August 2007
Volume 48, Issue 8
Free
Cornea  |   August 2007
Hypoxia Enhances the Expansion of Human Limbal Epithelial Progenitor Cells In Vitro
Author Affiliations
  • Hideyuki Miyashita
    From the Department of Ophthalmology, Keio University School of Medicine, Tokyo, Japan; and the
  • Kazunari Higa
    From the Department of Ophthalmology, Keio University School of Medicine, Tokyo, Japan; and the
    Cornea Center and Department of Ophthalmology, and the Tokyo Dental College, Ichikawa, Japan.
  • Naoko Kato
    From the Department of Ophthalmology, Keio University School of Medicine, Tokyo, Japan; and the
  • Tetsuya Kawakita
    Cornea Center and Department of Ophthalmology, and the Tokyo Dental College, Ichikawa, Japan.
  • Satoru Yoshida
    From the Department of Ophthalmology, Keio University School of Medicine, Tokyo, Japan; and the
    Cornea Center and Department of Ophthalmology, and the Tokyo Dental College, Ichikawa, Japan.
  • Kazuo Tsubota
    From the Department of Ophthalmology, Keio University School of Medicine, Tokyo, Japan; and the
    Cornea Center and Department of Ophthalmology, and the Tokyo Dental College, Ichikawa, Japan.
  • Shigeto Shimmura
    From the Department of Ophthalmology, Keio University School of Medicine, Tokyo, Japan; and the
    Cornea Center and Department of Ophthalmology, and the Tokyo Dental College, Ichikawa, Japan.
Investigative Ophthalmology & Visual Science August 2007, Vol.48, 3586-3593. doi:https://doi.org/10.1167/iovs.07-0077
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Hideyuki Miyashita, Kazunari Higa, Naoko Kato, Tetsuya Kawakita, Satoru Yoshida, Kazuo Tsubota, Shigeto Shimmura; Hypoxia Enhances the Expansion of Human Limbal Epithelial Progenitor Cells In Vitro. Invest. Ophthalmol. Vis. Sci. 2007;48(8):3586-3593. https://doi.org/10.1167/iovs.07-0077.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

purpose. To demonstrate the effects of hypoxia on proliferation and differentiation of human limbal epithelial cells in vitro.

methods. Primary human limbal epithelial cells were harvested from the rim of donor corneas. Colony-forming efficiency (CFE) and cell proliferation were observed in standard (20% O2) or hypoxic (2% O2) culture conditions. Cell cycle, forward scatter (FSC) and side scatter (SCC) of cells were analyzed by flow cytometry. Proliferating cells were also observed by pulse labeling (2 hours) with BrdU and Ki67 staining. Apoptosis was detected by TUNEL assay. Isolated colonies were examined by immunohistochemistry against K15, p63, involucrin, and K3. Involucrin expression was also analyzed by Western blot analysis.

results. Both CFE and proliferation of limbal epithelial cells was significantly enhanced in hypoxia. Flow cytometry revealed a higher fraction of hypoxic cells in the G0/G1-phase and fewer cells in the S-phase, compared with normoxia. However, there was no difference in the uptake of BrdU during a 2-hour pulse, suggesting that hypoxic colonies contained rapidly cycling cells. Apoptotic cells were sparse in both groups, and hypoxic cells showed lower FSC compared with normoxic cells. Although there was no difference in the staining pattern of K15, p63, and Ki67, cells cultivated in normoxia expressed higher levels of the differentiation markers involucrin and K3. Significantly higher involucrin expression was also observed by Western blot.

conclusions. Hypoxic culture (2%) enhances proliferation while inhibiting differentiation of limbal epithelial cells in vitro.

Recent advances in the study of limbal epithelial stem cells include the use of cell sorting to isolate the so-called side-population (SP) cells in the limbus, 1 2 which is based on a method first described for isolating bone marrow stem cells. 3 However, there are still no reports of expanding SP cells as an undifferentiated population in vitro, suggesting that this method is not effective in isolating progenitor cells for further ex vivo expansion. Hayashi et al. 4 recently reported the expression of N-cadherin in patches of basal limbal epithelial cells (LECs), which may be used to isolate stem cells by cell sorting, since N-cadherin is a membrane surface protein. 
However, from the point of view of regenerative medicine, the primary goal is to enhance the amount of progenitor cells that can be expanded ex vivo, regardless of the method of primary isolation. The availability of a serum-free, low-Ca2+ medium has significantly improved the yield of cells with high proliferative potential. 5 It is well known that both Ca2+ and serum triggers differentiation in several lines of epithelial cells. 6 7 8 9 Using irradiated or mitomycin-treated 3T3 feeder cells is another method used to isolate colonies of limbal progenitor cells. 10 11 The amniotic membrane with or without an intact epithelial layer was shown to maintain cocultivated epithelial cells in a less-differentiated state, 12 13 14 suggesting that the microenvironment surround the cells is crucial in maintaining an undifferentiated state. However, the use of feeder cells and amniotic substrates is considered a “black box” that introduces several uncharacterized factors to the culture system. Using xenogenic cells may also pose ethical problems when transplanting epithelial sheets cocultured with 3T3 feeder cells. 
Recently, the use of hypoxic incubation was reported to enhance progenitor cells in the bone marrow, 15 neural cells, 16 and epidermal keratinocytes. 17 18 We therefore hypothesized that hypoxia can also be used to induce immature cells to expand from the limbus. Although the cornea is exposed to atmospheric oxygen, it is possible that lower oxygen levels are maintained in the limbal stem cell environment. Recently, oxidative stress was shown to suppress quiescence of stem cells in the bone marrow, 19 further indicating that hypoxia may be beneficial for stem cells in general. In the present study, low levels of O2 (2%) induced the selective proliferation of undifferentiated LECs. Hypoxic cells express lower levels of differentiation markers and form larger colonies, suggesting that hypoxia may help maintain progenitor cells during ex vivo expansion of cultivated epithelial cell sheets. 
Materials and Methods
Tissue Preparation and Cell Culture
Human LECs were isolated from the limbus of eye bank corneas after the central buttons were used for transplantation. Iris, endothelium, and conjunctiva were surgically removed from corneal limbus, and the limbus was treated with 2.5 U/mL Dispase II (Roche, Basel, Switzerland) in F12/DMEM at 4°C overnight. The epithelium was separated from the stroma with a cell scraper and dispersed in 0.05% trypsin EDTA at 37°C for 30 minutes. LECs were suspended in serum-free low-Ca2+ medium (defined K-SFM; Invitrogen, Carlsbad, CA) consisting of 10 ng/mL human recombinant EGF (Invitrogen), 100 ng/mL cholera toxin (Calbiochem; Merck KGaA, Darmstadt, Germany), antibiotics, and growth supplement supplied by the manufacturer. Unless indicated otherwise, 5 × 104 LECs were seeded in 25-cm2 flasks. Flasks were cultured either in 5% CO2 at 37°C as a normoxic control or in hypoxia (2% O2 and 5% CO2 at 37°C) using an N2/CO2 multigas incubator (APM.-30D; Astec, Fukuoka, Japan). Medium was changed every 3 to 4 days. 
CFE and Cell Proliferation
LECs were inoculated in 60-mm dishes at 1000 cells/dish and cultured for 10 to 14 days. Cultured cells were stained with rhodamine B (Wako Pure Chemical Industries, Ltd., Osaka, Japan) for 30 minutes. Colony-forming efficiency (CFE) was calculated as number of colonies/number of inoculated cells. Five independent experiments were performed. 
LEC proliferation was observed in 24-well plates initially seeded with 2500 cells per well. Two wells in each condition were fixed with 70% ethanol every 2 days for up to 24 days. Medium was changed every 2 days. After all wells were fixed, the plates were stained with eosin for 1 hour. Images were scanned, and eosin-stained area was measured using Scion Image software (Scion Corp., Frederick, MD). Four independent experiments were performed. 
Flow Cytometry
LECs (5 × 104 cells) were cultured in 25-cm2 flasks for 12 days in normoxia and hypoxia, as just described, and dispersed by enzyme treatment (37°C, 10 minutes; TrypLE Express; Invitrogen). For forward scatter (FSC) and side scatter (SSC) analysis, the cells were resuspended in 0.1% sodium azide in PBS and analyzed by flow cytometry (EPICS XL; Beckman Coulter, Hialeah, FL). For cell-cycle analysis, the cells were resuspended in a solution containing 4 mM sodium citrate (pH 7.6; Wako), 0.2% Nonidet P-40 (Calbiochem), and 50 μg/mL propidium iodide (Wako). After incubation on ice for 30 minutes, the cell suspensions were treated with 250 μg/mL RNase A (Fermentas, Hanover, MD) for 15 minutes at 37°C to remove double-stranded RNA. Cells were measured by flow cytometry at an excitation wavelength of 488 nm. Data analysis was performed with commercial software (FlowJo; Tree Star, Inc., Ashland, OR). 
BrdU Labeling
LECs (104/well) were cultured in gelatin-coated two-well chamber slides for 7 days, and BrdU (final, 10 μM) was added to the culture for 2 hours. After fixing with methanol at RT for 10 minutes, cells were treated with 1 N HCl at RT for 1 hour, and BrdU was detected as described later. Thirty randomly selected clones in each group were photographed, and the percentage of BrdU+ cells in each colony was calculated. Four independent experiments were performed. 
Reverse Transcription–Polymerase Chain Reaction
Total RNA was extracted from LECs cultured for 10 to 14 days by using the SV total RNA isolation system (Promega Corp., Madison, WI), and cDNA was synthesized by using AVM reverse transcriptase XL (Takara; Bio Inc., Tsu, Japan). The same amount of cDNA was amplified by PCR (GeneAmp 9700; Applied Bioscience, Inc, [ABI], Foster City, CA) and the following primer pairs: GAPDH mRNA, forward (5′-GACCACAGTCCATGCCATCAC-3′), and reverse (5′-TCCACCACCCTGTTGCTGTAG-3′); involucrin mRNA, forward (5′- TGTTCCTCCTCCAGTCAATACC-3′), and reverse (5′-TCCCAGTTGCTCATCTCTCTTG-3′); K3 mRNA, forward (5′-GACAATAATCGTTCCCTGG-3′), and reverse (5′-TTGCGGTAGGTGGCGATCT-3′); K15 mRNA, forward (5-GAGAACTCACTGGCCGAGAC-3′), and reverse (5-GGGACGTTTCTCCTGCAATA-3′); ΔNp63α mRNA, forward (5-CTGGAAAACAATGCCCAGAC-3′) and reverse (5-ATCGCATGTCGAAATTGCTC-3′). 20 PCR products were analyzed by agarose gel electrophoresis. 
Western Blot Analysis
Cell pellets were dissolved with lysis buffer (50 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1% Nonidet P-40) and homogenated. Samples were incubated for 40 minutes at 4°C, and then centrifuged at 15,000 rpm for 30 minutes at 4°C. Protein concentration of the supernatant was determined by the DC protein assay (Bio-Rad Laboratory, Hercules, CA). All samples were then diluted in 2× sample buffer, containing 100 mM Tris-HCl (pH 6.8), 4% SDS (Invitrogen), 20% Glycerol (Wako), and 12% 2-mercaptoethanol (Wako), and boiled. Ten micrograms of each sample were loaded on a 10% Bis-Tris gel (Novex NuPAGE; Invitrogen) and transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Billerica, MA). The membranes were blocked with 5% skim milk (Difco Laboratories, Detroit, MI), 1.5% normal donkey serum, and PBS for 60 minutes at room temperature. Membranes were reacted with an anti-involucrin antibody (SY5; Abcam, Cambridge, UK) and β-actin (mabcam 8226; Abcam) for 60 minutes at room temperature. After three washes in Tris-buffered saline with Tween 20 (TBST), donkey biotinylated anti-mouse IgG (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) was added for 30 minutes at room temperature. Protein bands were visualized by the avidin-biotin complex (Vectastain ABC Elite Kit; Vector Laboratories, Burlingame, CA), with diaminobenzidine (DAB; Vector Laboratories) as substrate. The plot profile of the bands was analyzed with the NIH image 1.63 software (available by ftp at zippy.nimh.nih.gov/ or at http://rsb.info.nih.gov/nih-image; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD). 
Immunostaining
LECs (5 × 103 cells/ well) were cultured in gelatin-coated, four-well chamber slides and fixed with 2% paraformaldehyde (PFA, Wako) for the immunostaining of involucrin, K3, p63, with cold acetone for Ki67, or with cold methanol for K15. PFA-fixed cells were permeabilized with 0.1% Triton X-100 (Sigma-Aldrich, St. Louis, MO) (involucrin, K3, p63). After background staining was blocked with 10% normal donkey serum, the cells were treated with the following monoclonal primary antibodies: anti-K3 (clone AE5; Progen, Heidelberg, Germany), anti-involucrin (SY5, YLEM, Rome, Italy), anti-Ki67 antigen (MIB-1; Dako, Glostrup, Denmark), anti-p63 (4A4; Calbiochem), anti-K15 (LHK15; Laboratory Vision, Fremont, CA), and anti-BrdU (Chemicon International, Temecula, CA). The cells were then treated with Alexa Fluor 488– or 555–conjugated secondary antibodies (Invitrogen) or rhodamine (Jackson ImmunoResearch)- or Cy3 (Chemicon)-conjugated secondary antibodies. The nuclei were counterstained with 4′,6′-diamino-2-phenylindole (1 mg/mL, DAPI; Dojindo Laboratories, Tokyo, Japan) or TO-PRO-3 (Invitrogen). 
TUNEL Assay
The terminal deoxynucleotidyl transferase (TdT)–mediated dUTP nick-end labeling (TUNEL) technique was performed to detect apoptosis (In situ cell death detection kit; TMR red; Roche Diagnostics, Indianapolis, IN). Chamber slides were fixed with 4% paraformaldehyde (Wako) for 1 hour and permeabilized in 0.1% Triton X-100, 0.1% sodium citrate at 4°C for 2 minutes. The slides were incubated with TUNEL reaction mixture for 1 hour at 37°C. After washing, the slides were counterstained with DAPI (1 mg/mL; Dojindo Laboratories). 
Results
CFE and Cell Proliferation
Figure 1shows colony formation by limbal epithelial cells in normoxia and hypoxia after 10 to 14 days. Normoxic colonies (Fig. 1A)were smaller compared with colonies formed in hypoxia (Fig. 1B) . In addition to colony size, colony-forming efficiency was significantly higher in hypoxic cultures (Fig. 1C) . Hypoxia yielded higher colonies at all seeding densities investigated in a preliminary experiment (Supplementary Fig. S1 and Methods). We further examined whether hypoxia caused an increase in cell proliferation rate. Despite the same initial seeding density, hypoxic colonies had a significantly higher number of cells per colony after 7 days (Fig. 2A) . When the area covered by epithelial cells was used as a measure of cell proliferation, we found that hypoxic cultures reached confluence faster than normoxic cultures (Fig. 2B)and that the average area covered by the cells was significantly higher after 16 days, until the cells became confluent (Fig. 2C)
Cell Cycle Analysis
We examined the cell-cycle profile of LECs cultured in hypoxia to determine whether the increased proliferation can be explained by a higher ratio of dividing cells. Flow cytometry of cells in the proliferative stage (12 days) showed that hypoxic cells were predominantly in the G0/G1-phase of the cell cycle, whereas the number of cells in the DNA-replicating stage (S-phase) was more evident in normoxic cells (Fig. 3A) . The higher number of cells in the S-phase and fewer cells in the G0/G1-phase in normoxia was statistically significant compared with hypoxia (Fig. 3B)
Cell Proliferation Markers and Apoptosis
To explain the paradoxical results of flow cytometry, we pulse labeled cells for 2 hours to observe the total number of dividing cells during the period. The ratio of BrdU+ cells was not only similar, but also tended to be higher in the hypoxia group (Figs. 4A 4B 4C) . When cells were stained with the proliferation marker Ki67, both normoxic (Fig. 4D)and hypoxic (Fig. 4E)colonies were positive, especially in cells toward the center of the colony. This trend was more evident in hypoxic cells, which constantly produced larger colonies. However, the ratio of Ki67+ cells within the entire colony was not different between both groups. To rule out the possibility of increased apoptosis in normoxia as the reason for smaller colonies and slower proliferation, we performed TUNEL staining of colonies cultured for 12 days. However, cells undergoing apoptosis were sparse in both groups during this phase of proliferation (Figs. 4F 4G) . There was also no difference in PI-positive dead cells during earlier phases of cell culture (data not shown). 
Phenotype of Limbal Epithelial Cells in Hypoxia
Differentiation of LECs cultivated in hypoxia and normoxia were compared. Flow cytometry revealed that hypoxic LECs were smaller, as shown by lower FSC in cells cultivated in hypoxia (Fig. 5A) . Although there was no significant difference in FSC, the average cell density was significantly higher in hypoxic colonies due to the presence of enlarged cells scattered in normoxic colonies (Supplementary Fig. S2 and Methods). There was no difference in the expression of the progenitor markers K15 and p63 when examined by RT-PCR (Fig. 5B)and immunocytochemistry (Figs. 5C5D5E5F) . However, the differentiation markers involucrin and K3 showed stronger staining in normoxic colonies (Figs. 6A6C)compared with hypoxic colonies (Figs. 6B6D) . The difference was confirmed by RT-PCR (Fig. 6E)and by quantitating protein expression of involucrin by Western blots (Fig. 6F) . These results show that cells cultured in normoxia are driven toward differentiation. 
Discussion
Maintaining limbal progenitor cells in an immature state in vitro is vital in the study of limbal stem cell biology and for cultivating transplantable epithelial cell sheets. Recently, several studies on tissue stem cells have used hypoxic culture conditions to propagate cells in an immature state. Reduced oxygen was shown to enhance the survival of central nervous system (CNS) stem cells. 21 Hypoxia not only promoted proliferation of CNS stem cells, but also reduced apoptosis. 16 The effects of hypoxia on epithelial cells was reported in epidermal keratinocytes cultured in 2% O2 that showed increased colony-forming ability 18 and increased population doubling producing a larger yield of epithelial cells to form epithelial sheets. 17  
We hypothesized that hypoxia may also be effective in expanding limbal epithelial progenitor cells. As expected, culturing primary LECs in 2% O2 without feeder cells produced larger colonies than did culturing in normoxia (Figs. 1A 1B) . Furthermore, CFE was higher in hypoxic cultures (Figs. 1C 1D) . We further observed the effects of hypoxia on cell proliferation by seeding primary LECs at higher densities, and found that cell growth was enhanced in hypoxia giving rise to larger initial colonies that reached confluence earlier than normoxic cultures (Fig. 2)
A clear enhancement of cell proliferation prompted us to examine the cell cycle profile of cultured LECs by flow cytometry. Of interest, we found that a significantly higher number of cells were in the S- and G2/M-phases during normoxia, and that most of the cells in hypoxia were in the G0/G1-phase (Fig. 3) . This observation seems to contradict the notion of a higher proliferation rate in hypoxia, since flow cytometry results indicated a larger number of cells undergoing cell division in normoxia. Because flow cytometry only shows the state of cells at a specific time point, we pulse labeled cells with BrdU for 2 hours, to identify the cells entering the cell division cycle during this time frame. There was no difference in BrdU+ cells between both groups, with a tendency toward a higher ratio in hypoxia (Figs. 4A 4B 4C) . There was also no difference in the expression of the proliferation marker Ki67 (Figs. 4D 4E) , as well as apoptotic cells shown by TUNEL staining (Figs. 4F 4G) . The results suggest that the larger colonies and increased cell proliferation in hypoxia is due to the presence of rapidly cycling cells and explains why more cells were in the nondividing G1-phase of the cell cycle in hypoxia at a specific time point, whereas the number of BrdU-labeled cells during a 2-hour pulse was the same. 
The next step of the study was to examine the differentiation status of cells in both conditions. Although flow cytometry suggested that hypoxic cells were slightly smaller than normoxic cells (Fig. 5A) , there was no difference in the expression of the progenitor markers K15 and p63 examined by RT-PCR (Fig. 5B)and immunocytochemistry (Figs. 5C 5D 5E 5F) . However, normoxic cells expressed higher levels of the expression markers involucrin and K3 (Fig. 6) . Normoxic colonies also had a significantly lower cell density due to the presence of scattered large cells (Supplementary Fig. S2). These results clearly show that hypoxic conditions maintain LECs in a more undifferentiated state. Because primary LECs were from the same donor source for each experiment, we can assume that the initial density of stem cells was the same. An increase in rapidly proliferating, undifferentiated cells in hypoxia may indicate that lower oxygen levels facilitate the proliferation of transient amplifying (TA) cells. 
The mechanisms involved in the maintenance of progenitors during hypoxia are still not clear. One of the major intracellular regulators during hypoxia is hypoxia-inducible factor (HIF1)-α, and several reports have already suggested a role for HIF1-α in the inhibition of adipocyte differentiation during hypoxia. 22 HIF1-α was also shown to interact with Notch-responsive promoters during Notch activation in hypoxia to block neuronal and myogenic differentiation. 23 However, Notch seems to induce differentiation in keratinocytes, 24 suggesting that several pathways exist in the maintenance of the undifferentiated state in hypoxia. Perhaps oxygen at atmospheric levels alone may be a source of reactive oxygen species that may drive stem cells from a quiescent state as was shown in bone marrow stem cells. 19  
We have shown that limbal epithelial progenitor cells can be efficiently expanded in serum-free, feeder-free medium in hypoxic conditions. This protocol may have an impact on the way ex vivo expansion is performed in the future. For example, maintaining progenitor cells in a less differentiated state may allow the engineering of transplantable epithelial sheets from even a single cell source. One question that remains to be clarified is whether hypoxia can maintain quiescent stem cells during cultivation. Although further studies are needed resolve this question, similar observations found in various other cell types suggest that a hypoxic microenvironment is a key component of the epithelial stem-cell niche. 
 
Figure 1.
 
CFE in normoxic and hypoxic cultures. LECs seeded at clonal densities formed colonies in both normoxia (A) and hypoxia (B) without the use of feeder cells. Individual colonies were larger in hypoxia (C), and CFE was statistically higher in the hypoxia group than in the normoxia control (D: n = 5, Student’s t-test). Scale bars, 100 μm.
Figure 1.
 
CFE in normoxic and hypoxic cultures. LECs seeded at clonal densities formed colonies in both normoxia (A) and hypoxia (B) without the use of feeder cells. Individual colonies were larger in hypoxia (C), and CFE was statistically higher in the hypoxia group than in the normoxia control (D: n = 5, Student’s t-test). Scale bars, 100 μm.
Figure 2.
 
Enhanced cell proliferation in hypoxia. LEC proliferation was enhanced in hypoxia, with larger initial colonies (A) and rapid expansion of cells within the culture dish (B). The average area covered by proliferating cells was significantly higher in hypoxia by day 16 until the cells reached confluence (C: n = 4, *P < 0.05 Mann-Whitney test). N, normoxia; H, hypoxia.
Figure 2.
 
Enhanced cell proliferation in hypoxia. LEC proliferation was enhanced in hypoxia, with larger initial colonies (A) and rapid expansion of cells within the culture dish (B). The average area covered by proliferating cells was significantly higher in hypoxia by day 16 until the cells reached confluence (C: n = 4, *P < 0.05 Mann-Whitney test). N, normoxia; H, hypoxia.
Figure 3.
 
Cell-cycle profile of cultured LECs. Flow cytometry showed a prominent peak of G0/G1 cells in the hypoxia group, whereas normoxic cells were distributed throughout the cell cycle (A). The average ratio of cells in the G0/G1-phase was significantly higher in the hypoxia group, whereas cells in the S- and G2/M-phase were significantly higher in the normoxic group (B: n = 6, Student’s t-test). N, normoxia; H, hypoxia.
Figure 3.
 
Cell-cycle profile of cultured LECs. Flow cytometry showed a prominent peak of G0/G1 cells in the hypoxia group, whereas normoxic cells were distributed throughout the cell cycle (A). The average ratio of cells in the G0/G1-phase was significantly higher in the hypoxia group, whereas cells in the S- and G2/M-phase were significantly higher in the normoxic group (B: n = 6, Student’s t-test). N, normoxia; H, hypoxia.
Figure 4.
 
Cell proliferation and apoptosis markers. Although colonies cultured in hypoxia were constantly larger than those raised in normoxia, the ratio of BrdU+ cells (green) after a 2-hour pulse was similar in both groups (AC). Similarly, there was no difference in the ratio of Ki67+ (red) cells (D, E). TUNEL+ apoptotic cells (red) were sparse in both groups during the growth phase (day 12) of culture (F, G). Nuclei counter-stained with TO-PRO-3 (A, B) and DAPI (DG). Scale bars, 100 μm.
Figure 4.
 
Cell proliferation and apoptosis markers. Although colonies cultured in hypoxia were constantly larger than those raised in normoxia, the ratio of BrdU+ cells (green) after a 2-hour pulse was similar in both groups (AC). Similarly, there was no difference in the ratio of Ki67+ (red) cells (D, E). TUNEL+ apoptotic cells (red) were sparse in both groups during the growth phase (day 12) of culture (F, G). Nuclei counter-stained with TO-PRO-3 (A, B) and DAPI (DG). Scale bars, 100 μm.
Figure 5.
 
Expression of progenitor markers. Flow cytometry showed that hypoxic cells had slightly lower FSC, whereas SSC was similar (A). RT-PCR (B) and immunocytochemistry (CF) revealed that the expression of the progenitor markers K15 (red) and p63 (red) were similar in both groups. (CF) Nuclei counterstained with TO-PRO-3. Scale bar, 100 μm.
Figure 5.
 
Expression of progenitor markers. Flow cytometry showed that hypoxic cells had slightly lower FSC, whereas SSC was similar (A). RT-PCR (B) and immunocytochemistry (CF) revealed that the expression of the progenitor markers K15 (red) and p63 (red) were similar in both groups. (CF) Nuclei counterstained with TO-PRO-3. Scale bar, 100 μm.
Figure 6.
 
Expression of differentiation markers. Normoxic colonies expressed higher levels of the differentiation markers involucrin (red) (A) and K3 (red) (C) compared with hypoxic colonies (B, D). The difference was also confirmed by RT-PCR for involucrin and K3 (E) and by Western blot for involucrin (F). Semiquantitative analysis of Western blots showed that involucrin expression by normoxic cells was significantly higher in normoxia (n = 6, Student’s t-test). (AD) Nuclei were counterstained with DAPI. Scale bar, 200 μm.
Figure 6.
 
Expression of differentiation markers. Normoxic colonies expressed higher levels of the differentiation markers involucrin (red) (A) and K3 (red) (C) compared with hypoxic colonies (B, D). The difference was also confirmed by RT-PCR for involucrin and K3 (E) and by Western blot for involucrin (F). Semiquantitative analysis of Western blots showed that involucrin expression by normoxic cells was significantly higher in normoxia (n = 6, Student’s t-test). (AD) Nuclei were counterstained with DAPI. Scale bar, 200 μm.
Supplementary Materials
Colony-forming efficiency of HLEC seeded at various densities in normoxic and hypoxic culture. Hypoxic colonies were larger than normoxic colony in all seeding densities (A, rhodamine; B, stained dishes). Colony-forming efficiency in 300, 500, and 1000 cells/dish seeding densities were 1.1 ± 1.0 % under normoxia and 2.8 ± 1.6 % under hypoxia, 1.2 ± 0.5 % in normoxia and 2.6 ± 1.6 % in hypoxia, and 1.3 ± 0.6 % in normoxia and 2.6 ± 1.6 % in hypoxia, respectively. Difference in CFE reached statistical significance at densities of 300 and 1000 cells/dish (B: P < 0.05, paired t-test, n = 5). 
Cell densities of hypoxic and normoxic colonies. Cultures were fixed when colonies reached at same approximate size (A, B), and nuclei were stained with DAPI (C, D). Cell density of hypoxic colony (625.9 ± 136.4 cells per mm2, mean ± S.D., n = 20) was significantly higher than normoxic culture (478.1 ± 149.0 cells per mm2) (P < 0.01, paired t-test). Scale bar, 200 μm. 
The authors thank Mifuyu Ishiwata, Tomomi Sekiguchi, and Fumito Morito for their technical assistance and the staff of the Cornea Center Eye Bank for administrative support and Nobuhito Goda and Makoto Suematsu (Department of Biochemistry, Keio University School of Medicine, Tokyo, Japan) for use of their hypoxic culture chamber. 
WatanabeK, NishidaK, YamatoM, et al. Human limbal epithelium contains side population cells expressing the ATP-binding cassette transporter ABCG2. FEBS Lett. 2004;565:6–10. [CrossRef] [PubMed]
BudakMT, AlpdoganOS, ZhouM, LavkerRM, AkinciMA, WolosinJM. Ocular surface epithelia contain ABCG2-dependent side population cells exhibiting features associated with stem cells. J Cell Sci. 2005;118:1715–1724. [CrossRef] [PubMed]
GoodellMA, BroseK, ParadisG, ConnerAS, MulliganRC. Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J Exp Med. 1996;183:1797–1806. [CrossRef] [PubMed]
HayashiR, YamatoM, SugiyamaH, et al. N-cadherin is expressed by putative stem/progenitor cells and melanocytes in the human limbal epithelial stem cell niche. Stem Cells. 2007;25:289–296. [CrossRef] [PubMed]
KruseFE, TsengSC. Proliferative and differentiative response of corneal and limbal epithelium to extracellular calcium in serum-free clonal cultures. J Cell Physiol. 1992;151:347–360. [CrossRef] [PubMed]
HenningsH, MichaelD, ChengC, SteinertP, HolbrookK, YuspaSH. Calcium regulation of growth and differentiation of mouse epidermal cells in culture. Cell. 1980;19:245–254. [CrossRef] [PubMed]
BoyceST, HamRG. Calcium-regulated differentiation of normal human epidermal keratinocytes in chemically defined clonal culture and serum-free serial culture. J Invest Dermatol. 1983;81:33s–40s. [CrossRef] [PubMed]
HenningsH, HolbrookKA. Calcium regulation of cell-cell contact and differentiation of epidermal cells in culture: an ultrastructural study. Exp Cell Res. 1983;143:127–142. [CrossRef] [PubMed]
WattFM. Involucrin and other markers of keratinocyte terminal differentiation. J Invest Dermatol. 1983;81:100s–103s. [CrossRef] [PubMed]
LindbergK, BrownME, ChavesHV, KenyonKR, RheinwaldJG. In vitro propagation of human ocular surface epithelial cells for transplantation. Invest Ophthalmol Vis Sci. 1993;34:2672–2679. [PubMed]
TsengSC, KruseFE, MerrittJ, LiDQ. Comparison between serum-free and fibroblast-cocultured single-cell clonal culture systems: evidence showing that epithelial anti-apoptotic activity is present in 3T3 fibroblast-conditioned media. Curr Eye Res. 1996;15:973–984. [CrossRef] [PubMed]
GrueterichM, EspanaE, TsengSC. Connexin 43 expression and proliferation of human limbal epithelium on intact and denuded amniotic membrane. Invest Ophthalmol Vis Sci. 2002;43:63–71. [PubMed]
GrueterichM, EspanaEM, TsengSC. Modulation of keratin and connexin expression in limbal epithelium expanded on denuded amniotic membrane with and without a 3T3 fibroblast feeder layer. Invest Ophthalmol Vis Sci. 2003;44:4230–4236. [CrossRef] [PubMed]
HigaK, ShimmuraS, KawakitaT, et al. Proliferation and differentiation of transplantable epithelial sheets engineered with or without an amniotic membrane carrier. Invest Ophthalmol Vis Sci. 2007;48:597–604. [CrossRef] [PubMed]
CipolleschiMG, Dello SbarbaP, OlivottoM. The role of hypoxia in the maintenance of hematopoietic stem cells. Blood. 1993;82:2031–2037. [PubMed]
StuderL, CseteM, LeeSH, et al. Enhanced proliferation, survival, and dopaminergic differentiation of CNS precursors in lowered oxygen. J Neurosci. 2000;20:7377–7383. [PubMed]
Kino-okaM, AgatahamaY, HagaY, InoieM, TayaM. Long-term subculture of human keratinocytes under an anoxic condition. J Biosci Bioeng. 2005;100:119–122. [CrossRef] [PubMed]
NgoMA, SinitsynaNN, QinQ, RiceRH. Oxygen-dependent differentiation of human keratinocytes. J Invest Dermatol. 2007;127(354)361.
ItoK, HiraoA, AraiF, et al. Reactive oxygen species act through p38 MAPK to limit the lifespan of hematopoietic stem cells. Nat Med. 2006;12:446–451. [CrossRef] [PubMed]
KawasakiS, TaniokaH, YamasakiK, ConnonCJ, KinoshitaS. Expression and tissue distribution of p63 isoforms in human ocular surface epithelia. Exp Eye Res. 2006;82:293–299. [CrossRef] [PubMed]
MorrisonSJ, CseteM, GrovesAK, MelegaW, WoldB, AndersonDJ. Culture in reduced levels of oxygen promotes clonogenic sympathoadrenal differentiation by isolated neural crest stem cells. J Neurosci. 2000;20:7370–7376. [PubMed]
YunZ, MaeckerHL, JohnsonRS, GiacciaAJ. Inhibition of PPAR gamma 2 gene expression by the HIF-1-regulated gene DEC1/Stra13: a mechanism for regulation of adipogenesis by hypoxia. Dev Cell. 2002;2:331–341. [CrossRef] [PubMed]
GustafssonMV, ZhengX, PereiraT, et al. Hypoxia requires notch signaling to maintain the undifferentiated cell state. Dev Cell. 2005;9:617–628. [CrossRef] [PubMed]
RangarajanA, TaloraC, OkuyamaR, et al. Notch signaling is a direct determinant of keratinocyte growth arrest and entry into differentiation. EMBO J. 2001;20:3427–3436. [CrossRef] [PubMed]
Figure 1.
 
CFE in normoxic and hypoxic cultures. LECs seeded at clonal densities formed colonies in both normoxia (A) and hypoxia (B) without the use of feeder cells. Individual colonies were larger in hypoxia (C), and CFE was statistically higher in the hypoxia group than in the normoxia control (D: n = 5, Student’s t-test). Scale bars, 100 μm.
Figure 1.
 
CFE in normoxic and hypoxic cultures. LECs seeded at clonal densities formed colonies in both normoxia (A) and hypoxia (B) without the use of feeder cells. Individual colonies were larger in hypoxia (C), and CFE was statistically higher in the hypoxia group than in the normoxia control (D: n = 5, Student’s t-test). Scale bars, 100 μm.
Figure 2.
 
Enhanced cell proliferation in hypoxia. LEC proliferation was enhanced in hypoxia, with larger initial colonies (A) and rapid expansion of cells within the culture dish (B). The average area covered by proliferating cells was significantly higher in hypoxia by day 16 until the cells reached confluence (C: n = 4, *P < 0.05 Mann-Whitney test). N, normoxia; H, hypoxia.
Figure 2.
 
Enhanced cell proliferation in hypoxia. LEC proliferation was enhanced in hypoxia, with larger initial colonies (A) and rapid expansion of cells within the culture dish (B). The average area covered by proliferating cells was significantly higher in hypoxia by day 16 until the cells reached confluence (C: n = 4, *P < 0.05 Mann-Whitney test). N, normoxia; H, hypoxia.
Figure 3.
 
Cell-cycle profile of cultured LECs. Flow cytometry showed a prominent peak of G0/G1 cells in the hypoxia group, whereas normoxic cells were distributed throughout the cell cycle (A). The average ratio of cells in the G0/G1-phase was significantly higher in the hypoxia group, whereas cells in the S- and G2/M-phase were significantly higher in the normoxic group (B: n = 6, Student’s t-test). N, normoxia; H, hypoxia.
Figure 3.
 
Cell-cycle profile of cultured LECs. Flow cytometry showed a prominent peak of G0/G1 cells in the hypoxia group, whereas normoxic cells were distributed throughout the cell cycle (A). The average ratio of cells in the G0/G1-phase was significantly higher in the hypoxia group, whereas cells in the S- and G2/M-phase were significantly higher in the normoxic group (B: n = 6, Student’s t-test). N, normoxia; H, hypoxia.
Figure 4.
 
Cell proliferation and apoptosis markers. Although colonies cultured in hypoxia were constantly larger than those raised in normoxia, the ratio of BrdU+ cells (green) after a 2-hour pulse was similar in both groups (AC). Similarly, there was no difference in the ratio of Ki67+ (red) cells (D, E). TUNEL+ apoptotic cells (red) were sparse in both groups during the growth phase (day 12) of culture (F, G). Nuclei counter-stained with TO-PRO-3 (A, B) and DAPI (DG). Scale bars, 100 μm.
Figure 4.
 
Cell proliferation and apoptosis markers. Although colonies cultured in hypoxia were constantly larger than those raised in normoxia, the ratio of BrdU+ cells (green) after a 2-hour pulse was similar in both groups (AC). Similarly, there was no difference in the ratio of Ki67+ (red) cells (D, E). TUNEL+ apoptotic cells (red) were sparse in both groups during the growth phase (day 12) of culture (F, G). Nuclei counter-stained with TO-PRO-3 (A, B) and DAPI (DG). Scale bars, 100 μm.
Figure 5.
 
Expression of progenitor markers. Flow cytometry showed that hypoxic cells had slightly lower FSC, whereas SSC was similar (A). RT-PCR (B) and immunocytochemistry (CF) revealed that the expression of the progenitor markers K15 (red) and p63 (red) were similar in both groups. (CF) Nuclei counterstained with TO-PRO-3. Scale bar, 100 μm.
Figure 5.
 
Expression of progenitor markers. Flow cytometry showed that hypoxic cells had slightly lower FSC, whereas SSC was similar (A). RT-PCR (B) and immunocytochemistry (CF) revealed that the expression of the progenitor markers K15 (red) and p63 (red) were similar in both groups. (CF) Nuclei counterstained with TO-PRO-3. Scale bar, 100 μm.
Figure 6.
 
Expression of differentiation markers. Normoxic colonies expressed higher levels of the differentiation markers involucrin (red) (A) and K3 (red) (C) compared with hypoxic colonies (B, D). The difference was also confirmed by RT-PCR for involucrin and K3 (E) and by Western blot for involucrin (F). Semiquantitative analysis of Western blots showed that involucrin expression by normoxic cells was significantly higher in normoxia (n = 6, Student’s t-test). (AD) Nuclei were counterstained with DAPI. Scale bar, 200 μm.
Figure 6.
 
Expression of differentiation markers. Normoxic colonies expressed higher levels of the differentiation markers involucrin (red) (A) and K3 (red) (C) compared with hypoxic colonies (B, D). The difference was also confirmed by RT-PCR for involucrin and K3 (E) and by Western blot for involucrin (F). Semiquantitative analysis of Western blots showed that involucrin expression by normoxic cells was significantly higher in normoxia (n = 6, Student’s t-test). (AD) Nuclei were counterstained with DAPI. Scale bar, 200 μm.
Supplementary Methods
Supplementary Figure S1
Supplementary Figure S2
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×