Single cancer stem–like cells (CSC) from colorectal cancers can be functionally identified by their ability to form large lumen-containing colonies in three-dimensional Matrigel cultures. These colonies contain the three types of differentiated colorectal epithelial cells, and single cells obtained from them can reproduce themselves and form tumors efficiently in immunodeficient mice. In this study, we show how hypoxia affects these CSC-derived lumens to control differentiation of stem-like cells and enterocytes via the homeobox gene CDX1. Lumens were identified by F-actin staining and they expressed many characteristics associated with normal differentiated intestinal epithelium, including brush border enzymes, polarization, and tight junctions. RNA interference–mediated silencing of CDX1 reduced lumen formation. Inhibitory effects of hypoxia on lumen formation and stem cell differentiation, including suppression of CDX1 expression, could be mimicked by inhibiting prolyl-hydroxylases that activate HIF1, suggesting that HIF1 is a critical mediator of the effects of hypoxia in this setting. Cell line–derived lumens were phenotypically indistinguishable from colorectal tumor glandular structures used by pathologists to grade tumor differentiation. Parallel results to those obtained with established cell lines were seen with primary cultures from fresh tumors. This in vitro approach to functional characterization of CSCs and their differentiation offers a valid model to study colorectal tumor differentiation and differentiation of colorectal CSCs, with additional uses to enable high-throughput screening for novel anticancer compounds. Cancer Res; 73(18); 5798–809. Β©2013 AACR.

Colorectal tumors originate in the epithelial layer of the large intestine, which is lined with microinvaginations called crypts. A stem cell population resides at the crypt base and gives rise to the 3 main differentiated lineages in colon: enterocytes, goblet cells, and entero-endocrine cells (1). Enterocytes lining the crypts of the intestine express a layer of microvilli on their apical membranes known as the brush border, in which many differentiation markers such as villin are segregated in a polarized manner (2–4). Full development of the brush border and associated cytoskeletal and enzyme activities corresponds with intestinal differentiation (5, 6). Although cellular differentiation is ultimately deranged in colonic adenocarcinomas, colorectal tumors usually retain some degree of differentiation, such as glandular structure and expression of differentiation markers, and also frequently express brush border–enriched enzymes such as DPPIV/CD26 and alkaline phosphatase (7). The glandular form of tumors and general mucinous content are used by pathologists to help grade tumors (8). Tumors are classed as either well/moderately differentiated or β€œlow grade” if they contain many β€œneoplastic glands” that have a crypt-like structure or as poorly differentiated or β€œhigh grade” if they lack such glandular structures (9). Low-grade tumors generally exhibit a better prognosis for the patient (10).

Much evidence indicates that adenocarcinomas are driven by a subset of cells with characteristics of stem cells, including the ability to self-renew and to differentiate into all 3 colonic lineages (11). We have previously shown that a subpopulation of single cells derived from certain colorectal cell lines can, when grown for several weeks under 3-dimensional (3D) conditions, form either large crypt-like structures consisting of polarized cells surrounding a cell-free lumen or instead form small non-lumen colonies (12–15). The large lumen colonies express differentiation markers for all 3 colon cell lineages, indicating the multipotent nature of the original cell. Single cells derived from large lumen colonies were capable of forming new large lumen colonies, as well as small non-lumen colonies, whereas cells derived from small non-lumen colonies could only give rise to further small colonies (14). Lumen colony–derived cells were also more tumorigenic in mouse xenografts than cells derived from small non-lumen colonies. The proportion of cells that gives rise to lumen colonies could be increased by enrichment of cells with high expression of both the stem cell markers CD44 and CD24 (14). Collectively, these characteristics indicate that lumen formation is driven by high clonogenic cancer stem cells that can be derived from lumen-forming cell lines. Low clonogenic cells are unable to give rise to large lumen colonies and form small non-lumen colonies. Therefore, formation of lumens from single cells can be used to characterize cancer stem cell differentiation.

Interestingly, some colorectal cell lines such as DLD1 do not differentiate under 3D conditions and instead form disorganized colonies (14, 15). These cell lines retain self-renewal capacity of stem cells but lack the ability to fully differentiate and are correspondingly more aggressive in xenografts and more clonogenic than lumen-forming cell lines (14). One possibility is that differentiation is inhibited in these lines by reduced expression of key transcription factors such as CDX1. CDX1 is a gut transcription factor crucial to colorectal differentiation (16, 17), and it controls the transcription of a number of intestinal differentiation markers including villin (18), cytokeratin 20 (16), and FABP1 (19). In colorectal cancers, CDX1 is frequently transcriptionally silenced by promoter methylation (16, 17).

In this study, we have characterized the nature of differentiation of 3D colonies formed by different cell lines and primary tumor-derived cancer stem cells.

Cell culture

Details of cell line origins and validation can be found in the Supplementary Text. All cell lines were cultured in complete Dulbecco's Modified Eagle Media (DMEM) supplemented with 10% FBS and 1% penicillin/streptomycin (Invitrogen). Hypoxic cells were grown in humidified 1% oxygen and 10% CO2 environment using a MiniGalaxy incubator (RS Biotech Ltd). Medium was replaced 2 times per week after being prewarmed and equilibrated overnight in 1% oxygen. For dimethyloxaloylglycine (DMOG) treatment, culture medium containing 2 mmol/L DMOG or dimethyl sulfoxide (DMSO; vehicle control) was overlaid over the gels and this was changed every 24 hours.

For primary culture, 6 moderate/well-differentiated tumor specimens, obtained from fully consenting donors under National Research Ethics study, 07/H0606/120, were mechanically disrupted into small pieces and cultured in serum0free medium (Excell-620, Sigma-Aldrich Ltd.) supplemented with 2% StemPro hESC supplement (Life Technologies Inc.) on nonadherent plastic, until spheroids were formed. The cultures were checked by EpCam immunofluorescence and reverse transcriptase PCR to consist of CDX1-expressing epithelial cells (not shown). Established colonies were transferred to Matrigel and grown for 7 days, before fixation and immunostaining, as detailed below.

F-actin assay of colonies grown in Matrigel

Single-cell suspensions were achieved by fluorescence-activated cell sorting (FACS) or filtration through 30- and 20-ΞΌm filters (Celltrics, Partec GmbH). Five hundred cells were suspended in 40 ΞΌL of an ice-cold mixture of Matrigel and diluted 1:1 with ice-cold DMEM. Cell suspensions were seeded in triplicate into 96-well plates precoated with a solidified 1:1 Matrigel/DMEM layer (20 ΞΌL per well). Overlaid Matrigel was allowed to set for 20 minutes at 37Β°C, and culture medium was then added to the wells. Colonies were grown for 2 weeks, unless otherwise specified, with medium changes every 3 days. The medium was removed and 100 ΞΌL of 4% (v/v) paraformaldehyde in PBS was added for 20 minutes. Fluid was then removed and 100 ΞΌL of PBS containing 1:200 dilution of Triton-x was added for 10 minutes. The fluid was removed and the wells were washed 4 times with 50 mmol/L glycine in PBS. About 100 ΞΌL TRITC-phalloidin diluted at 1:1,000 in PBS was added and incubated at 4Β°C overnight. The colonies were then rinsed 3 times with PBS as above. A total of 100 ΞΌL of 10 ΞΌg/mL 4β€²6-diamidino-2-phenylindole (DAPI; Sigma) in PBS was added to each well and the plate was imaged using a Zeiss LSM510 laser scanning confocal microscope (Carl Zeiss Ltd.).

FACS analysis

Cells were harvested and 1 Γ— 106 cells and labeled with anti-CD44 and anti-CD24. Cell viability was assessed with Sytox Blue. Cells were run on a Dako MoFLow cytometer. Gates were set to exclude dead cells, debris, and doublets and the top or bottom 5% of gated CD44/CD24 (see Fig. 3). Labeled cells were collected into separate tubes containing 200 ΞΌL ice-cold DMEM. Cells were then embedded in Matrigel as described above and assayed after 2-week growth.

Western blotting

Western blotting was conducted on cell lysates as described previously (20).

In vivo tumorigenic assays

Single cells were injected subcutaneously into the flanks of 6- to 10-week-old female NOD/SCID immunodeficient mice, obtained from the John Radcliffe Hospital Biomedical Services (Oxford, UK) as described previously (14). Animal studies were conducted according to the University of Oxford institutional guidelines and within the limits of the Project License issued by the Home Office, United Kingdom.

Further details of materials and methods can be found in the Supplementary Data.

Lumen characterization

F-actin (filamentous actin) was identified as a useful marker of lumens by staining normal human colon cryosections with fluorescent phalloidin, which is specific for F-actin (Fig. 1A). In normal colon, F-actin is intensely enriched at the apical surface of enterocytes lining the crypt, corresponding to the brush border, presumably due to the presence of microvilli on colonic enterocytes (21, 22). We next examined F-actin labeling in single-cell–derived colonies from a panel of 6 colon cancer cell lines, grown in a 3D Matrigel matrix for 2 weeks. In an SW1222 colony, confocal Z-sectioning showed intense F-actin labeling on the apical cell membranes facing the lumen, which was surrounded by polarized cells expressing cytokeratin 20 on their basolateral membranes (Fig. 1B). LS180, SW1222, and C80 formed colonies consisting of polarized cells surrounding central lumens visible by light microscopy (Fig. 1C). All phase visible lumens were clearly labeled by intense F-actin staining. HT29, HCT116, and DLD1 did not form observable lumen colonies using phase contrast and did not exhibit organized bright foci of F-actin staining, although F-actin was visible in plasma membranes. A gallery of F-actin–labeled colonies from various cell lines is shown in Supplementary Fig. S1.

Figure 1.

Polarized F-actin is a marker of colonic brush borders in vivo and labels lumens in vitro. A, confocal images of normal human colon cryosections labeled with TRITC-phalloidin (red) and DAPI (blue) to visualize F-actin and nuclei, respectively. Left image shows longitudinal section; right image shows transverse section. Scale bar, 50 ΞΌm. B, z-stack gallery of a SW1222 colony labeled with F-actin (red), DAPI (blue), and cytokeratin 20 (green). Numbers represent distance in z-axis from top. Scale bar, 10 ΞΌm. C, single cells from various cell lines were grown in Matrigel for 6 days. Colonies were labeled with TRITC-phalloidin (red) to visualize F-actin and DAPI to visualize nuclei. Bar, 10 ΞΌm.

Figure 1.

Polarized F-actin is a marker of colonic brush borders in vivo and labels lumens in vitro. A, confocal images of normal human colon cryosections labeled with TRITC-phalloidin (red) and DAPI (blue) to visualize F-actin and nuclei, respectively. Left image shows longitudinal section; right image shows transverse section. Scale bar, 50 ΞΌm. B, z-stack gallery of a SW1222 colony labeled with F-actin (red), DAPI (blue), and cytokeratin 20 (green). Numbers represent distance in z-axis from top. Scale bar, 10 ΞΌm. C, single cells from various cell lines were grown in Matrigel for 6 days. Colonies were labeled with TRITC-phalloidin (red) to visualize F-actin and DAPI to visualize nuclei. Bar, 10 ΞΌm.

Close modal

We next examined whether other intestinal brush border markers were also present in lumens. Villin, an F-actin–interacting protein associated with intestinal differentiation, was strongly enriched in lumens and its staining overlapped that for F-actin (Fig. 2A). HCT116 and DLD1 did not express villin. In HT29, villin was generally expressed but it was not polarized and did not overlap F-actin. Ezrin, another F-actin–interacting protein enriched in the brush border, was also strongly enriched in lumens where it was clearly located at the apical cell membrane (Fig. 2B). Ezrin was only weakly present at non-apical plasma membranes. Ezrin was strongly expressed in HCT116 in a nonpolarized manner.

Figure 2.

Lumens represent enterocyte brush border differentiation but also contain secretory lineages. Single cells from specified cell lines were embedded in Matrigel and grown for 10 days. Colonies were fixed and labeled with phalloidin (red), DAPI (blue), and immunolabeled with anti-villin (A) or anti-ezrin (B). C and D, single SW1222 or C80 cells were grown in Matrigel for 10 days before immunolabeling with either anti-Muc2 antibodies (C) or anti-CEA (D). F-actin was colabeled with TRITC-phalloidin. Scale bar, 20 ΞΌm.

Figure 2.

Lumens represent enterocyte brush border differentiation but also contain secretory lineages. Single cells from specified cell lines were embedded in Matrigel and grown for 10 days. Colonies were fixed and labeled with phalloidin (red), DAPI (blue), and immunolabeled with anti-villin (A) or anti-ezrin (B). C and D, single SW1222 or C80 cells were grown in Matrigel for 10 days before immunolabeling with either anti-Muc2 antibodies (C) or anti-CEA (D). F-actin was colabeled with TRITC-phalloidin. Scale bar, 20 ΞΌm.

Close modal

The brush border enzyme CD26/dipeptidyl peptidase-4 (DPPIV; ref. 7) and the brush border–enriched tight junction marker Zo-1(23) were also polarized and enriched in lumen apical membranes (Supplementary Fig. S2A and S2B). In contrast, non-lumen–forming cell lines failed to show organized expression of either marker.

We next examined whether formation of F-actin–enriched lumens in colonies derived from a single-cell precluded the presence of secretory lineages. Co-staining of SW1222 colonies with phalloidin and anti-MUC2, a goblet cell–specific mucin (Fig. 2C), showed that goblet cells were present in F-actin colonies grown from single cells. We also examined expression of carcinoembryonic antigen (CEA); a membrane-associated and secreted glycoprotein commonly used as a cancer marker. CEA was present almost exclusively in the lumens of C80 and SW1222 colonies (Fig. 2D).

The above evidence indicates that lumen formation in colonies derived from single colorectal cells represents brush border differentiation but does not preclude differentiation along secretory lineages.

Cancer stem cell enrichment

To determine whether stem cell–enriched cell fractions gave rise to increased frequencies of lumens with F-actin/brush border differentiation, we FACS-sorted SW1222 and LS180 cells according to expression of cancer stem cell markers CD24 and CD44, as described previously (14). Single sorted cells were then plated in Matrigel and grown into colonies for 2 weeks. In Fig. 3A and B, the top 5% of CD24/44 (CD24 and CD44 high)-expressing SW1222 cells gave rise to significantly more large colonies with F-actin–labeled lumens than unsorted/bulk cell populations and significantly fewer small colonies with no lumens. The bottom 5% CD24/44-expressing cells (CD24 and CD44 low) cells showed very low clonogenicity, and resulting colonies were virtually all-small with absent or poorly developed F-actin foci. Similar results were obtained with LS180 cells (Fig. 3C and D). Thus, stem cell–enriched populations give rise to a larger proportion of larger colonies with increased brush border/enterocyte differentiation and a much smaller proportion of small non-lumen–forming colonies.

Figure 3.

F-actin can be used to identify lumen colonies in stem cell–enriched populations. A and C, scattergram plots showing the gating of the extreme 5% of CD44/CD24 high- and low-expressing cells populations of SW1222 or LS180 cells sorted by FACS. The gates were set so that 5% of the total positive populations were selected in both the top right quadrant (CD44/CD24 high) and the bottom left quadrant (CD44/CD24 low). B and D, isolated CD24/44-high and -low SW1222 or LS180 cells were grown for 2 weeks in Matrigel and stained with TRITC-phalloidin. Colonies were manually characterized into the following categories: large colony, diameter larger than 40 ΞΌm; lumen colony, F-actin foci larger than 10 ΞΌm. Bar graphs show total counts of different colony classifications from 3 replicate wells for each gating strategy. B, for SW1222 colonies, Fisher exact test showed that single cells with high CD24/44 expression gave rise to a significantly higher proportion of large lumen colonies at the expense of small non-lumen colonies when compared with unsorted cells (P = 0.0007). P values represent Fisher exact test of significance of the change in the proportion of lumen to non-lumen colonies and indicate a significant shift towards lumen colony formation for cells enriched for CD24/44 expression. Ο‡2 2 Γ— 4 table also showed that single cells with high CD24/44 expression gave rise to a significantly higher proportion of lumen colonies to non-lumen colonies irrespective of colony size (P = 0.0003), compared with unsorted cells. The CD24/44 low-expressing cells showed a much lower clonogenicity than either unsorted or CD24/44-high cells. Although there was a clear trend towards an increase in the ratio of small non-lumen colonies to large lumen colonies, the very small number of colonies precluded statistical analysis. D, for LS180, the Fisher exact test showed a similar significant increase in the proportion of large lumen to small non-lumen colonies derived from cells sorted for high CD24/44 expression, compared with CD24/44 low-expressing cells (P = 0.0002), and a Ο‡2 2 Γ— 4 table showed a significant difference between CD24/44-high and -low when comparing all lumen colonies to non-lumen colonies irrespective of colony size (P = 0.0001).

Figure 3.

F-actin can be used to identify lumen colonies in stem cell–enriched populations. A and C, scattergram plots showing the gating of the extreme 5% of CD44/CD24 high- and low-expressing cells populations of SW1222 or LS180 cells sorted by FACS. The gates were set so that 5% of the total positive populations were selected in both the top right quadrant (CD44/CD24 high) and the bottom left quadrant (CD44/CD24 low). B and D, isolated CD24/44-high and -low SW1222 or LS180 cells were grown for 2 weeks in Matrigel and stained with TRITC-phalloidin. Colonies were manually characterized into the following categories: large colony, diameter larger than 40 ΞΌm; lumen colony, F-actin foci larger than 10 ΞΌm. Bar graphs show total counts of different colony classifications from 3 replicate wells for each gating strategy. B, for SW1222 colonies, Fisher exact test showed that single cells with high CD24/44 expression gave rise to a significantly higher proportion of large lumen colonies at the expense of small non-lumen colonies when compared with unsorted cells (P = 0.0007). P values represent Fisher exact test of significance of the change in the proportion of lumen to non-lumen colonies and indicate a significant shift towards lumen colony formation for cells enriched for CD24/44 expression. Ο‡2 2 Γ— 4 table also showed that single cells with high CD24/44 expression gave rise to a significantly higher proportion of lumen colonies to non-lumen colonies irrespective of colony size (P = 0.0003), compared with unsorted cells. The CD24/44 low-expressing cells showed a much lower clonogenicity than either unsorted or CD24/44-high cells. Although there was a clear trend towards an increase in the ratio of small non-lumen colonies to large lumen colonies, the very small number of colonies precluded statistical analysis. D, for LS180, the Fisher exact test showed a similar significant increase in the proportion of large lumen to small non-lumen colonies derived from cells sorted for high CD24/44 expression, compared with CD24/44 low-expressing cells (P = 0.0002), and a Ο‡2 2 Γ— 4 table showed a significant difference between CD24/44-high and -low when comparing all lumen colonies to non-lumen colonies irrespective of colony size (P = 0.0001).

Close modal

Effect of hypoxia and Hif1a stabilization on differentiation and CDX1 expression

Lumen formation can be inhibited by long-term hypoxia, accompanied by a downregulation of CDX1 in an Hif1-dependent manner (15). To determine whether F-actin was similarly downregulated by hypoxia, we compared LS180 and SW1222 cells grown in Matrigel for 14 days in either normoxia or hypoxia (1% O2). Figure 4A shows representative images of F-actin labeling of the colonies under each condition. Hypoxic colonies were smaller and F-actin focus size and intensity were reduced, compared with normoxia. The number of F-actin foci appeared to be similar. Quantification of colonies indicated a significant drop in the proportion of colonies with F-actin–labeled lumens and an increase of colonies without well-defined lumens (Fig. 4B). These results show that hypoxia inhibits brush border development in colon cancer stem cells.

Figure 4.

A, hypoxia and Hif activation reduces F-actin foci formation. A, single cells of either LS180 or SW1222 were grown in Matrigel under normal 20% oxygen or 1% oxygen for 14 days. The colonies were then labeled with phalloidin (red). Scale bar, 50 ΞΌm. B, colony counts of the same experiment showing total number of colonies of LS180 cells in defined categories, from 3 replicates. Small colony defined as any colony with a radius smaller than 250 ΞΌm; lumen colony, colony with 1 or more F-actin foci larger than 10 ΞΌm. Forty colonies counted from each of 3 wells for hypoxia and normoxia. Hypoxia induced a significant difference in the proportion of lumen to non-lumen colonies (P = 0.0235, Fisher exact test 2 Γ— 2 table), compared with colonies grown in 20% oxygen. P value represents a Fisher exact test of significance of the change in the proportion of lumen to non-lumen colonies and indicates a significant shift towards non-lumen colony formation in the presence of DMOG. C, lumen F-actin formation is inhibited by DMOG. Single cells were grown in Matrigel for 1 week in the presence of 2 mmol/L DMOG or vehicle control (DMSO). Colonies were then fixed and F-actin labeled with phalloidin. D, graph showing total colonies with large or small lumens. A total of 40 colonies were counted from each of 3 wells for each condition. Lumen colony, colony with 1 or more F-actin foci larger than 15 ΞΌm. For LS180 and C80, DMOG significantly inhibited lumen formation (P ≀ 0.0001, 2 Γ— 2 tables, Fisher exact test of proportion of lumen to non-lumen colonies), compared with colonies grown in DMSO control. SW1222 showed a nonstatistically significant trend towards reduced lumens by DMOG (P = 0.157, Fisher exact test).

Figure 4.

A, hypoxia and Hif activation reduces F-actin foci formation. A, single cells of either LS180 or SW1222 were grown in Matrigel under normal 20% oxygen or 1% oxygen for 14 days. The colonies were then labeled with phalloidin (red). Scale bar, 50 ΞΌm. B, colony counts of the same experiment showing total number of colonies of LS180 cells in defined categories, from 3 replicates. Small colony defined as any colony with a radius smaller than 250 ΞΌm; lumen colony, colony with 1 or more F-actin foci larger than 10 ΞΌm. Forty colonies counted from each of 3 wells for hypoxia and normoxia. Hypoxia induced a significant difference in the proportion of lumen to non-lumen colonies (P = 0.0235, Fisher exact test 2 Γ— 2 table), compared with colonies grown in 20% oxygen. P value represents a Fisher exact test of significance of the change in the proportion of lumen to non-lumen colonies and indicates a significant shift towards non-lumen colony formation in the presence of DMOG. C, lumen F-actin formation is inhibited by DMOG. Single cells were grown in Matrigel for 1 week in the presence of 2 mmol/L DMOG or vehicle control (DMSO). Colonies were then fixed and F-actin labeled with phalloidin. D, graph showing total colonies with large or small lumens. A total of 40 colonies were counted from each of 3 wells for each condition. Lumen colony, colony with 1 or more F-actin foci larger than 15 ΞΌm. For LS180 and C80, DMOG significantly inhibited lumen formation (P ≀ 0.0001, 2 Γ— 2 tables, Fisher exact test of proportion of lumen to non-lumen colonies), compared with colonies grown in DMSO control. SW1222 showed a nonstatistically significant trend towards reduced lumens by DMOG (P = 0.157, Fisher exact test).

Close modal

To determine whether inhibition of lumen formation was dependent on Hif transcription factors, we cultured colonies in the presence or absence of the prolyl-hydroxylase inhibitor DMOG. Prolyl-hydroxylases hydroxylate Hif transcription factors, marking them for degradation and so DMOG interferes with this process leading to accumulation of Hif1Ξ± (24). We first confirmed that DMOG treatment induced expression of Hif1 and Hif1-responsive genes in cell lines by immunolabeling for Glut1 and carbonic anhydrase IX (CAIX) and immunoblotting for Hif1 (Supplementary Fig. S3A–S3C). We then treated colonies grown from single cells in Matrigel for 7 days and with DMOG for a further 5 days. The effect on F-actin labeling was determined with TRITC-phalloidin. Figure 4C shows that DMOG inhibited F-actin lumen formation in C80 and LS180 and SW1222. Large F-actin lumens were inhibited and, instead, many colonies exhibited small F-actin foci. There was no obvious effect on colony size. Quantitation of the colony morphologies showed a significant reduction in the proportion of colonies with large, well-defined F-actin lumens and an increase in the proportion of colonies with small F-actin foci (Fig. 4D).

We next determined whether DMOG treatment induced downregulation of CDX1 and its target cytokeratin 20 (16). Immunolabeling of DMOG-treated colonies and vehicle-treated controls showed a marked downregulation of CDX1 in response to DMOG, although cytokeratin 20 expression remained virtually unchanged (Fig. 5A and B). We also looked at the effect of DMOG on goblet cell lineage differentiation. Immunolabeling colonies with an anti-goblet cell antibody indicated that DMOG suppressed goblet cell differentiation in both the C80 and LS180 cell lines (Fig. 5C and D).

Figure 5.

Prolyl-hydroxylase inhibition downregulates CDX1 and suppresses goblet cell differentiation. A and B, 10-day-old colonies grown in Matrigel from single cells of either C80 (A) or SW1222 (B) treated with 2 mmol/L DMOG or DMSO for 4 days. Colonies were imaged by phase contrast and coimmunolabeled with anti-CDX1 and anti-cytokeratin 20 (CK20). Bar, 20 ΞΌm. C, single LS180 or C80 cells were plated on plastic at 500 cells per well in a 96-well plate and allowed to grow for 1 week in total. On day three, 2 mmol/L DMOG or DMSO (vehicle control) was added to the wells. The cells were fixed and labeled with DAPI (blue), phalloidin (red), and anti-goblet cell mucin (PR5D5; green). Bar, 20 ΞΌm. D, bar graph showing quantification of goblet cell–positive colonies in vehicle- or DMOG-treated wells. Of 120 colonies looked at, from a total of 3 wells for each condition, the number that contained goblet cells and the number that did not are given in the bar graph. DMOG caused a significant decrease in the proportion of goblet cell containing colonies compared with DMSO controls for both C80 and LS180 (P = 0.0030, P = 0.0001, respectively, using Fisher exact test for the relevant 2 Γ— 2 table).

Figure 5.

Prolyl-hydroxylase inhibition downregulates CDX1 and suppresses goblet cell differentiation. A and B, 10-day-old colonies grown in Matrigel from single cells of either C80 (A) or SW1222 (B) treated with 2 mmol/L DMOG or DMSO for 4 days. Colonies were imaged by phase contrast and coimmunolabeled with anti-CDX1 and anti-cytokeratin 20 (CK20). Bar, 20 ΞΌm. C, single LS180 or C80 cells were plated on plastic at 500 cells per well in a 96-well plate and allowed to grow for 1 week in total. On day three, 2 mmol/L DMOG or DMSO (vehicle control) was added to the wells. The cells were fixed and labeled with DAPI (blue), phalloidin (red), and anti-goblet cell mucin (PR5D5; green). Bar, 20 ΞΌm. D, bar graph showing quantification of goblet cell–positive colonies in vehicle- or DMOG-treated wells. Of 120 colonies looked at, from a total of 3 wells for each condition, the number that contained goblet cells and the number that did not are given in the bar graph. DMOG caused a significant decrease in the proportion of goblet cell containing colonies compared with DMSO controls for both C80 and LS180 (P = 0.0030, P = 0.0001, respectively, using Fisher exact test for the relevant 2 Γ— 2 table).

Close modal

Thus, DMOG treatment leads to rapid loss of CDX1 expression but not cytokeratin 20 and inhibits both brush border/enterocyte and secretory goblet cell differentiation.

To confirm that CDX1 downregulation could impair F-actin lumen formation, we used an LS174T-derivative cell line LS174T-CDX1 with stable expression of a construct encoding a short hairpin RNA targeting CDX1, in which CDX1 was downregulated by RNA interference (RNAi; ref. 16). An LS174T line, LS174T-vector, expressing an empty vector was used as a control, as these cells maintain normal CDX1 levels. Single cells from either LS174T siCDX1 or LS174T-vector were embedded in Matrigel and allowed to grow into colonies for 10 days, before fixation and phalloidin labeling to visualize F-actin. Figure 6A and B shows that low CDX1 significantly reduced the proportion of colonies with well-defined F-actin lumens. We also manually quantified the total number of lumens per colony in the same data and found that CDX1 knockdown also reduced this number, that is, CDX1 knockdown inhibited lumens per colony (Fig. 6C). We next took advantage of the CDX1 RNAi F-actin data to validate a semiautomatic approach to lumen formation that would facilitate high-throughput analysis of stem cell differentiation using public ImageJ software. Figure 6D, i, shows a whole well-composite image of an entire well from a 96-well plate that contained F-actin–labeled SW1222 colonies grown in 3D conditions. ImageJ software was then used to automatically recognize and analyze the colonies (Fig. 6D, ii). The software was able to detect lumens accurately in a well-differentiated colony and was able to distinguish that colony from a poorly differentiated colony lacking F-actin foci (Fig. 6D, iii and iv, respectively). To validate this approach, we used our semiautomatic method to analyze the LS174T-vector and LS174T-CDX1 colonies manually counted in Fig. 6C and found a good correlation between the automated and manual analyses (Fig. 6E). Thus, our approach enables tissue plate wells to be automatically scanned and analyzed for changes in lumen formation.

Figure 6.

CDX1 knockdown reduces F-actin lumen formation in LS174T colonies. A, F-actin labeling of LS174T vector and LS174T CDX1 RNAi cultured from single cells in Matrigel for 1.5 weeks. B, bar chart showing total counts of 200 colonies in each of three replicate wells, categorized into lumen or non-lumen colonies. Colonies were defined as non-lumen if they contained no F-actin foci with a diameter above 8 ΞΌm. Knockdown of CDX1 caused a significant reduction in the proportion of lumen colonies, compared with vector control colonies (P ≀ 0.0001, Fisher exact test). C, manual counts of averaged numbers of F-actin lumens per colony of LS174T-vector or LS174T-CDX1 colonies from the same experiment as used for B. A total of 120 colonies were analyzed (40 each from 3 replicates). The number of lumens in LS174T-CDX1 was significantly reduced compared with the vector control (Student t test, P ≀ 0.0001). Error bar, SD. D, semiautomated analysis of F-actin lumen formation using ImageJ. D, i, composite image of an entire well from a 96-well plate containing F-actin–labeled SW1222 colonies grown in Matrigel from single cells for 10 days. D, ii, masked (outlined) colonies automatically identified by the ImageJ algorithm. D, iii, a well-differentiated lumen colony is shown with 7 automatically detected F-actin foci. D, iv, a small poorly differentiated colony with only 1 F-actin focus automatically detected. E, automatic F-actin lumen counting was applied to the LS174T data used for C and a significantly reduced number of F-actin lumens per colony in the LS174T-CDX1 knockdown colonies was detected (Student t test, P = 0.001).

Figure 6.

CDX1 knockdown reduces F-actin lumen formation in LS174T colonies. A, F-actin labeling of LS174T vector and LS174T CDX1 RNAi cultured from single cells in Matrigel for 1.5 weeks. B, bar chart showing total counts of 200 colonies in each of three replicate wells, categorized into lumen or non-lumen colonies. Colonies were defined as non-lumen if they contained no F-actin foci with a diameter above 8 ΞΌm. Knockdown of CDX1 caused a significant reduction in the proportion of lumen colonies, compared with vector control colonies (P ≀ 0.0001, Fisher exact test). C, manual counts of averaged numbers of F-actin lumens per colony of LS174T-vector or LS174T-CDX1 colonies from the same experiment as used for B. A total of 120 colonies were analyzed (40 each from 3 replicates). The number of lumens in LS174T-CDX1 was significantly reduced compared with the vector control (Student t test, P ≀ 0.0001). Error bar, SD. D, semiautomated analysis of F-actin lumen formation using ImageJ. D, i, composite image of an entire well from a 96-well plate containing F-actin–labeled SW1222 colonies grown in Matrigel from single cells for 10 days. D, ii, masked (outlined) colonies automatically identified by the ImageJ algorithm. D, iii, a well-differentiated lumen colony is shown with 7 automatically detected F-actin foci. D, iv, a small poorly differentiated colony with only 1 F-actin focus automatically detected. E, automatic F-actin lumen counting was applied to the LS174T data used for C and a significantly reduced number of F-actin lumens per colony in the LS174T-CDX1 knockdown colonies was detected (Student t test, P = 0.001).

Close modal

Lumen formation is a feature of tumors in vivo

To determine whether lumen formation had relevance to tumors in vivo, we first examined actin and ezrin polarization in murine xenografts derived from the injection of HT29, HCT116, or SW1222 cells into the flanks of NOD/SCID mice. Mice were sacrificed after 1 month and resulting tumors removed and processed for formalin-fixed, paraffin-embedded (FFPE) tissue sections, which were stained by hematoxylin and eosin or immunolabeled with anti-actin/ezrin (Fig. 7A). Hematoxylin and eosin staining of these xenograft tumors showed poorly differentiated high-grade tumor morphology for HCT116 and HT29 tumors that lacked lumens or actin/ezrin polarization. SW1222 tumors exhibited a well-differentiated phenotype with numerous lumens surrounded by polarized cells. Actin and ezrin immunolabeling showed that neither HT29 nor HCT116 exhibited polarization of these markers, whereas SW1222 showed marked luminal polarization. Thus in vivo, cell lines grow in a manner similar to in vitro growth in Matrigel.

Figure 7.

Lumen formation is a feature of tumors in vivo, and human primary low- and high-grade tumors show differences in ezrin/actin and CEA polarization. A, lumen-forming and non-lumen–forming cell lines form well-differentiated (low-grade) and poorly differentiated (high-grade) tumors, respectively, as xenografts in vivo. Hematoxylin and eosin or anti-actin/ezrin immunostaining of FFPE sections from murine xenograft tumors derived from SW1222, HCT116, or HT29 cells injected into NOD/SCID. Inset shows magnification. B, typical hematoxylin and eosin staining of FFPE tissue sections derived from a moderately differentiated (low-grade) primary colorectal tumor from a patient, with numerous neoplastic glands visible (left), and a poorly differentiated (high-grade) human colorectal tumor, with no glands (right). C and D, anti-actin/anti-ezrin or anti-CEA/villin labeling of FFPE tissue sections derived from the same patient derived high- and low-grade tumors shown in B. E, characterization of lumen formation in primary cultures derived from a low-grade tumor. Left, phase contrast image of typical primary cultured spheroid cancer cell colony derived from the same low-grade human colorectal tumor shown in B–D. Right, F-actin labeling of these primary colonies following growth in Matrigel for 7 days. F, summary of lumen formation efficiency in Matrigel of a panel of 6 primary colorectal cancer cultures derived from 6 separate patients.

Figure 7.

Lumen formation is a feature of tumors in vivo, and human primary low- and high-grade tumors show differences in ezrin/actin and CEA polarization. A, lumen-forming and non-lumen–forming cell lines form well-differentiated (low-grade) and poorly differentiated (high-grade) tumors, respectively, as xenografts in vivo. Hematoxylin and eosin or anti-actin/ezrin immunostaining of FFPE sections from murine xenograft tumors derived from SW1222, HCT116, or HT29 cells injected into NOD/SCID. Inset shows magnification. B, typical hematoxylin and eosin staining of FFPE tissue sections derived from a moderately differentiated (low-grade) primary colorectal tumor from a patient, with numerous neoplastic glands visible (left), and a poorly differentiated (high-grade) human colorectal tumor, with no glands (right). C and D, anti-actin/anti-ezrin or anti-CEA/villin labeling of FFPE tissue sections derived from the same patient derived high- and low-grade tumors shown in B. E, characterization of lumen formation in primary cultures derived from a low-grade tumor. Left, phase contrast image of typical primary cultured spheroid cancer cell colony derived from the same low-grade human colorectal tumor shown in B–D. Right, F-actin labeling of these primary colonies following growth in Matrigel for 7 days. F, summary of lumen formation efficiency in Matrigel of a panel of 6 primary colorectal cancer cultures derived from 6 separate patients.

Close modal

Because similar glandular structures (neoplastic glands) are a feature of human colorectal adenocarcinomas and are used by pathologists to grade tumors, we examined whether they labeled with the same markers as in vitro lumens. We compared 2 colorectal tumors that had been graded previously as either moderately differentiated (low-grade) or poorly differentiated (high-grade). Hematoxylin and eosin staining of tissue sections from the low-grade tumor confirmed the presence of numerous neoplastic glands, which were absent in the poorly differentiated tumors (Fig. 7B). Immunolabeling of sections from the same tumors indicated that lumens in the low-grade tumor were composed of polarized cells expressing actin and ezrin on the apical membranes, whereas no polarization of these markers was observed in the poorly differentiated tumors (Fig. 7C). Typical of tumors, cellular debris was present in the lumens. Actin/ezrin-positive cells present in the stromal surrounding the glands were probably myofibroblasts (Supplementary Fig. S4). Lumen glands were strongly positive for polarized CEA, whereas CEA was nonpolarized and more cell membrane associated in the high-grade tumor (Fig. 7D). Thus, at least 3 markers of in vitro lumens also mark in vivo glands, indicating that in vitro lumens from cell lines are virtually identical in nature to lumens in primary tumors.

To determine whether lumen formation was a feature of primary tumor cells in vitro, we derived a primary tumor cell culture from the same low-grade tumor used in Fig. 7B–D. When grown under conditions that promote stem cell populations (25–28), these primary cells grew as well-organized spheroids (Fig. 7E, left). Following a week-long suspension in Matrigel, the colonies formed large central lumens marked by F-actin, (Fig. 7E, right). Three of 6 primary cultures derived from separate patients were capable of lumen formation (Table 1). Thus, we conclude that lumen formation represents the same process in vivo as it does in vitro and can therefore be used to identify stem cell populations in primary tumor–derived cultures.

Single stem cells from colorectal cancer lines, when grown under 3D conditions, can differentiate into polarized structures with remarkable structural similarity to normal intestinal crypts. Lumens express many characteristics of normal intestinal brush borders, including microvilli structural proteins (polarized F-actin, villin, ezrin), enzymatic activity (polarized DPPIV), cell surface and secreted glycoprotein (CEA), and tight junctions (TJP1/Zo-1). Thus, colorectal cancer stem cells that produce lumen colonies can be considered to be differentiating predominantly along the enterocyte lineage, the most common colonic lineage. Nevertheless, goblet cells were also often present in colonies with F-actin lumens that had been grown from single cells, illustrating their multipotent nature. Importantly, we also show that several of these markers of in vitro lumen brush border formation are also expressed in a polarized manner by neoplastic glands present in primary tumors, which are similarly composed of polarized cells surrounding a lumen. Thus, it is most likely that these neoplastic lumens also indicate stem cell differentiation in the same way as their in vitro counterparts. This is supported by our observation that primary tumor spheroid cultures derived from moderately/well-differentiated tumors could also form polarized lumen colonies when transferred into Matrigel.

Unlike lumen colonies and well-differentiated tumors, non-lumen cell lines, such as HCT116, totally lacked cellular polarization of several polarity markers such as villin, a situation mirrored in vivo by high-grade tumors. Thus, cell lines can also be separated into high or low β€œgrades” based on their polarization in Matrigel.

One possible explanation for this is that these cells lack expression of CDX1 (16), and this may disrupt their ability to efficiently polarize and differentiate. Ectopic re-expression of CDX1 in HCT116 and DLD1 can lead to the formation of primitive lumens in Matrigel (14, 15). Correspondingly, CDX1 and the related transcription factor CDX2 have been shown to be important for polarization of intestinal cells (29, 30). A lack of polarization and differentiation will lead to an increase in the population of stem cells and indeed we have found previously that HCT116 does not contain subpopulations of cells with different tumor-forming capacity, unlike the lumen-forming lines (14). Similarly, high-grade tumors have a significantly worse clinical prognosis and exhibit a more β€œstem-like” transcription profile (31, 32).

Tumor formation is frequently associated with hypoxia, and our previous studies have shown that prolonged hypoxia inhibits lumen formation (15). Correspondingly, hypoxia leads to a failure to polarize F- actin correctly, as did treatment of colonies with the Hif1Ξ± activator DMOG. DMOG also inhibited CDX1 expression and goblet cell differentiation, although cytokeratin 20 expression was not affected, probably due to a difference in turnover rate between CDX1 and cytokeratin 20. As the CDX1 promoter contains a number of putative Hif-binding sites (unpublished data), it is possible that downregulation of CDX1 is directly induced by Hif-mediated transcriptional inhibition. It seems likely that the loss of F-actin lumen expression in response to hypoxia/DMOG treatment is triggered by a loss of CDX1. This is supported by the observation that targeted knockdown of CDX1 in LS174T cells leads to a significant reduction in the size and intensity of F-actin–labeled lumens. However, we cannot rule out a role for other hypoxia-regulated transcription factors in differentiation suppression.

Specific cancer stem cell markers remain elusive. While markers such as LGR5 (33), ephrin type-B receptor 2 (EphB2; ref. 34), and Lrig1 (35) appear to mark well-defined stem cell populations in normal intestine, their expression in tumors does not appear to be entirely specific to cancer stem cells as low clonogenic cells are also marked, at least for EphB2 (31), and LGR5 (36, 37). F-actin lumen formation should therefore be useful for defining the cutoff in expression of markers such as LGR5 and EphB2 that define cancer stem cell populations from non-stem cells. The ability of single cancer stem cells to grow into differentiated lumen colonies provides an unambiguous tool to estimate stem cell populations. Our approach therefore allows for the functional characterization of CSCs and their differentiation, independent of expression of stem cell surface markers. Furthermore, the F-actin lumen assay can be adapted for high-throughput applications, which should be useful for identifying drugs that can target stem cells.

In summary, we have characterized the formation of lumen colonies from cell line–derived single cells to represent brush border/enterocyte differentiation and show the extent of heterogeneity between cell lines and primary tumor cultures. Thus, the cancer stem cells from the cell lines, in their ability to form lumen colonies, have a striking similarity to normal intestinal stem cells, which also divide to form the colonic crypt.

No potential conflicts of interest were disclosed.

Conception and design: N. Ashley, T. Yeung, W.F. Bodmer

Development of methodology: N. Ashley

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): N. Ashley, T. Yeung

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): N. Ashley, W.F. Bodmer

Writing, review, and/or revision of the manuscript: N. Ashley, W.F. Bodmer

Study supervision: W.F. Bodmer

The authors thank Kevin Clarke Jenny Wilding and Steve Taylor for technical and conceptual input.

This work was funded, in part, by the EpiDiaCan FP7 European Union Grant no: FP7-241783.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Potten
β€ˆ
CS
,
Booth
β€ˆ
C
,
Pritchard
β€ˆ
DM
.β€ˆ
The intestinal epithelial stem cell: the mucosal governor
.
Int J Exp Pathol
β€ˆ
1997
;
78
:
219
–
43
.
2.
Lacroix
β€ˆ
B
,
Kedinger
β€ˆ
M
,
Simon-Assmann
β€ˆ
P
,
Rousset
β€ˆ
M
,
Zweibaum
β€ˆ
A
,
Haffen
β€ˆ
K
.β€ˆ
Developmental pattern of brush border enzymes in the human fetal colon. Correlation with some morphogenetic events
.
Early Hum Dev
β€ˆ
1984
;
9
:
95
–
103
.
3.
Holmes
β€ˆ
R
.β€ˆ
The intestinal brush border
.
Gut
β€ˆ
1971
;
12
:
668
–
77
.
4.
Holmes
β€ˆ
R
,
Lobley
β€ˆ
RW
.β€ˆ
Intestinal brush border revisited
.
Gut
β€ˆ
1989
;
30
:
1667
–
78
.
5.
Carboni
β€ˆ
JM
,
Howe
β€ˆ
CL
,
West
β€ˆ
AB
,
Barwick
β€ˆ
KW
,
Mooseker
β€ˆ
MS
,
Morrow
β€ˆ
JS
.β€ˆ
Characterization of intestinal brush border cytoskeletal proteins of normal and neoplastic human epithelial cells. A comparison with the avian brush border
.
Am J Pathol
β€ˆ
1987
;
129
:
589
–
600
.
6.
West
β€ˆ
AB
,
Isaac
β€ˆ
CA
,
Carboni
β€ˆ
JM
,
Morrow
β€ˆ
JS
,
Mooseker
β€ˆ
MS
,
Barwick
β€ˆ
KW
.β€ˆ
Localization of villin, a cytoskeletal protein specific to microvilli, in human ileum and colon and in colonic neoplasms
.
Gastroenterology
β€ˆ
1988
;
94
:
343
–
52
.
7.
Zweibaum
β€ˆ
A
,
Hauri
β€ˆ
HP
,
Sterchi
β€ˆ
E
,
Chantret
β€ˆ
I
,
Haffen
β€ˆ
K
,
Bamat
β€ˆ
J
, et al
β€ˆ
Immunohistological evidence, obtained with monoclonal antibodies, of small intestinal brush border hydrolases in human colon cancers and foetal colons
.
Int J Cancer
β€ˆ
1984
;
34
:
591
–
8
.
8.
Hamilton
β€ˆ
SR
,
Aaltonen
β€ˆ
LA
.β€ˆ
World Health Organization classification of tumors: pathology and genetics of tumors of the digestive system
.
IARC Press
;
Lyon, France
;β€ˆ
2000
.
9.
Hinoi
β€ˆ
T
,
Tani
β€ˆ
M
,
Lucas
β€ˆ
PC
,
Caca
β€ˆ
K
,
Dunn
β€ˆ
RL
,
Macri
β€ˆ
E
, et al
β€ˆ
Loss of CDX2 expression and microsatellite instability are prominent features of large cell minimally differentiated carcinomas of the colon
.
Am J Pathol
β€ˆ
2001
;
159
:
2239
–
48
.
10.
O'Connell
β€ˆ
JB
,
Maggard
β€ˆ
MA
,
Ko
β€ˆ
CY
.β€ˆ
Colon cancer survival rates with the new American Joint Committee on Cancer sixth edition staging
.
J Natl Cancer Inst
β€ˆ
2004
;
96
:
1420
–
5
.
11.
Ashley
β€ˆ
N
.β€ˆ
Regulation of intestinal cancer stem cells
.
Cancer Lett.
β€ˆ
2012 Apr 27
.
[Epub ahead of print]
.
12.
Richman
β€ˆ
P
,
Bodmer
β€ˆ
W
.β€ˆ
Control of differentiation in human colorectal carcinoma cell lines: epithelial-mesenchymal interactions
.
Pathol J
β€ˆ
1988
;
156
:
197
–
211
.
13.
Del Buono
β€ˆ
R
,
Pignatelli
β€ˆ
M
,
Bodmer
β€ˆ
WF
,
Wright
β€ˆ
NA
.β€ˆ
The role of the arginine-glycine-aspartic acid-directed cellular binding to type I collagen and rat mesenchymal cells in colorectal tumour differentiation
.
Differentiation
β€ˆ
1991
;
46
:
97
–
103
.
14.
Yeung
β€ˆ
TM
,
Gandhi
β€ˆ
SC
,
Wilding
β€ˆ
JL
,
Muschel
β€ˆ
R
,
Bodmer
β€ˆ
WF
.β€ˆ
Cancer stem cells from colorectal cancer-derived cell lines
.
Proc Natl Acad Sci U S A
β€ˆ
2010
;
107
:
3722
–
7
.
15.
Yeung
β€ˆ
TM
,
Gandhi
β€ˆ
SC
,
Bodmer
β€ˆ
WF
.β€ˆ
Hypoxia and lineage specification of cell line-derived colorectal cancer stem cells
.
Proc Natl Acad Sci U S A
β€ˆ
2011
;
108
:
4382
–
7
.
16.
Chan
β€ˆ
CW
,
Wong
β€ˆ
NA
,
Liu
β€ˆ
Y
,
Bicknell
β€ˆ
D
,
Turley
β€ˆ
H
,
Hollins
β€ˆ
L
, et al
β€ˆ
Gastrointestinal differentiation marker Cytokeratin 20 is regulated by homeobox gene CDX1
.
Proc Natl Acad Sci U S A
β€ˆ
2009
;
106
:
1936
–
41
.
17.
Wong
β€ˆ
NA
,
Britton
β€ˆ
MP
,
Choi
β€ˆ
GS
,
Stanton
β€ˆ
TK
,
Bicknell
β€ˆ
DC
,
Wilding
β€ˆ
JL
, et al
β€ˆ
Loss of CDX1 expression in colorectal carcinoma: promoter methylation, mutation, and loss of heterozygosity analyses of 37 cell lines
.
Proc Natl Acad Sci U S A
β€ˆ
2004
;
101
:
574
–
9
.
18.
Arango
β€ˆ
D
,
Al-Obaidi
β€ˆ
S
,
Williams
β€ˆ
DS
,
Dopeso
β€ˆ
H
,
Mazzolini
β€ˆ
R
,
Corner
β€ˆ
G
, et al
β€ˆ
Villin expression is frequently lost in poorly differentiated colon cancer
.
Am J Pathol
β€ˆ
2012
;
180
:
1509
–
21
.
19.
Staloch
β€ˆ
LJ
,
Divine
β€ˆ
JK
,
Witten
β€ˆ
JT
,
Simon
β€ˆ
TC
.β€ˆ
C/EBP and Cdx family factors regulate liver fatty acid binding protein transgene expression in the small intestinal epithelium
.
Biochim Biophys Acta
β€ˆ
2005
;
1731
:
168
–
78
.
20.
Ashley
β€ˆ
N
,
Poulton
β€ˆ
J
.β€ˆ
Anticancer DNA intercalators cause p53-dependent mitochondrial DNA nucleoid re-modelling
.
Oncogene
β€ˆ
2009
;
28
:
3880
–
91
.
21.
Pittman
β€ˆ
FE
,
Pittman
β€ˆ
JC
.β€ˆ
An electron microscopic study of epithelium of normal human sigmoid colonic mucosa
.
Gut
β€ˆ
1966
;
7
:
644
–
61
.
22.
Bretscher
β€ˆ
A
,
Weber
β€ˆ
K
.β€ˆ
Localization of actin and microfilament-associated proteins in the microvilli and terminal web of the intestinal brush border by immunofluorescence microscopy
.
J Cell Biol
β€ˆ
1978
;
79
:
839
–
45
.
23.
Polak-Charcon
β€ˆ
S
,
Shoham
β€ˆ
J
,
Ben-Shaul
β€ˆ
Y
.β€ˆ
Tight junctions in epithelial cells of human fetal hindgut, normal colon, and colon adenocarcinoma
.
J Natl Cancer Inst
β€ˆ
1980
;
65
:
53
–
62
.
24.
Chan
β€ˆ
DA
,
Sutphin
β€ˆ
PD
,
Denko
β€ˆ
NC
,
Giaccia
β€ˆ
AJ
.β€ˆ
Role of prolyl hydroxylation in oncogenically stabilized hypoxia-inducible factor-1alpha
.
J Biol Chem
β€ˆ
2002
;
277
:
40112
–
7
.
25.
Vermeulen
β€ˆ
L
,
Todaro
β€ˆ
M
,
de Sousa Mello
β€ˆ
F
,
Sprick
β€ˆ
MR
,
Kemper
β€ˆ
K
,
Perez
β€ˆ
Alea M
, et al
β€ˆ
Single-cell cloning of colon cancer stem cells reveals a multi-lineage differentiation capacity
.
Proc Natl Acad Sci U S A
β€ˆ
2008
;
105
:
13427
–
32
.
26.
Emmink
β€ˆ
BL
,
Van Houdt
β€ˆ
WJ
,
Vries
β€ˆ
RG
,
Hoogwater
β€ˆ
FJ
,
Govaert
β€ˆ
KM
,
Verheem
β€ˆ
A
, et al
β€ˆ
Differentiated human colorectal cancer cells protect tumor-initiating cells from irinotecan
.
Gastroenterology
β€ˆ
2011
;
141
:
269
–
78
.
27.
Lombardo
β€ˆ
Y
,
Scopelliti
β€ˆ
A
,
Cammareri
β€ˆ
P
,
Todaro
β€ˆ
M
,
Iovino
β€ˆ
F
,
Ricci-Vitiani
β€ˆ
L
, et al
β€ˆ
Bone morphogenetic protein 4 induces differentiation of colorectal cancer stem cells and increases their response to chemotherapy in mice
.
Gastroenterology
β€ˆ
2011
;
140
:
297
–
309
.
28.
Odoux
β€ˆ
C
,
Fohrer
β€ˆ
H
,
Hoppo
β€ˆ
T
,
Guzik
β€ˆ
L
,
Stolz
β€ˆ
DB
,
Lewis
β€ˆ
DW
, et al
β€ˆ
A stochastic model for cancer stem cell origin in metastatic colon cancer
.
Cancer Res
β€ˆ
2008
;
68
:
6932
–
41
.
29.
Keller
β€ˆ
MS
,
Ezaki
β€ˆ
T
,
Guo
β€ˆ
RJ
,
Lynch
β€ˆ
JP
.β€ˆ
Cdx1 or Cdx2 expression activates E-cadherin-mediated cell-cell adhesion and compaction in human COLO 205 cells
.
Am J Physiol Gastrointest Liver Physiol
β€ˆ
2004
;
287
:
G104
–
114
.
30.
Gao
β€ˆ
N
,
Kaestner
β€ˆ
KH
.β€ˆ
Cdx2 regulates endo-lysosomal function and epithelial cell polarity
.
Genes Dev
β€ˆ
2010
;
24
:
1295
–
305
.
31.
Merlos-Suarez
β€ˆ
A
,
Barriga
β€ˆ
FM
,
Jung
β€ˆ
P
,
Iglesias
β€ˆ
M
,
Cespedes
β€ˆ
MV
,
Rossell
β€ˆ
D
, et al
β€ˆ
The intestinal stem cell signature identifies colorectal cancer stem cells and predicts disease relapse
.
Cell Stem Cell
β€ˆ
2011
;
8
:
511
–
24
.
32.
de Sousa
β€ˆ
E
,
Melo
β€ˆ
F
,
Colak
β€ˆ
S
,
Buikhuisen
β€ˆ
J
,
Koster
β€ˆ
J
,
Cameron
β€ˆ
K
,
de Jong
β€ˆ
JH
, et al
β€ˆ
Methylation of cancer-stem-cell-associated Wnt target genes predicts poor prognosis in colorectal cancer patients
.
Cell Stem Cell
β€ˆ
2011
;
9
:
476
–
85
.
33.
Barker
β€ˆ
N
,
van Es
β€ˆ
JH
,
Kuipers
β€ˆ
J
,
Kujala
β€ˆ
P
,
van den Born
β€ˆ
M
,
Cozijnsen
β€ˆ
M
, et al
β€ˆ
Identification of stem cells in small intestine and colon by marker gene Lgr5
.
Nature
β€ˆ
2007
;
449
:
1003
–
7
.
34.
Jung
β€ˆ
P
,
Sato
β€ˆ
T
,
Merlos-Suarez
β€ˆ
A
,
Barriga
β€ˆ
FM
,
Iglesias
β€ˆ
M
,
Rossell
β€ˆ
D
, et al
β€ˆ
Isolation and in vitro expansion of human colonic stem cells
.
Nat Med
β€ˆ
2011
;
17
:
1225
–
7
.
35.
Powell
β€ˆ
AE
,
Wang
β€ˆ
Y
,
Li
β€ˆ
Y
,
Poulin
β€ˆ
EJ
,
Means
β€ˆ
AL
,
Washington
β€ˆ
MK
, et al
β€ˆ
The pan-ErbB negative regulator Lrig1 is an intestinal stem cell marker that functions as a tumor suppressor
.
Cell
β€ˆ
2012
;
149
:
146
–
58
.
36.
Kemper
β€ˆ
K
,
Prasetyanti
β€ˆ
PR
,
De Lau
β€ˆ
W
,
Rodermond
β€ˆ
H
,
Clevers
β€ˆ
H
,
Medema
β€ˆ
JP
.β€ˆ
Monoclonal antibodies against Lgr5 identify human colorectal cancer stem cells
.
Stem Cells
β€ˆ
2012
;
30
:
2378
–
86
.
37.
Kobayashi
β€ˆ
S
,
Yamada-Okabe
β€ˆ
H
,
Suzuki
β€ˆ
M
,
Natori
β€ˆ
O
,
Kato
β€ˆ
A
,
Matsubara
β€ˆ
K
, et al
β€ˆ
LGR5-positive colon cancer stem cells interconvert with drug-resistant LGR5-negative cells and are capable of tumor reconstitution
.
Stem Cells
β€ˆ
2012
;
30
:
2631
–
44
.