Foundation, and Karolinska Institute. We appreciate the assistance of our colleagues Mark Lal, Zhijie Xiao, and Jill O'Loughlin (Karolinska Institute, Sweden).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Potential Conflicts of Interest
  9. Acknowledgements
  10. References
  11. Supporting Information
FilenameFormatSizeDescription
SC-07-0389_Suppl_Fig_1.tif1800KSupplemental Figure 1
SC-07-0389_Suppl_Fig_2.tif1618KSupplemental Figure 2
SC-07-0389_Suppl_Fig_Legends.pdf13KSupplemental Figure Legends
SC-07-0389_Suppl_Table_1.eps225KSupplemental Table 1
SC-07-0389_Suppl_Table_2.pdf47KSupplemental Table 2

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.