https://orcid.org/0000-0002-8489-8262
Figures
Abstract
Background
CD44 and CD133 are stem cell markers in colorectal cancer (CRC). CD44 has distinctive isoforms with different oncological properties like total CD44 (CD44T) and variant CD44 (CD44V). Clinical significance of such markers remains elusive.
Methods
Sixty colon cancer were examined for CD44T/CD44V and CD133 at mRNA level in a quantitative PCR, and clarified for their association with clinicopathological factors.
Results
(1) Both CD44T and CD44V showed higher expression in primary colon tumors than in non-cancerous mucosas (p<0.0001), while CD133 was expressed even in non-cancerous mucosa and rather decreased in the tumors (p = 0.048). (2) CD44V expression was significantly associated with CD44T expression (R = 0.62, p<0.0001), while they were not correlated to CD133 at all in the primary tumors. (3) CD44V/CD44T expressions were significantly higher in right colon cancer than in left colon cancer (p = 0.035/p = 0.012, respectively), while CD133 expression were not (p = 0.20). (4) In primary tumors, unexpectedly, CD44V/CD44T/CD133 mRNA expressions were not correlated with aggressive phenotypes, but CD44V/CD44T rather significantly with less aggressive lymph node metastasis/distant metastasis (p = 0.040/p = 0.039, respectively). Moreover, both CD44V and CD133 expressions were significantly decreased in liver metastasis as compared to primary tumors (p = 0.0005 and p = 0.0006, respectively).
Citation: Kaida T, Fujiyama Y, Soeno T, Yokota M, Nakamoto S, Goto T, et al. (2023) Less demand on stem cell marker-positive cancer cells may characterize metastasis of colon cancer. PLoS ONE 18(4): e0277395. https://doi.org/10.1371/journal.pone.0277395
Editor: Alvaro Galli, CNR, ITALY
Received: April 17, 2022; Accepted: October 26, 2022; Published: April 25, 2023
Copyright: © 2023 Kaida et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the article and its Supporting Information files.
Funding: The authors received no specific funding for this work.
Competing interests: Dr. Naitoh reports financial relationship outside the submitted work: grants and personal fees from Chugai pharm, Taiho pharm, Kaken pharm, Daiichi-Sankyo, and Eli Lilly Japan, personal fees from Takeda pharm, Merck, Bayer, and Boehringer Ingelheim. Dr. Hiki reports financial relationship outside the submitted work: grants and personal fees from Abbott Japan, EA pharm, Johnson&Johnson, Otsuka pharm, Otsuka Pharm Factory Inc., Kaken pharm, Covidien Japan Inc., Takeda Pharm, Daiichi Sankyo, Taiho pharm, Tsumura, Terumo, and Miyarisan pharm, grants from Shionogi pharm, Chugai pharm, Hogy Medical, and Yakult Honsha, personal fees from NHK, Pfizer Japan, AstraZeneca, Nihon pharm, Olympus, Novartis pharm, Intuitive Surgical G.K., Ono pharm, Kaigen pharm, QLife, Sumitomo Dainippon pharm, and Nestle Japan. Dr. Kumamoto reports financial relationship outside the submitted work: grants and personal fees from Taiho pharm, Takeda pharm, and Asahi Kasei pharm, grants from Kaken pharm, Eli Lilly Japan, Chugai pharm, Daiichi Sankyo, Covidien Japan Inc., Yakult Honsha, Nihon pharm, personal fees from Viatris pharm, Ono pharm, AMCO, Novartis pharm, Olympus, Japan Blood Products Organization, and Terumo. Dr.Sangai reports financial relationship outside the submitted work: grants and personal fees from Chugai pharm, Taiho pharm, Eisai, and Kyowa Kirin, grants from Takeda pharm, and Nippon Kayaku, personal fees from Pfizer Japan, AstraZeneca, Novartis pharm, Daiichi Sankyo, Eli Lilly Japan, and Maruho. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
Colorectal cancer (CRC) is the recent leading cause of deaths among human cancers [1]. Conventional type of CRC is a genetic disease, and genomic mutations are the most frequent among human cancers [2]; mutations of APC (Adenomatous polyposis coli), K-ras, and TP53 (Tumor protein p53) are uniquely frequent in 81%, 43%, and 60%, respectively, by new generation sequence (NGS). Molecular target therapy had succeeded according to K-ras mutation status in CRC [3]. The latest promising preclinical studies also revealed that APC/β-catenin pathway could be targeted by TNIK (TRAF and NCK-interacting protein kinase) inhibition [4], while gain of function of TP53 is also an emerging potential target for tumor cells with TP53 mutation [5, 6].
On the other hand, an alternate target strategy of CRC was based on the cancer stem cell theory [7]. Cancer stem cells are supposed to be resistant to the various anti-cancer therapies and an obstacle against treatments, and stem cell biomarkers have been identified such as CD44, CD133, Ascl2, and Lgr5 in CRC [8–10].
Ishimoto et al. demonstrated that among the various CD44 transcripts, only CD44 variants (CD44V) including CD44v9 can interact with xCT (cystine transporter) through CD44v9 exon counterpart and established a function for CD44V in regulation of ROS (reactive oxygen species) as an antioxidant molecule. CD44V can bind with xCT to augment cystine influx into cytoplasm, leading to high concentration of intracellular cysteine, and tumor growth through cysteine transportation was demonstrated to play an important role in cancer stemness (cancer-initiation capacity) in the context of its NRF2 (NF-E2-related factor 2) association [11, 12].
Interestingly, CD44V expression is frequently seen in primary CRC tissues, while non-cancerous colon mucosa tissues expressed only CD44S (standard form with no variants sequence) [13, 14]. CD44V expression was the most early proposed to be a cancer progression marker in CRC [15, 16], whereas it finally failed to demonstrate it [17, 18]. As a result, an interest in CD44 expression profiles as progression markers of CRC had gradually declined, and has not been sufficiently validated yet at present. Such controversial issues are afterwards repeated even in the quantitative analysis [19, 20].
Allowing for the recent advanced understanding of CD44 variant isoforms, we hypothesized that clinical significance of CD44V putatively representing cancer stemness properties may be different from CD44T. In this study, we for the first time developed real time PCR for CD44v8-v10 as the most optimal representative for CD44V expression together with CD44T expression and an emerging stem cell marker CD133 in primary colon tumors, because such differential quantitative assessments have never been done in primary colon cancer. Therefore, we thought that we could report new findings by developing quantitative assessment systems, especially for the CD44 variant form.
Materials and methods
Patients
We collected both tumor and corresponding non-cancerous colon mucosa specimens from 60 consecutive colon cancer patients who underwent colectomy at the Department of Lower Gastrointestinal Surgery, Kitasato University Hospital, Japan, from January 2018 through December 2018. We also collected 33 liver metastasis specimens from 22 consecutive CRC cancer patients (including 8 colon cancer) with liver metastasis who underwent liver resection at the Department of General Pediatric and Hepatobiliary Pancreatic Surgery, Kitasato University Hospital, Japan, from June 2020 through December 2020. Colon cancer patients were classified by the 8th TNM classification of the Union for International Cancer Control (UICC). All tissues were collected at the Kitasato University Hospital, and informed consent was obtained. Informed consent was obtained in writing with the patient’s consent. This study was performed with approval of the Ethics Committee of the Kitasato University School of Medicine (approve number B20-382).
Cell lines and cultures
We used CRC cell lines; COLO205, COLO320, DLD1, HCT116, and LOVO were provided by the Cell Response Center for Biomedical Research Institute of Development, Aging and Cancer, Tohoku University (Sendai, Japan), SW480 cells were purchased from the American Type Culture Collection (ATCC) (Manassas, VA, USA). HCT15 cells were purchased form the RIKEN BioResouce Center (Tsukuba, Japan), and LS174T cells were purchased from the European collection of authenticated cultures (Porton Down, Salisbury, UK). LOVO and SW480 were grown in mixed medium (1:1 of RPMI1640: F-12 HAMS purchased from Sigma-Aldrich N6658, St. Louis, MO, USA) and Leibovitz L15 medium (ThermoFisher SCIENTIFIC 11415064, Walltham, MA, USA), while all other cell lines were grown in RPMI1640 medium (Gibco, Carlsbad, CA, USA), supplemented with 10% fetal bovine serum and penicillin-streptomycin (Gibco).
RNA extraction and cDNA generation
During surgery, all specimens were immediately collected and stored in RNAlaterTM RNA Stabilization Solution (ThermoFisher SCIENTIFIC, Invitrogen) at 4 degree overnight. The next day, the specimens were sliced down and stored at -80 degree. RNA extraction from homogenized specimens and cell lines was performed using RNeasy Mini Kit (Qiagen) following the manufacture’s protocol. Two micrograms of total RNA were reversed transcribed into complementary DNA (cDNA) using oligo-d(T) primer and the SuperScript Ⅲ reverse transcriptase kit (ThermoFisher SCIENTIFIC, Invitrogen).
Semi-quantitative RT-PCR
The pooled assay mix contained primers at a final concentration of 0.2 mL, 1mL of dNTP mixture, 1.5mL of MgCl2, 5 mL of PCR buffer, and 0.2 mL of Platinum Taq DNA polymerase (Invitrogen) in a final volume of 50mL. The thermal cycling conditions for CD44T and CD44V were an initial hold at 95 degree for 3 min and 35 cycles of 30 second at 95 degree, 30 second at annealing temperature (60 degree), 30 second at 72 degree, and 10 min at 72 degree. Both Primer F1 and Primer R1 were designed for gene amplification from exon 5 to exon 15 (Fig 1A), and both Primer F2 and Primer R2 were made for gene amplification from exon 8 to exon 10 (Fig 1B). The primer sequences are shown as follows (S1 Table).
(a) CD44 transcripts in CRC included both CD44S (standard) and CD44V (variants), and CD44V always included variant 8–10 from the reference 17. PrimerF1 and primerR1 can amplify both CD44S and CD44V. (b) For CD44V (v8-10), primerF2 and primerR2 were selected on the variant 8 and variant 10, respectively, with probe 2. On the other hand, for CD44T (total CD44), primerF3 and primerR3 were selected on the exon 3 and exon 5, respectively, with probe 3. (c) In semi-quantitative RT-PCR (F1-R1), multiple bands representing CD44 Variants and CD44S (185 bp) were seen in 8 CRC cell lines (upper panel). Using F2-R2 primer sets, CD44V (157 bp) could be amplified as a single band (middle panel). The lower panel showed β-actin expression as internal controls. (d) In quantitative real time PCR, CD44V and CD44T transcripts were measured in 8 CRC cell lines. Transcriptional discrepancy was seen in LOVO cells. (e) In CRC cell lines, there were close association (R = 0.83, P = 0.0092) of transcripts between CD44T and CD44V.
Quantitative RT-PCR (Q-RT-PCR)
Gene expression levels were assessed by Q-RT-PCR with iQTM Supermix Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). PCR conditions for CD44V (v8-10) (Fig 1B) were an initial hold at 95 degree 3 min and 40 cycles of 20 second at 95 degree, 30 second at annealing temperature (60 degree), 30 second at 72 degree in a 25 mL reaction volume containing 200 nmol/L fluorescent probe (probe 2) and 12.5 mL iQTM Supermix. Target gene expression levels (Ex-Value through the text) were normalized to β-actin expression and analyzed using the comparative cycle threshold method (Ex-value was defined as target gene expression/ β-actin expression x 100).
Variant 9 was encompassed by primers of both variant 8 and variant 10 designed for quantitative PCR for CD44V, whereas CD44T was represented by intracellular structure for CD44 commonly with all isoforms shared.
Statistical analysis
Continuous variables were evaluated using the Student’s t test, and categorical variables were evaluated using Fisher’s exact test or Chi-square test, as appropriate. A p value below 0.05 was considered to indicate statistical significance. All calculations were performed using JMP version 14 software (SAS institute Japan, Tokyo, Japan).
Results
CD44 expression in CRC cell lines
CD44 transcripts from CRC tumors are comprised of various isoforms including both CD44S and CD44V using pooled colonic tumor tissues by sequencing as indicated in Fig 1A as previous report [17], and interestingly all variant forms of CD44V inevitably included exon8-10. CD44V encompassed v9 exon which includes a region bound with cysteine transporter xCT to regulate redox status in cancer cells and sustain stemness [11]. Thus, CD44V is considered to represent cancer stemness.
Using both PrimerF1 and PrimerR1 (Fig 1A), CD44 Variants (multiple bands) were detected in CRC cell lines except LOVO cells, while CD44S was also expressed in CRC cell lines except LS174T (top panel of Fig 1C). Early reports described that CD44S was detected only in non-cancerous mucosas [21], however almost CRC cell lines clearly expressed it in our current study.
Using both PrimerF2 and PrimerR2 (Fig 1B), CD44V (v8-v10) which may represent total amounts of variants in a single band were strongly expressed in CRC cell lines except LOVO cells, which deemed consistent with CD44 Variants made by primerF1/primerR1 (middle panel of Fig 1C). β-actin was also shown as an internal control (bottom panel of Fig 1C).
Quantitative real time RT-PCR was then developed for CD44V and CD44T using their specific primers (primerF2/R2 and primerF3/R3, respectively) with TaqMan probes (probe2 and probe3, respectively) (Fig 1D), and HCT116 cells were proved to show the strongest bands for both forms and used as a positive control for CD44V/T expression.
Real time RT-PCR confirmed that the intense expression of CD44V was confirmed to be recognized in HCT116, LS174T, and DLD1 cells in order of expression level, and that LOVO cell was also additional for CD44T expression. In LOVO cells, CD44S is predominantly expressed differently from other cell lines. Interestingly, there was close association in expression between CD44V and CD44T in CRC cell lines except LOVO cells (R = 0.83, p = 0.0092, Fig 1E).
CD44V/CD44T expression in primary colon cancer
We then investigated CD44V expression from both primary tumors and corresponding non-cancerous mucosa tissues of the 60 colon cancer patients. CD44V expression was significantly increased in primary colon cancer tissues as compared to the corresponding non-cancerous mucosa tissues (p<0.0001) (Fig 2A).
(a) CD44V expression is elevated in 60 primary colon cancer tissues as compared to in 60 non-cancerous mucosa tissues (p<0.0001). (b) ROC curve of CD44V expression value to differentiate cancer from non-cancerous mucosa showed high AUC of 0.95, and the most optimal cut-off value was determined to be 1.54 (sensitivity of 0.93 and specificity 0.90, p<0.0001). (c) CD44V expression was significantly higher in right colon cancer than in left colon cancer (p = 0.035). (d) There was no significant difference of CD44V expression according to T factors. (e) There was significant difference of CD44V expression according to N factors (p = 0.04), and CD44V expression was rather reduced in N2 factor. (f) There was no significant difference of CD44V expression according to M factors. (g) Using the cut-off value of 5.71, right colon cancer included 12 cases with high CD44V expression, while left colon cancer had 2 cases with high CD44V expression. In right colon cancer, MSI-H cases were shown in red bars.
Using ROC (receiver operating characteristics) curve to discriminate tumor tissues from mucosa tissues, area under curve (AUC) was proved to be 0.95 (Fig 2B), and the most optimal cut-off value of the CD44V expression level determined in real time RT-PCR was 1.54. Using this optimal cut-off value, CD44V expression was high in 92% (55/60) in primary colon cancer tissues, while it was high in 10% (6/60) non-cancerous mucosa tissues (p<0.0001). These data indicated that CD44V expression is elevated specifically in primary colon cancer tissues.
We also investigated CD44T expression from both primary tumors and corresponding non-cancerous mucosa tissues of the 60 colon cancer patients. CD44T expression was significantly increased in primary colon cancer tissues as compared to the corresponding non-cancerous mucosa tissues (p<0.0001) (S1A Fig).
In ROC curve to discriminate tumor tissues from mucosa tissues, AUC was proved to be 0.84 in CD44T (S1B Fig), which is inferior to CD44V expression, and its optimal cut-off value determined in real time RT-PCR was 5.71. Using this optimal cut-off value, CD44T expression was high in 78% (47/60) in primary colon cancer tissues, while it was high in 18% (11/60) non-cancerous mucosa tissues (p<0.0001).
These data indicated that CD44T expression is also elevated specifically in primary colon cancer tissues, however its sensitivity and specificity to cancer tissues is much inferior to CD44V. This low specificity may be because CD44T transcripts included CD44S transcripts that were supposed to be expressed in non-cancerous mucosa tissues [21]. Interestingly, nevertheless, CD44T expression is closely associated with CD44V expression in primary colon cancer (R = 0.62, p<0.0001) as confirmed in CRC cell lines.
Clinicopathological significance of CD44V expression in primary colon cancer
We then examined clinicopathological characteristics of CD44V expression by real time RT-PCR in 60 colon cancers. In student’s t test, CD44V expression was significantly different between right colon and left colon cancer (p = 0.035) (Fig 2C), while there was no statistical difference with regard to other clinicopathological factors such as TNM factors except lymph node metastasis (p = 0.04) (Fig 2D–2F). Unexpectedly, CD44V expression rather tended to be low in aggressive colon cancer with lymph node metastasis or distant metastasis.
We then analyzed clinicopathological characteristics after separating right colon cancer from left colon cancer. The most optimal cut-off value of CD44V was newly determined to discriminate right colon cancer from left colon cancer by ROC curve analysis. It clarified that the most optimal one was proved to be 6.72, and using this cut-off value, CD44V expression was high in 39% (12/31) in right colon cancer, while it was high in 7% (2/29) in left colon cancer (p = 0.004) (Fig 2G).
As high expression of CD44V was uniquely seen in right colon cancer, clinicopathological characteristics were again analyzed, limited to right colon cancer (Table 1). As a result, high CD44V expression was significantly correlated with young age (p = 0.041) and less lymph node metastasis (p = 0.060).
CD44V expression and microsatellite instability (MSI) in right colon cancer
The unique clinicopathological features of high CD44V expression (right colon cancer and younger patients) are reminiscent of phenotypes of microsatellite instability-high (MSI-H) colon cancer. So, we analyzed MSI status using the 5 Bethesda panel markers, and explored the association of high CD44V expression with MSI status in right colon cancer. First, MSI-H cases were seen in 8 (26%) out of the 31 right colon cancer, which is consistent with the previous reports (20%~) [22, 23]. In right colon cancer, MSI-H status did not correlate with any clinicopathological factors.
Unexpectedly, MSI-H was rather rarely (1/12, 8%) found in right colon cancer with high CD44V expression, as compared with those with low CD44V expression (7/19, 37%) (p = 0.077, Table 1, and see red bars of left panel in Fig 2G). This finding is, however, consistent with less aggressive biological features of MSI-H phenotypes.
Clinicopathological significance of CD44T expression in primary colon cancer
We will then describe clinicopathological characteristics of CD44T expression by real time RT-PCR in 60 colon cancers. In student’s t test, CD44T expression was again significantly different between right colon and left colon cancer (p = 0.012) (S1C Fig), while there was no statistical difference with regard to other clinicopathological factors such as TNM factors except distant metastasis (M factor) (p = 0.039) (S1D–S1F Fig). CD44T expression tended to be less aggressive colon cancer like CD44V, either.
It clarified that the most optimal value to discriminate right colon cancer from left colon cancer was proved to be 7.24, and using this cut-off value, CD44T expression was high in 71% (22/31) in right colon cancer, while it was high in 34% (10/29) in left colon cancer (p = 0.005) (S1G Fig). CD44T expression did not correlate with MSI-H status in right colon cancer.
Clinicopathological significance of CD133 expression in primary colon cancer
We then investigated CD133 expression in CRC cell lines, and in both primary tumors and corresponding non-cancerous mucosa tissues of the 60 colon cancer patients. Surprisingly, CD133 expression was silenced in many CRC cell lines which are tumorigenic like DLD1 cells [24] (Fig 3A), and rather decreased in primary colon cancer tissues as compared to the corresponding non-cancerous mucosa tissues as a whole (p = 0.048) (Fig 3B). Reduced expression of CD133 in the specific human cancers is, however, consistent with the early papers describing cancer-specific promoter DNA methylation of CD133 especially in early colorectal cancer [25, 26].
(a) In semi-quantitative RT-PCR, CD133 transcripts were seen in Colo205 and HCT116 among 8 CRC cell lines (lower panel). Quantitative PCR is consistent with the semi-quantitative RT-PCR (upper panel). (b) CD133 expression was significantly reduced in 60 colon cancer tissues as compared to in 60 non-cancerous mucosa tissues (p = 0.048). (c) CD133 expressions in colon cancer tissues were significantly associated with those in non-cancerous mucosa tissues (R = 0.26, p = 0.026). (d) Public database of microdissection followed by microarray revealed that CD133 expression is rather higher in non-cancerous tissues than in cancer tissues. N.E., normal epithelia N.S., normal stroma C.E., cancer epithelia C.S., cancer stroma (e) CD133 expression in colon cancer tumor tissues was not different according to T factors (upper panel), whereas CD133 expression in non-cancerous mucosa tissues was significantly higher in T4 than in non-T4 (p = 0.048). (f) CD44T expression was not significantly different between primary colon cancer and liver metastasis (p = 0.24, left panel). On the other hand, CD44V expression was significantly lower in liver metastasis than in primary colon cancers (p = 0.0006, middle panel). CD133 expression was significantly lower in liver metastasis than in primary colon cancers, either (p = 0.0005, right panel).
Interestingly, CD133 expression in colon cancer tissues is weakly correlated with that in the corresponding non-cancerous mucosa tissues (Fig 3C, R = 0.28, p = 0.030), suggesting that CD133 expression is partially representative of the corresponding non-cancerous mucosa. Microdissection analysis of epithelium and stroma components of both CRC tissues and the corresponding non-cancerous mucosa tissues again recapitulates considerable expression of CD133 of the non-cancerous mucosa using the public database [27] (Fig 3D), which is consistent with our data.
We then examined clinicopathological characteristics of CD133 expression by real time RT-PCR in 60 colon cancers. In student’s t test, CD133 expression in the colon cancer tissues was significantly different only in non-cancerous mucosa according to pT factor (p = 0.048) by ANOVA (Fig 3E), while there was no statistical difference with regard to other clinicopathological factors in both tumors and non-cancerous mucosa (S2 Fig).
If the most optimal cut-off value of CD133 expression to discriminate colon cancer tissues from the corresponding non-cancerous mucosa is selected as 1.20 (AUC = 0.62, p = 0.10), high expression of CD133 is again significantly associated with right colon cancer, but not with other clinicopathological factors in primary colon tumors (S2 Table).
Finally, we compared expressions of both CD133 and CD44V/T in both liver metastasis of CRC (n = 33) and primary colon tumors (n = 60), and demonstrated that expressions of both CD133 and CD44V, but not CD44T, were significantly reduced in liver metastasis as compared to primary colon tumors (Fig 3F). CD44T expression was uniquely higher in liver metastasis than in primary colon tumors, however there was no statistical difference.
Discussion
Cancer stem cell hypothesis has been long focused on because cancer stemness was demonstrated to be functionally associated with drug (treatment) resistance [28–30]. Several subsequent reports proposed that cancer stemness may be correlated even with metastatic ability representing patient poor prognosis [31–33], hence proposing that cancer stemness is considered to be targeted to control cancer metastasis. However, there have remained elusive whether stem cell markers are really predictive of metastatic states of clinical cancers. In this study, we would like to explore this important clinical question. In this study, we did not include NANOG, SOX2, and OCT4, poorly differentiated markers in embryonic stem cells, because poorly differentiated tumors are rare in CRC.
Together with our early reports regarding Ascl2, one of the stem cell markers of colon cancer [34, 35], we herein investigated other well-known stem cell markers such as CD44V and CD133 at mRNA level in primary colon cancer. As a result, stem cell markers of CD44V showed remarkable increase of their expression in primary colon tumor tissues like Ascl2 [34, 35], but not associate with aggressive phenotypes of colon cancer at all. These findings suggested that cancer stemness caused by Ascl2 and CD44V does not necessarily represent metastatic potential in clinical samples of the primary tumors.
On the other hand, in this study CD133 expression is expressed in primary colon tumor tissues, whereas the corresponding non-cancerous mucosa expressed considerable amounts of CD133 transcripts, either. We wondered if CD133 mRNA expression is more abundantly expressed in non-cancerous mucosa, because an early report clearly described significant increase of CD133 expression in colon tumor tissues (p = 0.001) [36]. To validate our confusing finding, we additionally investigated CD133 mRNA expression in microdissection analysis of CRC using the public database [27], and still confirmed its considerable expression of non-cancerous mucosa epithelium (Fig 3D). Immunohistochemistry by other researchers also supported our finding, because differentiated colon mucosa showed definite expression of CD133 [37]. These findings indicated that a functional role of CD133 may be different between cancer cells and non-cancerous mucosa.
CD133 mRNA does not correlate with colon cancer metastasis at all in our current study, either. However, our data can not deny the early findings that CD133 protein can correlate with poor survival of CRC [38, 39], because such data was confirmed at protein level differently from our current study. Moreover, CD133 is a well-established cancer stem cell marker and has been definitely demonstrated to play a critical role in treatment resistance against many cancers including colon cancer [28, 40].
HDAC6 (histone deacetylase complex 6), and β-catenin form a ternary complex with CD133 to contextually stabilize β-catenin and CD133 proteins through HDAC6 deacetylase activity to suppress mesenchymal to epithelial differentiation [41]. Thus, CD133 may function as APC/ β-catenin signaling amplifier only in case of transformed colon cells.
CD133 expression was recently reported to be upregulated in CRC tumors that have a hyperactivated Ras-Raf-MEK-ERK pathway and is therefore related to mutations in K-ras or B-raf [42]. Moreover, recently K-ras mutation was demonstrated to activate cancer stem cells, contributing to colorectal tumorigenesis and metastasis in CRC cells harboring APC mutations [43]. These findings can explain why CD133 can play a pivotal role in CRC progression, but not in non-cancerous mucosa.
In our study, nevertheless transcripts expression of stem cell markers did not represent aggressive metastasis at all, as stem cell markers were rather significantly lower in liver metastasis than in primary colon tumors (Fig 3F). Intriguingly, less transcripts of stem cell markers may represent more stromal cells escorts of non-stem tumor cells in metastasis. This mechanism may reach high cost performance during cancer progression, because stem cell population generally occupy a small portion of total tumor cell population [40].
Colon cancer metastasis representing poor prognosis may be associated with specific cancer-associated fibroblast (CAF) harboring unique molecular profiles [44], and stem cell functionality is microenvironmentally defined during tumor expansion [45]. We therefore interpreted our result that transcripts of stem cell markers can be remarkably diluted due to rigorous stromal activation and looked low in metastasis.
CD44V (v8-v10) encompassed v9 which included binding region with cystine transporter xCT to regulate redox status through membrane xCT and promote tumor growth [11], supporting importance of cysteine metabolism in cancer initiation capacity representing cancer stemness [12]. On the other hand, recently soluble CD44 shed from the cancer cell surface promotes tumor metastasis by triggering macrophage IL-1 production in tumor associated macrophage (TAM) [46]. The former can represent CD44V, while the latter may reflect CD44T in this study, postulating a differential role of the individual CD44 isoform status during tumor progression. And CD44V expression representing cancer initiation capacity is significantly reduced in aggressive tumor tissues.
Asieh Sadeghi et al. reported that CD44 expression was significantly inversely associated with pathological tumors, lymph nodes, substages of metastasis (pTNM) and lymphatic infiltration [47]. This is in line with our results. Moreover, our finding regarding lymph node involvement is confirmed in a more specified situation (only in right colon cancer). This difference may be due to the evaluation method. Compared to immunohistochemistry, quantitative PCR is considered to be a more objective method for evaluating gene expressions. In this regard, our study is more accurate and objective than that of Asieh Sadeghi’s results. However, our number of patients was much smaller than in their study and we would like to increase our number of patients in the near future. In addition, PCR evaluations of bulky tumors did not consider heterogeneous expression of proteins compared to immunohistochemistry. Nevertheless, similar results obtained from different methodologies were considered to make them reasonable.
We also examined 218 CRC cases in The Cancer Genome Atlas (TCGA) database. In our case, rectal cancer cases were excluded because of their molecular modification by preoperative chemoradiotherapy, whereas rectal cancer was included in left colon cancer in the TCGA database. Although the variant of CD44 is unknown in the TCGA database, CD44 expression was significantly higher in right colon cancer than in left colon cancer as in our current study (S3A Fig). In addition, M factor and CD44 expression tended to be inversely correlated (S3B Fig). These findings support our results.
Recent reports have shown that the interaction of tumor cells with stromal cells such as cancer-related fibroblasts (CAF), tumor-related macrophages, and various forms of T cells, in other words tumor microenvironment (TME), is involved potential tumor progression in distant metastasis [48, 49]. Thus, together with our current study, we interpreted that less demand on stem cell marker-positive cancer cells may characterize metastasis of colon cancer, and believe that TME targeting may be important in controlling tumor metastasis.
In conclusion, our quantitative assessment of cancer stem cell markers revealed their unique expression patterns in right colon cancer, and moreover, we herein concluded that all the stem cell markers (CD44V, CD133, and Ascl2) investigated in our current study can not represent metastatic markers by themselves. This may represent less demand of stemness in terms of cancer metastasis, and TME control might be a bona-fide therapeutic target.
Supporting information
S1 Table. The primer sequences used in the real-time PCR experiments.
https://doi.org/10.1371/journal.pone.0277395.s001
(DOCX)
S2 Table. Clinicopathological factors and CD133 expression in primary colon cancer.
https://doi.org/10.1371/journal.pone.0277395.s002
(DOCX)
S1 Fig. Clinical significance of CD44T expression in primary colon cancer tissues.
(a) CD44T expression is elevated in 60 primary colon cancer tissues as compared to in 60 non-cancerous mucosa tissues (p<0.0001). (b) ROC curve of CD44T expression value to differentiate cancer from non-cancerous mucosa showed high AUC of 0.84, and the most optimal cut-off value was determined to be 5.71 (sensitivity of 0.78 and specificity 0.82, p<0.0001). (c) CD44T expression was significantly higher in right colon cancer than in left colon cancer (p = 0.012). (d) There was no significant difference of CD44T expression according to T factors. (e) There was marginally significant difference of CD44T expression according to N factors (p = 0.10), and CD44T expression was rather reduced in N2 factor. (f) There was significant difference of CD44V expression according to M factors (p = 0.039). CD44T was significantly reduced in M1 than in M0. (g) Using the cut-off value of 7.24, right colon cancer included 22 cases with high CD44V expression, while left colon cancer had 10 cases with high CD44V expression. In right colon cancer, MSI-H cases were shown in red bars.
https://doi.org/10.1371/journal.pone.0277395.s003
(PDF)
S2 Fig. Clinical significance of CD133 expression in primary colon cancer tissues.
(a) There was no significant difference in the expression of CD133 between right and left primary colon cancer in non cancerous mucosa tissues (p = 0.34). And similar results were obtained for N factors and M factors (p = 0.58 and p = 0.12, respectively). (b) As well as (a), there was no significant difference in the expression of CD133 between right and left primary colon cancer in colon cancer tumor tissues (p = 0.20) as well as N factors and M factors(p = 0.12 and p = 0.28, respectively).
https://doi.org/10.1371/journal.pone.0277395.s004
(PDF)
S3 Fig. Clinical significance of CD44 expression in primary colon cancer tissues in the TCGA database.
(a) There was significant difference in the expression of CD44 between right and left primary colon cancer (p = 0.00030). (b) M factor and CD44 expression tended to be inversely correlated (p = 0.089).
https://doi.org/10.1371/journal.pone.0277395.s005
(PDF)
Acknowledgments
We thank T. Miyake, Y. Matsuo, K. Fujitani for their technical assistance and advice.
References
- 1. Ferlay J, Colombet M, Soerjomataram I, et al: Estimating the global cancer incidence and mortality in 2018: GLOBOCAN sources and methods. Int J Cancer. 2019; 144:1941–1953. pmid:30350310
- 2. Comprehensive molecular characterization of human colon and rectal cancer. Nature. 2012; 487:330–337.
- 3. Karapetis CS, Khambata-Ford S, Jonker DJ, et al: K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med. 2008; 359:1757–1765. pmid:18946061
- 4. Masuda M, Uno Y, Ohbayashi N, et al: TNIK inhibition abrogates colorectal cancer stemness. Nat Commun. 2016; 7:12586. pmid:27562646
- 5. Schulz-Heddergott R, Stark N, Edmunds SJ, et al: Therapeutic Ablation of Gain-of-Function Mutant p53 in Colorectal Cancer Inhibits Stat3-Mediated Tumor Growth and Invasion. Cancer Cell. 2018; 34:298–314 e7. pmid:30107178
- 6. Tanaka T, Watanabe M, Yamashita K: Potential therapeutic targets of TP53 gene in the context of its classically canonical functions and its latest non-canonical functions in human cancer. Oncotarget. 2018; 9:16234–16247. pmid:29662640
- 7. Dalerba P, Dylla SJ, Park IK, et al: Phenotypic characterization of human colorectal cancer stem cells. Proc Natl Acad Sci U S A. 2007; 104:10158–10163. pmid:17548814
- 8. Todaro M, Francipane MG, Medema JP, et al: Colon cancer stem cells: promise of targeted therapy. Gastroenterology. 2010; 138:2151–2162. pmid:20420952
- 9. de Sousa EMF, Colak S, Buikhuisen J, et al: Methylation of cancer-stem-cell-associated Wnt target genes predicts poor prognosis in colorectal cancer patients. Cell Stem Cell. 2011; 9:476–485. pmid:22056143
- 10. Giakountis A, Moulos P, Zarkou V, et al: A Positive Regulatory Loop between a Wnt-Regulated Non-coding RNA and ASCL2 Controls Intestinal Stem Cell Fate. Cell Rep. 2016; 15:2588–2596. pmid:27292638
- 11. Ishimoto T, Nagano O, Yae T, et al: CD44 variant regulates redox status in cancer cells by stabilizing the xCT subunit of system xc(-) and thereby promotes tumor growth. Cancer Cell. 2011; 19:387–400. pmid:21397861
- 12. Kang YP, Torrente L, Falzone A, et al: Cysteine dioxygenase 1 is a metabolic liability for non-small cell lung cancer. Elife. 2019; 8:e45572. pmid:31107239
- 13. Rodriguez C, Monges G, Rouanet P, et al: CD44 expression patterns in breast and colon tumors: a PCR-based study of splice variants. Int J Cancer. 1995; 64:347–354. pmid:7591309
- 14. Goodison S, Yoshida K, Sugino T, et al: Rapid analysis of distinctive CD44 RNA splicing preferences that characterize colonic tumors. Cancer Res. 1997; 57:3140–3144. pmid:9242440
- 15. Tanabe KK, Ellis LM, Saya H: Expression of CD44R1 adhesion molecule in colon carcinomas and metastases. Lancet. 1993; 341:725–726. pmid:8095628
- 16. Mulder JW, Kruyt PM, Sewnath M, et al: Colorectal cancer prognosis and expression of exon-v6-containing CD44 proteins. Lancet. 1994; 344:1470–1472. pmid:7526103
- 17. Gotley DC, Fawcett J, Walsh MD, et al: Alternatively spliced variants of the cell adhesion molecule CD44 and tumour progression in colorectal cancer. Br J Cancer. 1996; 74:342–351. pmid:8695347
- 18. Givehchian M, Wörner S, Sträter J, et al: No evidence for cancer-related CD44 splice variants in primary and metastatic colorectal cancer. Eur J Cancer. 1998; 34:1099–1104. pmid:9849461
- 19. Yamada Y, Itano N, Narimatsu H, et al: Receptor for hyaluronan-mediated motility and CD44 expressions in colon cancer assessed by quantitative analysis using real-time reverse transcriptase-polymerase chain reaction. Jpn J Cancer Res. 1999; 90:987–992. pmid:10551329
- 20. Yamada Y, Itano N, Narimatsu H, et al: CD44 variant exon 6 expressions in colon cancer assessed by quantitative analysis using real time reverse transcriptase-polymerase chain reaction. Oncol Rep. 2003; 10:1919–1924. pmid:14534719
- 21. Matsumura Y, Tarin D: Significance of CD44 gene products for cancer diagnosis and disease evaluation. Lancet. 1992; 340:1053–1058. pmid:1357452
- 22. Ward R, Meagher A, Tomlinson I, et al: Microsatellite instability and the clinicopathological features of sporadic colorectal cancer. Gut. 2001; 48:821–829. pmid:11358903
- 23. Yaeger R, Chatila WK, Lipsyc MD, et al: Clinical Sequencing Defines the Genomic Landscape of Metastatic Colorectal Cancer. Cancer Cell. 2018; 33:125–136 e3. pmid:29316426
- 24. Shirasawa S, Furuse M, Yokoyama N, et al: Altered growth of human colon cancer cell lines disrupted at activated Ki-ras. Science. 1993; 260:85–88. pmid:8465203
- 25. Yi JM, Tsai HC, Glöckner SC, et al: Abnormal DNA methylation of CD133 in colorectal and glioblastoma tumors. Cancer Res. 2008; 68:8094–8103. pmid:18829568
- 26. Hibi K, Sakata M, Kitamura YH, et al: Demethylation of the CD133 gene is frequently detected in advanced colorectal cancer. Anticancer Res. 2009; 29:2235–2237. pmid:19528487
- 27. Nishida N, Nagahara M, Sato T, et al: Microarray analysis of colorectal cancer stromal tissue reveals upregulation of two oncogenic miRNA clusters. Clin Cancer Res. 2012; 18:3054–3070. pmid:22452939
- 28. Bao S, Wu Q, McLendon RE, et al: Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature. 2006; 444:756–760. pmid:17051156
- 29. Todaro M, Alea MP, Di Stefano AB, et al: Colon cancer stem cells dictate tumor growth and resist cell death by production of interleukin-4. Cell Stem Cell. 2007; 1:389–402. pmid:18371377
- 30. Ma S, Lee TK, Zheng BJ, et al: CD133+ HCC cancer stem cells confer chemoresistance by preferential expression of the Akt/PKB survival pathway. Oncogene. 2008; 27:1749–1758. pmid:17891174
- 31. Hermann PC, Huber SL, Herrler T, et al: Distinct populations of cancer stem cells determine tumor growth and metastatic activity in human pancreatic cancer. Cell Stem Cell. 2007; 1:313–323. pmid:18371365
- 32. Ieta K, Tanaka F, Haraguchi N, et al: Biological and genetic characteristics of tumor-initiating cells in colon cancer. Ann Surg Oncol. 2008; 15:638–648. pmid:17932721
- 33. Zeppernick F, Ahmadi R, Campos B, et al: Stem cell marker CD133 affects clinical outcome in glioma patients. Clin Cancer Res. 2008; 14:123–129. pmid:18172261
- 34. Tanaka T, Kojima K, Yokota K, et al: Comprehensive Genetic Search to Clarify the Molecular Mechanism of Drug Resistance Identifies ASCL2-LEF1/TSPAN8 Axis in Colorectal Cancer. Ann Surg Oncol. 2019; 26:1401–1411. pmid:30706227
- 35. Yokota K, Tanaka Y, Harada H, et al: WiNTRLINC1/ASCL2/c-Myc Axis Characteristics of Colon Cancer with Differentiated Histology at Young Onset and Essential for Cell Viability. Ann Surg Oncol. 2019; 26:4826–4834. pmid:31549316
- 36. Artells R, Moreno I, Díaz T, et al: Tumour CD133 mRNA expression and clinical outcome in surgically resected colorectal cancer patients. Eur J Cancer. 2010; 46:642–649. pmid:20005089
- 37. Karbanová J, Missol-Kolka E, Fonseca AV, et al: The stem cell marker CD133 (Prominin-1) is expressed in various human glandular epithelia. J Histochem Cytochem. 2008; 56:977–993. pmid:18645205
- 38. Kojima M, Ishii G, Atsumi N, et al: Immunohistochemical detection of CD133 expression in colorectal cancer: a clinicopathological study. Cancer Sci. 2008; 99:1578–1583. pmid:18754869
- 39. Horst D, Kriegl L, Engel J, et al: CD133 expression is an independent prognostic marker for low survival in colorectal cancer. Br J Cancer. 2008; 99:1285–1289. pmid:18781171
- 40. Ricci-Vitiani L, Lombardi DG, Pilozzi E, et al: Identification and expansion of human colon-cancer-initiating cells. Nature. 2007; 445:111–115. pmid:17122771
- 41. Mak AB, Nixon AM, Kittanakom S, et al: Regulation of CD133 by HDAC6 promotes β-catenin signaling to suppress cancer cell differentiation. Cell Rep. 2012; 2:951–963.
- 42. Kemper K, Versloot M, Cameron K, et al: Mutations in the Ras-Raf Axis underlie the prognostic value of CD133 in colorectal cancer. Clin Cancer Res. 2012; 18:3132–3141. pmid:22496204
- 43. Moon BS, Jeong WJ, Park J, et al: Role of oncogenic K-Ras in cancer stem cell activation by aberrant Wnt/β-catenin signaling. J Natl Cancer Inst. 2014; 106:djt373.
- 44. Li H, Courtois ET, Sengupta D, et al: Reference component analysis of single-cell transcriptomes elucidates cellular heterogeneity in human colorectal tumors. Nat Genet. 2017; 49:708–718. pmid:28319088
- 45. Lenos KJ, Miedema DM, Lodestijn SC, et al: Stem cell functionality is microenvironmentally defined during tumour expansion and therapy response in colon cancer. Nat Cell Biol. 2018; 20:1193–1202. pmid:30177776
- 46. Jang JH, Kim DH, Lim JM, et al: Breast Cancer Cell-Derived Soluble CD44 Promotes Tumor Progression by Triggering Macrophage IL1β Production. Cancer Res. 2020; 80:1342–1356.
- 47. Sadeghi Asieh, Roudi Raheleh, Mirzaei Alireza, et al: CD44 epithelial isoform inversely associates with invasive characteristics of colorectal cancer. Biomark Med. 2019; 13(6):419–426. pmid:30942083
- 48. Cristina Galindo-Pumariǹo Manuel Collado, Herrera Mercedes, et al: Tumor Microenvironment in Metastatic Colorectal Cancer: The Arbitrator in Patients’ Outcome. Cancers (Basel). 2021; 13(5):1130. pmid:33800796
- 49. Gallo Gaetano, Vescio Giuseppina, Paola Gilda De, et al: Therapeutic Targets and Tumor microenvironment in Colorectal Cancer. J Clin Med. 2021; 10(11):2295. pmid:34070480