- Email: [email protected]
The expression of CD44v6 in colon: from normal to malignant
Annals of Diagnostic Pathology 20 (2016) 19–23
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology
The expression of CD44v6 in colon: from normal to malignant Alaa Afify, MD a,⁎, Blythe Durbin-Johnson b, Avnit Virdi, BS a, Heidi Jess, MD a,1 a b
Department of Pathology and Laboratory Medicine, University of California, Davis Medical Center, PATH Bldg, 4400 V St, Sacramento, CA, 95817 Department of Public Health Sciences, Division of Biostatistics, School of Medicine, University of California, Davis One Shields Ave, Med Sci 1C, Room 145, Davis, CA, 95616
a r t i c l e
i n f o
Keywords: CD44v6 Colon Polyp Malignant
a b s t r a c t CD44v6, an integral transmembrane protein belonging to a family of adhesion molecule receptors, plays an important role in tumor growth, progression and metastasis. The purpose of this study was to evaluate the expression of CD44v6 in normal, hyperplastic, adenomatous, and malignant colonic epithelium and to determine its correlation with tumor pathologic stage and lymph node metastasis. We examined the immunohistochemical expression of CD44v6 in normal colonic tissue (n = 25), hyperplastic polyps (n = 45), tubular adenomas (n = 57), tubulovillous adenomas (n = 25), villous adenomas (n = 9), adenocarcinomas stage I (n = 26), adenocarcinomas stage III (n = 26), and lymph node metastasis (n = 26). The percentage of positive cells and the staining intensity were assessed and scored. Statistical analysis was performed using logistic regression and McNemar test. All normal colonic tissue and hyperplastic polyps showed CD44v6 staining confined to the base of the crypt. In tubular adenomas, the dysplastic surface adenomatous epithelium expressed CD44v6 in 49 (86%) cases. CD44v6 was expressed in the glandular areas of tubulovillous adenomas in 21 (84%) cases and in the villous portion in 18 (72%) cases. All villous adenomas expressed CD44v6. CD44v6 was expressed in 23 (88%) cases of stage I adenocarcinomas, in 24 (92%) cases of stage III adenocarcinomas, and in 9 (35%) cases of metastatic adenocarcinomas. We concluded that the gain of CD44v6 expression in premalignant and malignant colonic lesions suggests that CD44v6 may be functionally involved in the adenoma-to-carcinoma progression. CD44v6 did not correlate to tumor pathologic stage and is lost during the acquisition of migratory function by metastatic tumor cells. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Colorectal cancer is the second leading cause of cancer-related deaths in the United States and the third most common cancer in men and women. The American Cancer Society estimates that, in the United States in 2015, about 132,700 cases will be diagnosed with colorectal cancer and about 49,700 people will die of the disease [1]. Colorectal cancers commonly develop from preexisting polyps. Other risk factors include long-standing inflammatory bowel disease, certain types of diets, physical inactivity, obesity, smoking, and heavy alcohol consumption. Histologically, colonic polyps are classified as hyperplastic (90%) and adenomatous (10%), with the latter encompassing tubular, tubulovillous, and villous adenomas. Tubular adenomas are encountered most frequently (80%-86%) followed by tubulovillous adenomas (8%16%) and villous adenomas (5%). Adenoma-to-carcinoma progression is well established as evidenced by the presence of 1 or more synchronous adenomas in one-third of operative specimens containing colon
⁎ Corresponding author. Tel.: +1 916 734 2525; fax: +1 916 734 2560. E-mail addresses: amafi[email protected] (A. Afify), [email protected] (B. Durbin-Johnson), [email protected] (A. Virdi), [email protected] (H. Jess). 1 Present/permanent address: Department of Pathology, Pathology Sciences Medical Group, 183 E 8th Ave, Chico, CA, 95926. http://dx.doi.org/10.1016/j.anndiagpath.2015.10.010 1092-9134/© 2015 Elsevier Inc. All rights reserved.
cancer, the risk of colon cancer is increased with the number of adenomatous polyps, and adenomatous tissue is frequently found contiguous to carcinoma. Molecular genetic studies also support an adenoma-tocarcinoma sequence with the accumulation of a number of genetic and epigenetic mutation [2,3]. Adenoma-to-carcinoma progression, tumor cell invasion, and metastasis involve a series of complex interactions between malignant cells and peritumoral stroma. These interactions are accomplished through transmembrane receptors on tumor cells that interact with stromal extracellular matrix molecules. One of these receptors, CD44, binds to the extracellular matrix component hyaluronan [4]. CD44 is an 85- to 90-kDa integral transmembrane glycoprotein belonging to a distinct family of adhesion molecule receptors. Multiple variant isoforms exist (CD44v2-v10, including CD44v6) arising from alternate mRNA splicing between exons 5 and 16 [5]. The variation at the gene level translates to variations on the proximal segment of extracellular portion of the CD44 receptor. These variations alter the cell-to-cell and cell-to-matrix adhesion properties of the receptor. There has been recent intense interest in the association of variant CD44v6 expression and tumor progression and metastasis [6,7]. Although CD44v6 expression has been correlated with higher tumor stage and metastatic potential in many tumors, its expression in colorectal tumors has been inconsistent and confusing.
20
A. Afify et al. / Annals of Diagnostic Pathology 20 (2016) 19–23
The purpose of this study was to evaluate the expression and distribution of CD44v6 during the progression of colorectal neoplasia by analyzing normal, hyperplastic, adenomatous, and malignant colonic epithelium and to determine its correlation with tumor pathologic stage and lymph node metastasis.
consists of an adenomatous and villous component. Therefore, in each case of TV polyp, we evaluated the expression of CD44v6 in the adenomatous and villous component separately. The stains results were tabulated, and statistical analysis was performed. 2.4. Statistical analysis
2. Materials and methods 2.1. Case selection This retrospective study was conducted following approval by the institutional review board. A total of 234 cases were identified via the center’s laboratory information system. The study included normal colonic tissue (n = 25), hyperplastic polyps (HP, n = 45), tubular adenomas (TA, n = 57), tubulovillous adenomas (TV, n = 25), villous adenomas (VA, n = 9), adenocarcinomas stage I (n = 26), adenocarcinomas stage III (n = 26), and adenocarcinoma metastatic to lymph nodes (n = 26). Tumors were staged according to American Joint Committee on Cancer [8]. The clinical histories were reviewed, and only cases obtained from screening colonoscopy with no history of cancer, colonic polyp, or inflammatory bowel disease and with normal pathology were included in the study as examples of normal tissue. Hematoxylin and eosin– stained slides of each case were reviewed to confirm the diagnosis. A representative slide was selected from each case, and the corresponding formalin-fixed, paraffin-embedded tissue blocks were retrieved from the surgical pathology archives. To compare the CD44v6 expression in the primary tumor and the metastatic tumor, we retrieved 2 tissue blocks from every case of stage III adenocarcinoma: 1 block from the primary tumor (n = 26) and the second block from the lymph node harboring the metastatic adenocarcinoma (n = 26). 2.2. Immunohistochemical staining for CD44v6 Five-micrometer–thick sections were cut from paraffin-embedded tissue and placed onto coated slides, deparaffinized in xylene, and rehydrated in descending grades of ethanol (100%-70%). Sections were then microwaved at 97°C in 150 mL of 10 mmol/L citrate buffer (pH 6) for 15 minutes to induce epitope retrieval. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide in methanol (Vector, Burlingame, CA) for 10 minutes. Sections were incubated at room temperature for 1 hour with a monoclonal antibody against CD44v6, the monoclonal antibody VFF7 against the amino acid sequence encoded by exon CD44V6 (1:1500, Bender MedSystems, CA). The slides were incubated with a biotinylated secondary antibody at room temperature for 30 minutes and thereafter with horseradish peroxidase–conjugated avidin-biotin complex at room temperature for 1 hour. Immunostaining was visualized using 3,3′-diaminobenzidine tetrahydrochloride/H2O2 substrate and counterstained with hematoxylin. Sections from normal skin served as positive control for CD44v6. For negative controls, the primary antibody was substituted with an isotypic mouse immunoglobulin G. 2.3. Results interpretation Tissue sections from all cases were evaluated independently by 2 pathologists for the expression of CD44v6. Reactivity for CD44v6 was defined as uniform cytoplasmic membrane staining, and its intensity was graded as follows: 1, weak; 02, moderate; 3, strong. Weak stain was considered negative. The stain was considered positive if at least 10% of cells showed strong positive stain. The percentage of positive staining was classified as focal if foci of positive cells were separated by nonstaining areas and diffuse if more than 75% of the area of interest was positive. Reviewers agreed on the interpretation of all variables; in the event of a disagreement, a consensus interpretation was generated after simultaneous review of the slide. Histologically, the TV polyp
The odds of positive (3+ or 2+) CD44v6 intensity staining, of any (3+, 2+, or 1+) CD44v6 staining between groups, or of staining distribution (diffuse vs focal staining) were compared between groups using logistic regression; Firth’s [9] penalized likelihood method was used to address staining/group combinations with zero counts. Odds ratio, 95% confidence interval, and P value were tabulated and compared. The proportions of samples with positive CD44v6 staining were compared between stage III primary adenocarcinoma and matched lymph node metastasis samples using the McNemar test [10]. Analyses were conducted using SAS software for Windows, version 9.4 (SAS Institute, Cary, NC). 3. Results 3.1. Evaluation of histopathology Reevaluation of the hematoxylin and eosin–stained tissue sections confirmed the original diagnosis on all cases. Therefore, we included all the 234 cases we retrieved from our files. 3.2. CD44v6 staining results In all cases of normal colon, strong (3+) CD44v6 staining was noted only in rare cells located at the base of the crypt (Fig. 1A). The remaining epithelium of the crypts and the surface epithelium did not show any CD44v6 staining. In all 45 (100%) HP cases, the hyperplastic epithelium did not express CD44v6; however, rare positive (3+) epithelial cells were noted at the base of the crypt (Fig. 1B). Tubular adenomas expressed CD44v6 staining in 49 of 57 (86%) cases and were negative in 8 cases. The stain was diffuse in 39 cases and focal in 10 cases (3+ in 30 cases and 2+ in 19 cases). Analogous to normal colonic crypts and HP, TA showed rare CD44v6 positive (3+) epithelial cells at the base of the crypt, whereas the remaining epithelium of the crypt was negative (Figs. 2 and 3). In TV, we evaluated CD44v6 expression in the adenomatous (tubular) and villous component independently in each case. The adenomatous component of TV polyps expressed CD44v6 in 21 of 25 (84%) cases and was negative in 4 cases. The stain was diffuse in 15 cases and focal in 6 cases (3+ in 11 cases and 2+ in 10 cases). The villous component of the TV was positive for CD44v6 in 18 (72%) cases and negative in 7 cases. The stain was diffuse in 3 and focal in 15 (3+ in 3 and 2+ in 15 cases). All cases that showed positive CD44v6 staining in the villous component also exhibited positive staining in the adenomatous areas. Three cases were positive only in the adenomatous areas, and 4 cases were negative in both components (Fig. 3). Villous adenomas expressed CD44v6 in all cases (9/9; 100%). The stain was diffuse in 5 cases and focal in 4 cases (3+ in 8 cases and 2+ in 1 case). CD44v6 was expressed in 23 (88%) cases of adenocarcinomas stage I (3+ in 17 cases and 2+ in 6 cases) and in 24 (92%) cases of adenocarcinomas stage III (3+ in 21 cases and 2+ in 3 cases) (Fig. 4). Metastatic foci in the lymph nodes expressed CD44v6 in 9 (35%) of cases (3+ in 8 cases and 2+ in 1 cases). Stromal cells in all cases did not express CD44v6. Scattered lymphocytes were noted to be positive. 4. Statistical analysis results Tables 1, 2, and 3 summarize CD44v6 staining intensity, numbers of positive samples, and numbers of diffuse vs focal samples, respectively, by group. Table 4 shows numbers of positive samples in matched stage III adenocarcinoma and lymph node metastasis samples.
A. Afify et al. / Annals of Diagnostic Pathology 20 (2016) 19–23
21
Fig. 1. Normal colonic tissue (A) and hyperplastic polyp (B) expressed CD44v6 staining in few cells located at the crypt base.
4.1. Statistical analysis of staining by group The odds of positive (3+ or 2+) CD44v6 intensity staining, odds of any (3+, 2+, or 1+) CD44v6 staining between groups, and odds of staining distribution (diffuse vs focal) were compared between groups using logistic regression analysis. The odds of positive staining were significantly higher in all other groups compared with the normal and hyperplastic groups. The normal and hyperplastic groups did not differ significantly from each other, nor did other groups differ significantly from each other. The odds of strong (3+) CD44v6 staining were significantly higher in the stage I adenocarcinoma, stage III adenocarcinoma, TA, and TV compared with the hyperplastic samples. The odds of strong staining were also significantly higher in the stage I adenocarcinoma, stage III adenocarcinoma, and VA compared with the normal samples. The logistic regression analysis showed that the odds of any (3+, 2+, or 1+) CD44v6 staining between groups were significantly higher in all other groups compared with the normal and hyperplastic samples. In regard to the CD44v6 stain distribution (diffuse vs focal), the logistic regression analysis revealed that the odds of diffuse staining were significantly higher in the stage III adenocarcinoma and tubular adenomas or tubular areas in tubulovillous adenomas compared with the stage I adenocarcinoma groups. The odds of diffuse staining were significantly lower in the villous area compared with the stage III adenocarcinoma group.
of diffuse staining were significantly higher in the stage III adenocarcinoma group compared with the stage I adenocarcinoma group. The odds of positive staining or staining intensity did not differ significantly between the stage III and stage I adenocarcinoma groups. 4.3. Stage III adenocarcinoma vs matched lymph nodes The proportions of samples with positive CD44v6 staining were compared between stage III primary adenocarcinoma and matched lymph node metastasis samples using McNemar test (Table 4). The proportion of samples staining positive for CD44v6 was significantly higher in stage III adenocarcinoma samples compared with matched lymph node metastasis samples (P b .001). 5. Discussion
We compared CD44v6 results in stage I and stage III adenocarcinomas to determine if CD44v6 correlates with tumor stage. The odds
The normal colonic epithelium displays a remarkable self-renewal rate to replace the constant shedding of surface epithelium. This selfrenewal process is confined to the crypt base where stem cells are located [11]. Stem cells are undifferentiated, multipotent cells which when they divide give rise to 2 daughter cells, one being identical to the original cell, whereas the other has the potential to differentiate and is called transit-amplifying cell. Transit-amplifying cells move upward toward the crypt surface, differentiate, and finally die and are shed into the lumen. In the current study, we demonstrated that, in normal colonic mucosa and hyperplastic epithelium, CD44v6 is expressed exclusively in cells confined to the crypt base where stem cells are reported to reside. The remaining epithelium of the crypt, the crypt surface, and
Fig. 2. Tubular adenoma expressed CD44v6 staining in the dysplastic surface epithelium. Note the rare cells that stain positive at the crypt base.
Fig. 3. Tubular adenoma showing diffuse CD44v6 staining. Note the CD44v6-negative normal colonic mucosa at the polyp stalk.
4.2. Stage III vs stage I adenocarcinoma
22
A. Afify et al. / Annals of Diagnostic Pathology 20 (2016) 19–23 Table 2 CD44v6 Positive vs Negative Staining Results.
Normal Hyperplastic polyps Tubular adenomas Tubulovillous adenomas • Tubular Area • Villous Area Villous adenomas Stage I Adeno Ca Stage III Adeno Ca
Fig. 4. Primary colonic adenocarcinoma showed diffuse CD44v6 staining.
hyperplastic epithelial cells did not show any CD44v6 staining. Recent studies revealed the presence of cancer stem cells that have the ability to regenerate tumors [12]. These cancer stem cells share many characteristics with normal stem cells, including self-renewal and differentiation. CD44 is considered one of the pivotal cancer stem cell markers in the majority of cancers and is used as a surface marker for isolating cancer stem cells including breast, lung, pancreas, and colorectal [13–16]. Recently, EpCAMhigh/CD44+/CD166 + was suggested as a cancer stem cell phenotype for isolation of colorectal cancer stem cells [16]. In contrast to normal and benign hyperplastic epithelium, 86%, 84%, and 100% of adenomatous epithelium in TA, TV, and VA, respectively, gained the expression of CD44v6. These findings suggest that the induction in CD44v6 expression represents an early event in colon carcinogenesis and plays a role in the transition from normal to neoplastic colonic mucosa. A similar study performed by Kim et al [17] suggests that activation of CD44 gene expression is related to earlier events in the adenoma-carcinoma sequence, as it occurs before K-ras and p53 gene alterations in colorectal neoplasia. The intracellular domain of CD44v6 selectively interacts with several membrane-associated cytoskeletal proteins that play an important role in tumorigenesis and tumor progression. CD44v6 is required in hepatocyte growth factor–mediated activation of the c-Met tyrosine kinase pathway. This pathway plays a crucial role in inducing the proliferation, motility, adhesion, and invasion of colon cancer [18,19]. CD44 also interacts with osteopontin and regulates its cellular functions leading to tumor progression. Todaro et al [20] demonstrated that osteopontin and stromal-derived factor 1alpha induce CD44v6 expression in colorectal cancer progenitor cells, which increases the tumorigenic activity and metastasis. CD44v6 has been associated with metastatic spread and poor prognosis in animal models and several human cancers. The most compelling evidence has come from studies using animal models in which Table 1 CD44v6 staining intensity.
Normal Hyperplastic polyps Tubular adenomas Tubulovillous adenomas • Tubular area • Villous area Villous adenomas Stage I Adeno Ca Stage III Adeno Ca
3+
2+
1+
0
0 0 30 (53%)
0 0 19 (33%)
0 0 7 (12%)
25 (100%) 45 (100%) 1 (2%)
11 (44%) 8 (32%) 8 (89%) 17 (65%) 21 (81%)
10 (40%) 10 (40%) 1 (11%) 6 (23%) 3 (12%)
0 0 0 0 0
4 (16%) 7 (28%) 0 3 (12%) 2 (8%)
Positive
Negative
0 0 49 (86%)
25 (100%) 45 (100%) 8 (14%)
21 (84%) 18 (72%) 9 (100%) 23 (88%) 24 (92%)
4 (16%) 7 (28%) 0 3 (12%) 2 (8%)
Günthert et al [21] isolated CD44v6 from a metastatic rat pancreatic carcinoma cell line and confirmed that transfection of nonmetastatic tumor cells with CD44v6 enhanced metastasis. Metastasis in these models could be blocked by cotreatment with monoclonal antibodies directed against anti-CD44v6 [22]. CD44v6, has been reported to indicate a more aggressive phenotype in many tumors including breast, head and neck, lung, endometrium, and ovary [23,24]. However, conflicting data exist regarding the prognostic significance of this molecule in colon cancer. Increased expression of CD44v6 has been reported by some investigators to indicate a more aggressive phenotype and lymph node metastases and has been linked to adverse prognosis independent of Dukes stage [25,26]. Other reports, however, have failed to show any association between CD44v6 expression and lymph node metastases, survival, or long-term follow-up [27–29]. These contradictory results of CD44 expression could be due to preanalytic and analytic factors such as method and duration of tissue fixation, the use of diverse antibodies, and the definition of CD44 positive staining. The problem is compounded by comparing results obtained with different techniques. For instance, using immunohistochemistry technique, Zhao et al [30] reported the link between overexpression of CD44v6 and poor prognosis in colon cancer. They found that CD44v6 expression was observed in 72% of the colorectal cancer patients; and its expression was significantly associated with a poorly differentiated histology, greater invasion depth, lymph node metastasis, and a more advanced pathologic TNM stage. In another study, Jüngling et al [28] used reverse transcriptase polymerase chain reaction and Southern blotting to analyze CD44v6 expression in stage I-IV colorectal adenocarcinomas. They reported that CD44v6 expression failed to show any correlation with survival in long-term follow-up. Köbel et al suggest that conflicting results of CD44v6 as a prognostic factor may be related to the fact that the function of CD44v6 depends on the hyaluronan; CD44 ligand, content of cell-surrounding matrix [31]. They reported that tumor stage does not correlate to epithelial hyaluronan expression alone but to the coexpression of both epithelial hyaluronan and CD44v6. The same authors reported that hyaluronan exclusively stimulated in vitro invasion of CD44v6-expressing cells that was partly reversed by an anti-CD44v6 antibody. Their findings suggest that the adverse prognostic effect of CD44v6 in colorectal adenocarcinoma might be restricted to those tumors that have pericellular hyaluronan. We found no correlation between CD44v6 expression and tumor pathologic stage, as CD44v6 was expressed in Dukes stage I and stage III tumors in 92% and 96%, respectively. Interestingly, our results Table 3 CD44v6 Staining Distribution (Diffuse vs Focal) (Samples with +2 staining or higher).
Tubular adenomas Tubulovillous adenomas • Tubular area • Villous area Villous adenomas Stage I adenocarcinoma Stage III adenocarcinoma
Diffuse
Focal
39 (80%)
10 (20%)
15 (71%) 3 (17%) 5 (56%) 7 (30%) 21 (88%)
6 (29%) 15 (83%) 4 (44%) 16 (70%) 3 (13%)
A. Afify et al. / Annals of Diagnostic Pathology 20 (2016) 19–23 Table 4 CD44v6 staining results in matched stage III adenocarcinoma and lymph node samples. Lymph node Colon Negative Positive
Negative
Positive
2 (8%) 15 (58%)
0 9 (35%)
revealed that CD44V6 expression is lost in most metastatic tumors. We reported that 65% of metastatic adenocarcinomas in lymph nodes lost CD44v6 expression, in contrast to their primary tumor that expressed CD44v6 in 92% of the cases. This finding suggests that, in many cases of colon cancer metastasis, the expression of CD44V6 is lost during the acquisition of the migratory function. Similar to our findings Coppola et al [32] found that CD44v6 immunoreactivity was detected in 100% of the colonic adenomas and in 91% of colonic adenocarcinomas but only in 38% of nodal or liver metastases. Loss of CD44v6 expression also occurred at the invasive front in both primary and metastatic colonic adenocarcinomas and correlated with lymph node metastasis [33]. The explanation of this observation could be due to several factors. CD44 is a member of a large family of cell adhesion molecules that is responsible for adhesion between adjacent cells and between cells and the extracellular matrix. CD44 is subject to proteolytic cleavage in its proximal region of the extracellular membrane by proteases, such as membrane membrane-associated metalloproteases type I [34,35]. CD44 cleavage increases in cancer in response to production of hyaluronan oligosaccharides, fluctuation of extracellular Ca++, and activation of PKC and Ras [36,37]. CD44V6 has high affinity with its ligand hyaluronan, which could entrap tumor cells at the primary site. Shedding of receptors plays an important role in modulating cell-cell and cell-matrix adhesion. As a result, cancer cells will lose its cell-surface adhesion with neighbor cells, resulting in enhanced cellular migration and metastatic potential. Using Western blots, Okamoto et al [38] found a very high incidence of CD44 cleavage in colon carcinomas as compared with their matched normal tissues and other types of cancer. They reported that CD44 cleavage product can be detected in 90% of colon carcinomas, 67% of breast carcinomas, 45% of non–small cell lung carcinomas, and 25% of ovarian carcinomas. In conclusion, our study showed that, in normal and hyperplastic colonic epithelium, CD44v6 is exclusively expressed in cells located at the base of the crypt, the same location of colonic epithelial stem cells. The early induction CD44v6 in premalignant colonic lesions suggests that CD44v6 may be functionally involved early in the adenoma-tocarcinoma progression. CD44v6 did not correlate to tumor pathologic stage, and it is lost during the acquisition of migratory function, freeing tumor cells from their extracellular matrix and leading to metastases. Acknowledgments The project described was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through grant number UL1 TR000002. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. References [1] American Cancer Society, cancer facts & figures; 2015. [2] Carvalho B, Sillars-Hardebol AH, Postma C, et al. Colorectal adenoma to carcinoma progression is accompanied by changes in gene expression associated with ageing, chromosomal instability, and fatty acid metabolism. Cell Oncol 2012;35:53–63. [3] Li B, Shi XY, Liao DX, Cao BR, Luo CH, Cheng SJ. Advanced colorectal adenoma related gene expression signature may predict prognostic for colorectal cancer patients with adenoma-carcinoma sequence. Eur J Cancer 2015;51:409–20. [4] Banerji Suneale, Wright Alan J, Noble Martin, Mahoney DJ, Campbell ID, Day AJ, Jackson DG. Structures of the CD44-hyaluronan complex provide insight into a fundamental carbohydrate-protein interaction. Nat Struct Mol Biol 2007;14:234–9.
23
[5] Screaton GR, Bell MV, Jackson DG, Cornelis FB, Gerth U, Bell JI. Genomic structure of DNA encoding the lymphocyte homing receptor CD44 reveals at least 12 alternative spliced exons. Proc Natl Acad Sci U S A 1992;89:12160–4. [6] Tjhay F, Motohara T, Tayama S, Narantuya D, Fujimoto K, Guo J, Sakaguchi I. CD44 variant 6 is correlated with peritoneal dissemination and poor prognosis in patients with advanced epithelial ovarian cancer. Cancer Sci 2015. http://dx.doi.org/10.1111/ cas.12765. [7] Yu P, Zhou L, Ke W, Li K. Clinical significance of pAKT and CD44v6 overexpression with breast cancer. J Cancer Res Clin Oncol 2010;136(8):1283–92. [8] AJCC (American Joint Committee on Cancer) cancer staging manual. 7th ed. New York: Springer; 2010 143. [9] Firth D. Bias reduction of maximum likelihood estimates. Biometrika 1993;80:27–38. [10] McNemar Q. Note on the sampling error of the difference between correlated proportions or percentages. Psychometrika 1947;12(2):153–7. [11] Barker N, Van Oudenaarden A, Clevers H. Identifying the stem cell of the intestinal crypt: strategies and pitfalls. Cell Stem Cell 2012;11(4):452–60. [12] Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 2003;100:3983–8. [13] Du L, Wang H, He Zhang LJ, Ni B, Wang X. CD44 is of functional importance for colorectal cancer stem cells. Clin Cancer Res 2008;21:6751–60. [14] Botchkina IL, Rowehl RA, Rivadeneira Karpeh DE, Crawford H, Dufour A. Phenotypic subpopulations of metastatic colon cancer stem cells: genomic analysis. Cancer Genomics Proteomics 2009;6:19–30. [15] Dalerba P, Dylla SJ, Park IK, Liu R, Wang X, Cho RW. Phenotypic characterization of human colorectal cancer stem cells. Proc Natl Acad Sci U S A 2007;104:10158–63. [16] Leung EL, Fiscus RR, Tung JW. Non–small cell lung cancer cells expressing CD44 are enriched for stem cell-like properties. PLoS One 2010;5(11):e14062. http://dx.doi. org/10.1371/journal.pone.0014062. [17] Kim H, Yang XL, Rosada C, Hamilton SR, August JT. CD44 expression in colorectal adenomas is an early event occurring prior to K-ras and p53 gene mutation. Arch Biochem Biophys 1994;310(2):504–7. [18] Elliott VA, Rychahou P, Zaytseva YY, Evers BM. Activation of c-Met and upregulation of CD44 expression are associated with the Metastaticm phenotype in the colorectal cancer liver metastasis model. Clin Transl Oncol 2014;16(3):285–92. [19] Kammula US, Kuntz EJ, Francone TD, Zeng Z, Shia J, Landmann RG. Molecular coexpression of the c-Met oncogene and hepatocyte growth factor in primary colon cancer predicts tumor stage and clinical outcome. Cancer Lett 2007;248:219–28. [20] Todaro M, Gaggianesi M, Catalano V, Benfante A, Iovino F, Biffoni M, et al. CD44v6 is a marker of constitutive and reprogrammed cancer stem cells driving colon cancer metastasis. Cell Stem Cell 2014;14:342–56. [21] Günthert U, Hofmann M, Rudy W, et al. A new variant of glycoprotein CD44 confers metastatic potential to rat carcinoma cells. Cell 1991;65(1):13–24. [22] Guo Y, Ma J, Wang J, et al. Inhibition of human melanoma growth and metastasis in vivo by anti-CD44 monoclonal antibody. Cancer Res 1994;54(6):1561–5. [23] Afify A, McNiel Maaya A, Braggin Jacquelyn, Bailey Helen. Expression of CD44s, CD44v6 and hyaluronan across the spectrum of normal-hyperplasia-carcinoma in breast. Appl Immunohistochem Mol Morphol 2008;16:121–7. [24] Afify AM, Tate S, Durbin-Johnson B, Rocke DM, Konia T. Expression of CD44s and CD44v6 in lung cancer and their correlation with prognostic factors. Int J Biol Markers 2011;26(1):50–7. [25] Li XD, Ji M, Wu J, Jiang JT, Wu CP. Clinical significance of CD44 variants expression in colorectal cancer. Tumori 2013;99(1):88–92. [26] Saito S, Okabe H, Watanabe M, et al. CD44v6 expression is related to mesenchymal phenotype and poor prognosis in patients with colorectal cancer. Oncol Rep 2013; 29(4):1570–8. [27] Delektorskaia VV, Perevoshchikov AG, Golovkov DA, Kushlinskiĭ NE. Immunohistochemical study of E-cadherin, beta-catenin and CD-44v6 expression in the cells of primary colon cancer and its metastases. Arkh Patol 2005;67(6):34–8. [28] Jüngling B, Menges M, Goebel R, Wittig BM, Weg-Remers S, Pistorius G. Expression of CD44v6 has no prognostic value in patients with colorectal cancer. Z Gastroenterol 2002;40(4):229–33. [29] Weg-Remers S, Anders M, Von Lampe B, et al. Decreased expression of CD44 splicing variants in advanced colorectal carcinomas. Eur J Cancer 1998;34(10):1607–11. [30] Zhao LH, Lin QL, Wei J, Huai YL, Wang KJ, Yan HY. CD44v6 expression in patients with stage II or stage III sporadic colorectal cancer is superior to CD44 expression for predicting progression. Int J Clin Exp Pathol 2015;8(1):692–701. [31] Köbel M, Weichert W, Cruwell K, Schmitt WD, Lautenschläger C, Hauptmann S. Epithelial hyaluronic acid and CD44v6 are mutually involved in invasion of colorectal adenocarcinomas and linked to patient prognosis. Virchows Arch 2004;445:456–64. [32] Coppola D, Hyacinthe M, Fu L, et al. CD44V6 expression in human colorectal carcinoma. Hum Pathol 1998;29(6):627–35. [33] Avoranta ST, Korkeila EA, Syrjänen KJ, Pyrhönen SO, Sundström JT. Lack of CD44 variant 6 expression in rectal cancer invasive front associates with early recurrence. World J Gastroenterol 2012;18(33):4549–56. [34] Okamoto I, Kawano Y, Tsuiki H, et al. CD44 cleavage induced by a membraneassociated metalloprotease plays a critical role in tumor cell migration. Oncogene 1999;18:1435–46. [35] Nagano O, Saya H. Mechanism and biological significance of CD44 cleavage. Cancer Sci 2004;95:930–5. [36] Okamoto I, Kawano Y, Matsumoto M, et al. Regulated CD44 cleavage under the control of protein kinase C, calcium influx, and the Rho family of small G proteins. J Biol Chem 1999;274:25525–34. [37] Kajita M, Itoh Y, Chiba T, et al. Membrane-type 1 matrix metalloproteinase cleaves CD44 and promotes cell migration. J Cell Biol 2001;153:893-04. [38] Okamoto I, Tsuiki H, Kenyon LC, Godwin AK, et al. Proteolytic cleavage of the CD44 adhesion molecule in multiple human tumors. Am J Pathol 2002;160(2):441–7.
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology
The expression of CD44v6 in colon: from normal to malignant Alaa Afify, MD a,⁎, Blythe Durbin-Johnson b, Avnit Virdi, BS a, Heidi Jess, MD a,1 a b
Department of Pathology and Laboratory Medicine, University of California, Davis Medical Center, PATH Bldg, 4400 V St, Sacramento, CA, 95817 Department of Public Health Sciences, Division of Biostatistics, School of Medicine, University of California, Davis One Shields Ave, Med Sci 1C, Room 145, Davis, CA, 95616
a r t i c l e
i n f o
Keywords: CD44v6 Colon Polyp Malignant
a b s t r a c t CD44v6, an integral transmembrane protein belonging to a family of adhesion molecule receptors, plays an important role in tumor growth, progression and metastasis. The purpose of this study was to evaluate the expression of CD44v6 in normal, hyperplastic, adenomatous, and malignant colonic epithelium and to determine its correlation with tumor pathologic stage and lymph node metastasis. We examined the immunohistochemical expression of CD44v6 in normal colonic tissue (n = 25), hyperplastic polyps (n = 45), tubular adenomas (n = 57), tubulovillous adenomas (n = 25), villous adenomas (n = 9), adenocarcinomas stage I (n = 26), adenocarcinomas stage III (n = 26), and lymph node metastasis (n = 26). The percentage of positive cells and the staining intensity were assessed and scored. Statistical analysis was performed using logistic regression and McNemar test. All normal colonic tissue and hyperplastic polyps showed CD44v6 staining confined to the base of the crypt. In tubular adenomas, the dysplastic surface adenomatous epithelium expressed CD44v6 in 49 (86%) cases. CD44v6 was expressed in the glandular areas of tubulovillous adenomas in 21 (84%) cases and in the villous portion in 18 (72%) cases. All villous adenomas expressed CD44v6. CD44v6 was expressed in 23 (88%) cases of stage I adenocarcinomas, in 24 (92%) cases of stage III adenocarcinomas, and in 9 (35%) cases of metastatic adenocarcinomas. We concluded that the gain of CD44v6 expression in premalignant and malignant colonic lesions suggests that CD44v6 may be functionally involved in the adenoma-to-carcinoma progression. CD44v6 did not correlate to tumor pathologic stage and is lost during the acquisition of migratory function by metastatic tumor cells. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Colorectal cancer is the second leading cause of cancer-related deaths in the United States and the third most common cancer in men and women. The American Cancer Society estimates that, in the United States in 2015, about 132,700 cases will be diagnosed with colorectal cancer and about 49,700 people will die of the disease [1]. Colorectal cancers commonly develop from preexisting polyps. Other risk factors include long-standing inflammatory bowel disease, certain types of diets, physical inactivity, obesity, smoking, and heavy alcohol consumption. Histologically, colonic polyps are classified as hyperplastic (90%) and adenomatous (10%), with the latter encompassing tubular, tubulovillous, and villous adenomas. Tubular adenomas are encountered most frequently (80%-86%) followed by tubulovillous adenomas (8%16%) and villous adenomas (5%). Adenoma-to-carcinoma progression is well established as evidenced by the presence of 1 or more synchronous adenomas in one-third of operative specimens containing colon
⁎ Corresponding author. Tel.: +1 916 734 2525; fax: +1 916 734 2560. E-mail addresses: amafi[email protected] (A. Afify), [email protected] (B. Durbin-Johnson), [email protected] (A. Virdi), [email protected] (H. Jess). 1 Present/permanent address: Department of Pathology, Pathology Sciences Medical Group, 183 E 8th Ave, Chico, CA, 95926. http://dx.doi.org/10.1016/j.anndiagpath.2015.10.010 1092-9134/© 2015 Elsevier Inc. All rights reserved.
cancer, the risk of colon cancer is increased with the number of adenomatous polyps, and adenomatous tissue is frequently found contiguous to carcinoma. Molecular genetic studies also support an adenoma-tocarcinoma sequence with the accumulation of a number of genetic and epigenetic mutation [2,3]. Adenoma-to-carcinoma progression, tumor cell invasion, and metastasis involve a series of complex interactions between malignant cells and peritumoral stroma. These interactions are accomplished through transmembrane receptors on tumor cells that interact with stromal extracellular matrix molecules. One of these receptors, CD44, binds to the extracellular matrix component hyaluronan [4]. CD44 is an 85- to 90-kDa integral transmembrane glycoprotein belonging to a distinct family of adhesion molecule receptors. Multiple variant isoforms exist (CD44v2-v10, including CD44v6) arising from alternate mRNA splicing between exons 5 and 16 [5]. The variation at the gene level translates to variations on the proximal segment of extracellular portion of the CD44 receptor. These variations alter the cell-to-cell and cell-to-matrix adhesion properties of the receptor. There has been recent intense interest in the association of variant CD44v6 expression and tumor progression and metastasis [6,7]. Although CD44v6 expression has been correlated with higher tumor stage and metastatic potential in many tumors, its expression in colorectal tumors has been inconsistent and confusing.
20
A. Afify et al. / Annals of Diagnostic Pathology 20 (2016) 19–23
The purpose of this study was to evaluate the expression and distribution of CD44v6 during the progression of colorectal neoplasia by analyzing normal, hyperplastic, adenomatous, and malignant colonic epithelium and to determine its correlation with tumor pathologic stage and lymph node metastasis.
consists of an adenomatous and villous component. Therefore, in each case of TV polyp, we evaluated the expression of CD44v6 in the adenomatous and villous component separately. The stains results were tabulated, and statistical analysis was performed. 2.4. Statistical analysis
2. Materials and methods 2.1. Case selection This retrospective study was conducted following approval by the institutional review board. A total of 234 cases were identified via the center’s laboratory information system. The study included normal colonic tissue (n = 25), hyperplastic polyps (HP, n = 45), tubular adenomas (TA, n = 57), tubulovillous adenomas (TV, n = 25), villous adenomas (VA, n = 9), adenocarcinomas stage I (n = 26), adenocarcinomas stage III (n = 26), and adenocarcinoma metastatic to lymph nodes (n = 26). Tumors were staged according to American Joint Committee on Cancer [8]. The clinical histories were reviewed, and only cases obtained from screening colonoscopy with no history of cancer, colonic polyp, or inflammatory bowel disease and with normal pathology were included in the study as examples of normal tissue. Hematoxylin and eosin– stained slides of each case were reviewed to confirm the diagnosis. A representative slide was selected from each case, and the corresponding formalin-fixed, paraffin-embedded tissue blocks were retrieved from the surgical pathology archives. To compare the CD44v6 expression in the primary tumor and the metastatic tumor, we retrieved 2 tissue blocks from every case of stage III adenocarcinoma: 1 block from the primary tumor (n = 26) and the second block from the lymph node harboring the metastatic adenocarcinoma (n = 26). 2.2. Immunohistochemical staining for CD44v6 Five-micrometer–thick sections were cut from paraffin-embedded tissue and placed onto coated slides, deparaffinized in xylene, and rehydrated in descending grades of ethanol (100%-70%). Sections were then microwaved at 97°C in 150 mL of 10 mmol/L citrate buffer (pH 6) for 15 minutes to induce epitope retrieval. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide in methanol (Vector, Burlingame, CA) for 10 minutes. Sections were incubated at room temperature for 1 hour with a monoclonal antibody against CD44v6, the monoclonal antibody VFF7 against the amino acid sequence encoded by exon CD44V6 (1:1500, Bender MedSystems, CA). The slides were incubated with a biotinylated secondary antibody at room temperature for 30 minutes and thereafter with horseradish peroxidase–conjugated avidin-biotin complex at room temperature for 1 hour. Immunostaining was visualized using 3,3′-diaminobenzidine tetrahydrochloride/H2O2 substrate and counterstained with hematoxylin. Sections from normal skin served as positive control for CD44v6. For negative controls, the primary antibody was substituted with an isotypic mouse immunoglobulin G. 2.3. Results interpretation Tissue sections from all cases were evaluated independently by 2 pathologists for the expression of CD44v6. Reactivity for CD44v6 was defined as uniform cytoplasmic membrane staining, and its intensity was graded as follows: 1, weak; 02, moderate; 3, strong. Weak stain was considered negative. The stain was considered positive if at least 10% of cells showed strong positive stain. The percentage of positive staining was classified as focal if foci of positive cells were separated by nonstaining areas and diffuse if more than 75% of the area of interest was positive. Reviewers agreed on the interpretation of all variables; in the event of a disagreement, a consensus interpretation was generated after simultaneous review of the slide. Histologically, the TV polyp
The odds of positive (3+ or 2+) CD44v6 intensity staining, of any (3+, 2+, or 1+) CD44v6 staining between groups, or of staining distribution (diffuse vs focal staining) were compared between groups using logistic regression; Firth’s [9] penalized likelihood method was used to address staining/group combinations with zero counts. Odds ratio, 95% confidence interval, and P value were tabulated and compared. The proportions of samples with positive CD44v6 staining were compared between stage III primary adenocarcinoma and matched lymph node metastasis samples using the McNemar test [10]. Analyses were conducted using SAS software for Windows, version 9.4 (SAS Institute, Cary, NC). 3. Results 3.1. Evaluation of histopathology Reevaluation of the hematoxylin and eosin–stained tissue sections confirmed the original diagnosis on all cases. Therefore, we included all the 234 cases we retrieved from our files. 3.2. CD44v6 staining results In all cases of normal colon, strong (3+) CD44v6 staining was noted only in rare cells located at the base of the crypt (Fig. 1A). The remaining epithelium of the crypts and the surface epithelium did not show any CD44v6 staining. In all 45 (100%) HP cases, the hyperplastic epithelium did not express CD44v6; however, rare positive (3+) epithelial cells were noted at the base of the crypt (Fig. 1B). Tubular adenomas expressed CD44v6 staining in 49 of 57 (86%) cases and were negative in 8 cases. The stain was diffuse in 39 cases and focal in 10 cases (3+ in 30 cases and 2+ in 19 cases). Analogous to normal colonic crypts and HP, TA showed rare CD44v6 positive (3+) epithelial cells at the base of the crypt, whereas the remaining epithelium of the crypt was negative (Figs. 2 and 3). In TV, we evaluated CD44v6 expression in the adenomatous (tubular) and villous component independently in each case. The adenomatous component of TV polyps expressed CD44v6 in 21 of 25 (84%) cases and was negative in 4 cases. The stain was diffuse in 15 cases and focal in 6 cases (3+ in 11 cases and 2+ in 10 cases). The villous component of the TV was positive for CD44v6 in 18 (72%) cases and negative in 7 cases. The stain was diffuse in 3 and focal in 15 (3+ in 3 and 2+ in 15 cases). All cases that showed positive CD44v6 staining in the villous component also exhibited positive staining in the adenomatous areas. Three cases were positive only in the adenomatous areas, and 4 cases were negative in both components (Fig. 3). Villous adenomas expressed CD44v6 in all cases (9/9; 100%). The stain was diffuse in 5 cases and focal in 4 cases (3+ in 8 cases and 2+ in 1 case). CD44v6 was expressed in 23 (88%) cases of adenocarcinomas stage I (3+ in 17 cases and 2+ in 6 cases) and in 24 (92%) cases of adenocarcinomas stage III (3+ in 21 cases and 2+ in 3 cases) (Fig. 4). Metastatic foci in the lymph nodes expressed CD44v6 in 9 (35%) of cases (3+ in 8 cases and 2+ in 1 cases). Stromal cells in all cases did not express CD44v6. Scattered lymphocytes were noted to be positive. 4. Statistical analysis results Tables 1, 2, and 3 summarize CD44v6 staining intensity, numbers of positive samples, and numbers of diffuse vs focal samples, respectively, by group. Table 4 shows numbers of positive samples in matched stage III adenocarcinoma and lymph node metastasis samples.
A. Afify et al. / Annals of Diagnostic Pathology 20 (2016) 19–23
21
Fig. 1. Normal colonic tissue (A) and hyperplastic polyp (B) expressed CD44v6 staining in few cells located at the crypt base.
4.1. Statistical analysis of staining by group The odds of positive (3+ or 2+) CD44v6 intensity staining, odds of any (3+, 2+, or 1+) CD44v6 staining between groups, and odds of staining distribution (diffuse vs focal) were compared between groups using logistic regression analysis. The odds of positive staining were significantly higher in all other groups compared with the normal and hyperplastic groups. The normal and hyperplastic groups did not differ significantly from each other, nor did other groups differ significantly from each other. The odds of strong (3+) CD44v6 staining were significantly higher in the stage I adenocarcinoma, stage III adenocarcinoma, TA, and TV compared with the hyperplastic samples. The odds of strong staining were also significantly higher in the stage I adenocarcinoma, stage III adenocarcinoma, and VA compared with the normal samples. The logistic regression analysis showed that the odds of any (3+, 2+, or 1+) CD44v6 staining between groups were significantly higher in all other groups compared with the normal and hyperplastic samples. In regard to the CD44v6 stain distribution (diffuse vs focal), the logistic regression analysis revealed that the odds of diffuse staining were significantly higher in the stage III adenocarcinoma and tubular adenomas or tubular areas in tubulovillous adenomas compared with the stage I adenocarcinoma groups. The odds of diffuse staining were significantly lower in the villous area compared with the stage III adenocarcinoma group.
of diffuse staining were significantly higher in the stage III adenocarcinoma group compared with the stage I adenocarcinoma group. The odds of positive staining or staining intensity did not differ significantly between the stage III and stage I adenocarcinoma groups. 4.3. Stage III adenocarcinoma vs matched lymph nodes The proportions of samples with positive CD44v6 staining were compared between stage III primary adenocarcinoma and matched lymph node metastasis samples using McNemar test (Table 4). The proportion of samples staining positive for CD44v6 was significantly higher in stage III adenocarcinoma samples compared with matched lymph node metastasis samples (P b .001). 5. Discussion
We compared CD44v6 results in stage I and stage III adenocarcinomas to determine if CD44v6 correlates with tumor stage. The odds
The normal colonic epithelium displays a remarkable self-renewal rate to replace the constant shedding of surface epithelium. This selfrenewal process is confined to the crypt base where stem cells are located [11]. Stem cells are undifferentiated, multipotent cells which when they divide give rise to 2 daughter cells, one being identical to the original cell, whereas the other has the potential to differentiate and is called transit-amplifying cell. Transit-amplifying cells move upward toward the crypt surface, differentiate, and finally die and are shed into the lumen. In the current study, we demonstrated that, in normal colonic mucosa and hyperplastic epithelium, CD44v6 is expressed exclusively in cells confined to the crypt base where stem cells are reported to reside. The remaining epithelium of the crypt, the crypt surface, and
Fig. 2. Tubular adenoma expressed CD44v6 staining in the dysplastic surface epithelium. Note the rare cells that stain positive at the crypt base.
Fig. 3. Tubular adenoma showing diffuse CD44v6 staining. Note the CD44v6-negative normal colonic mucosa at the polyp stalk.
4.2. Stage III vs stage I adenocarcinoma
22
A. Afify et al. / Annals of Diagnostic Pathology 20 (2016) 19–23 Table 2 CD44v6 Positive vs Negative Staining Results.
Normal Hyperplastic polyps Tubular adenomas Tubulovillous adenomas • Tubular Area • Villous Area Villous adenomas Stage I Adeno Ca Stage III Adeno Ca
Fig. 4. Primary colonic adenocarcinoma showed diffuse CD44v6 staining.
hyperplastic epithelial cells did not show any CD44v6 staining. Recent studies revealed the presence of cancer stem cells that have the ability to regenerate tumors [12]. These cancer stem cells share many characteristics with normal stem cells, including self-renewal and differentiation. CD44 is considered one of the pivotal cancer stem cell markers in the majority of cancers and is used as a surface marker for isolating cancer stem cells including breast, lung, pancreas, and colorectal [13–16]. Recently, EpCAMhigh/CD44+/CD166 + was suggested as a cancer stem cell phenotype for isolation of colorectal cancer stem cells [16]. In contrast to normal and benign hyperplastic epithelium, 86%, 84%, and 100% of adenomatous epithelium in TA, TV, and VA, respectively, gained the expression of CD44v6. These findings suggest that the induction in CD44v6 expression represents an early event in colon carcinogenesis and plays a role in the transition from normal to neoplastic colonic mucosa. A similar study performed by Kim et al [17] suggests that activation of CD44 gene expression is related to earlier events in the adenoma-carcinoma sequence, as it occurs before K-ras and p53 gene alterations in colorectal neoplasia. The intracellular domain of CD44v6 selectively interacts with several membrane-associated cytoskeletal proteins that play an important role in tumorigenesis and tumor progression. CD44v6 is required in hepatocyte growth factor–mediated activation of the c-Met tyrosine kinase pathway. This pathway plays a crucial role in inducing the proliferation, motility, adhesion, and invasion of colon cancer [18,19]. CD44 also interacts with osteopontin and regulates its cellular functions leading to tumor progression. Todaro et al [20] demonstrated that osteopontin and stromal-derived factor 1alpha induce CD44v6 expression in colorectal cancer progenitor cells, which increases the tumorigenic activity and metastasis. CD44v6 has been associated with metastatic spread and poor prognosis in animal models and several human cancers. The most compelling evidence has come from studies using animal models in which Table 1 CD44v6 staining intensity.
Normal Hyperplastic polyps Tubular adenomas Tubulovillous adenomas • Tubular area • Villous area Villous adenomas Stage I Adeno Ca Stage III Adeno Ca
3+
2+
1+
0
0 0 30 (53%)
0 0 19 (33%)
0 0 7 (12%)
25 (100%) 45 (100%) 1 (2%)
11 (44%) 8 (32%) 8 (89%) 17 (65%) 21 (81%)
10 (40%) 10 (40%) 1 (11%) 6 (23%) 3 (12%)
0 0 0 0 0
4 (16%) 7 (28%) 0 3 (12%) 2 (8%)
Positive
Negative
0 0 49 (86%)
25 (100%) 45 (100%) 8 (14%)
21 (84%) 18 (72%) 9 (100%) 23 (88%) 24 (92%)
4 (16%) 7 (28%) 0 3 (12%) 2 (8%)
Günthert et al [21] isolated CD44v6 from a metastatic rat pancreatic carcinoma cell line and confirmed that transfection of nonmetastatic tumor cells with CD44v6 enhanced metastasis. Metastasis in these models could be blocked by cotreatment with monoclonal antibodies directed against anti-CD44v6 [22]. CD44v6, has been reported to indicate a more aggressive phenotype in many tumors including breast, head and neck, lung, endometrium, and ovary [23,24]. However, conflicting data exist regarding the prognostic significance of this molecule in colon cancer. Increased expression of CD44v6 has been reported by some investigators to indicate a more aggressive phenotype and lymph node metastases and has been linked to adverse prognosis independent of Dukes stage [25,26]. Other reports, however, have failed to show any association between CD44v6 expression and lymph node metastases, survival, or long-term follow-up [27–29]. These contradictory results of CD44 expression could be due to preanalytic and analytic factors such as method and duration of tissue fixation, the use of diverse antibodies, and the definition of CD44 positive staining. The problem is compounded by comparing results obtained with different techniques. For instance, using immunohistochemistry technique, Zhao et al [30] reported the link between overexpression of CD44v6 and poor prognosis in colon cancer. They found that CD44v6 expression was observed in 72% of the colorectal cancer patients; and its expression was significantly associated with a poorly differentiated histology, greater invasion depth, lymph node metastasis, and a more advanced pathologic TNM stage. In another study, Jüngling et al [28] used reverse transcriptase polymerase chain reaction and Southern blotting to analyze CD44v6 expression in stage I-IV colorectal adenocarcinomas. They reported that CD44v6 expression failed to show any correlation with survival in long-term follow-up. Köbel et al suggest that conflicting results of CD44v6 as a prognostic factor may be related to the fact that the function of CD44v6 depends on the hyaluronan; CD44 ligand, content of cell-surrounding matrix [31]. They reported that tumor stage does not correlate to epithelial hyaluronan expression alone but to the coexpression of both epithelial hyaluronan and CD44v6. The same authors reported that hyaluronan exclusively stimulated in vitro invasion of CD44v6-expressing cells that was partly reversed by an anti-CD44v6 antibody. Their findings suggest that the adverse prognostic effect of CD44v6 in colorectal adenocarcinoma might be restricted to those tumors that have pericellular hyaluronan. We found no correlation between CD44v6 expression and tumor pathologic stage, as CD44v6 was expressed in Dukes stage I and stage III tumors in 92% and 96%, respectively. Interestingly, our results Table 3 CD44v6 Staining Distribution (Diffuse vs Focal) (Samples with +2 staining or higher).
Tubular adenomas Tubulovillous adenomas • Tubular area • Villous area Villous adenomas Stage I adenocarcinoma Stage III adenocarcinoma
Diffuse
Focal
39 (80%)
10 (20%)
15 (71%) 3 (17%) 5 (56%) 7 (30%) 21 (88%)
6 (29%) 15 (83%) 4 (44%) 16 (70%) 3 (13%)
A. Afify et al. / Annals of Diagnostic Pathology 20 (2016) 19–23 Table 4 CD44v6 staining results in matched stage III adenocarcinoma and lymph node samples. Lymph node Colon Negative Positive
Negative
Positive
2 (8%) 15 (58%)
0 9 (35%)
revealed that CD44V6 expression is lost in most metastatic tumors. We reported that 65% of metastatic adenocarcinomas in lymph nodes lost CD44v6 expression, in contrast to their primary tumor that expressed CD44v6 in 92% of the cases. This finding suggests that, in many cases of colon cancer metastasis, the expression of CD44V6 is lost during the acquisition of the migratory function. Similar to our findings Coppola et al [32] found that CD44v6 immunoreactivity was detected in 100% of the colonic adenomas and in 91% of colonic adenocarcinomas but only in 38% of nodal or liver metastases. Loss of CD44v6 expression also occurred at the invasive front in both primary and metastatic colonic adenocarcinomas and correlated with lymph node metastasis [33]. The explanation of this observation could be due to several factors. CD44 is a member of a large family of cell adhesion molecules that is responsible for adhesion between adjacent cells and between cells and the extracellular matrix. CD44 is subject to proteolytic cleavage in its proximal region of the extracellular membrane by proteases, such as membrane membrane-associated metalloproteases type I [34,35]. CD44 cleavage increases in cancer in response to production of hyaluronan oligosaccharides, fluctuation of extracellular Ca++, and activation of PKC and Ras [36,37]. CD44V6 has high affinity with its ligand hyaluronan, which could entrap tumor cells at the primary site. Shedding of receptors plays an important role in modulating cell-cell and cell-matrix adhesion. As a result, cancer cells will lose its cell-surface adhesion with neighbor cells, resulting in enhanced cellular migration and metastatic potential. Using Western blots, Okamoto et al [38] found a very high incidence of CD44 cleavage in colon carcinomas as compared with their matched normal tissues and other types of cancer. They reported that CD44 cleavage product can be detected in 90% of colon carcinomas, 67% of breast carcinomas, 45% of non–small cell lung carcinomas, and 25% of ovarian carcinomas. In conclusion, our study showed that, in normal and hyperplastic colonic epithelium, CD44v6 is exclusively expressed in cells located at the base of the crypt, the same location of colonic epithelial stem cells. The early induction CD44v6 in premalignant colonic lesions suggests that CD44v6 may be functionally involved early in the adenoma-tocarcinoma progression. CD44v6 did not correlate to tumor pathologic stage, and it is lost during the acquisition of migratory function, freeing tumor cells from their extracellular matrix and leading to metastases. Acknowledgments The project described was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through grant number UL1 TR000002. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. References [1] American Cancer Society, cancer facts & figures; 2015. [2] Carvalho B, Sillars-Hardebol AH, Postma C, et al. Colorectal adenoma to carcinoma progression is accompanied by changes in gene expression associated with ageing, chromosomal instability, and fatty acid metabolism. Cell Oncol 2012;35:53–63. [3] Li B, Shi XY, Liao DX, Cao BR, Luo CH, Cheng SJ. Advanced colorectal adenoma related gene expression signature may predict prognostic for colorectal cancer patients with adenoma-carcinoma sequence. Eur J Cancer 2015;51:409–20. [4] Banerji Suneale, Wright Alan J, Noble Martin, Mahoney DJ, Campbell ID, Day AJ, Jackson DG. Structures of the CD44-hyaluronan complex provide insight into a fundamental carbohydrate-protein interaction. Nat Struct Mol Biol 2007;14:234–9.
23
[5] Screaton GR, Bell MV, Jackson DG, Cornelis FB, Gerth U, Bell JI. Genomic structure of DNA encoding the lymphocyte homing receptor CD44 reveals at least 12 alternative spliced exons. Proc Natl Acad Sci U S A 1992;89:12160–4. [6] Tjhay F, Motohara T, Tayama S, Narantuya D, Fujimoto K, Guo J, Sakaguchi I. CD44 variant 6 is correlated with peritoneal dissemination and poor prognosis in patients with advanced epithelial ovarian cancer. Cancer Sci 2015. http://dx.doi.org/10.1111/ cas.12765. [7] Yu P, Zhou L, Ke W, Li K. Clinical significance of pAKT and CD44v6 overexpression with breast cancer. J Cancer Res Clin Oncol 2010;136(8):1283–92. [8] AJCC (American Joint Committee on Cancer) cancer staging manual. 7th ed. New York: Springer; 2010 143. [9] Firth D. Bias reduction of maximum likelihood estimates. Biometrika 1993;80:27–38. [10] McNemar Q. Note on the sampling error of the difference between correlated proportions or percentages. Psychometrika 1947;12(2):153–7. [11] Barker N, Van Oudenaarden A, Clevers H. Identifying the stem cell of the intestinal crypt: strategies and pitfalls. Cell Stem Cell 2012;11(4):452–60. [12] Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 2003;100:3983–8. [13] Du L, Wang H, He Zhang LJ, Ni B, Wang X. CD44 is of functional importance for colorectal cancer stem cells. Clin Cancer Res 2008;21:6751–60. [14] Botchkina IL, Rowehl RA, Rivadeneira Karpeh DE, Crawford H, Dufour A. Phenotypic subpopulations of metastatic colon cancer stem cells: genomic analysis. Cancer Genomics Proteomics 2009;6:19–30. [15] Dalerba P, Dylla SJ, Park IK, Liu R, Wang X, Cho RW. Phenotypic characterization of human colorectal cancer stem cells. Proc Natl Acad Sci U S A 2007;104:10158–63. [16] Leung EL, Fiscus RR, Tung JW. Non–small cell lung cancer cells expressing CD44 are enriched for stem cell-like properties. PLoS One 2010;5(11):e14062. http://dx.doi. org/10.1371/journal.pone.0014062. [17] Kim H, Yang XL, Rosada C, Hamilton SR, August JT. CD44 expression in colorectal adenomas is an early event occurring prior to K-ras and p53 gene mutation. Arch Biochem Biophys 1994;310(2):504–7. [18] Elliott VA, Rychahou P, Zaytseva YY, Evers BM. Activation of c-Met and upregulation of CD44 expression are associated with the Metastaticm phenotype in the colorectal cancer liver metastasis model. Clin Transl Oncol 2014;16(3):285–92. [19] Kammula US, Kuntz EJ, Francone TD, Zeng Z, Shia J, Landmann RG. Molecular coexpression of the c-Met oncogene and hepatocyte growth factor in primary colon cancer predicts tumor stage and clinical outcome. Cancer Lett 2007;248:219–28. [20] Todaro M, Gaggianesi M, Catalano V, Benfante A, Iovino F, Biffoni M, et al. CD44v6 is a marker of constitutive and reprogrammed cancer stem cells driving colon cancer metastasis. Cell Stem Cell 2014;14:342–56. [21] Günthert U, Hofmann M, Rudy W, et al. A new variant of glycoprotein CD44 confers metastatic potential to rat carcinoma cells. Cell 1991;65(1):13–24. [22] Guo Y, Ma J, Wang J, et al. Inhibition of human melanoma growth and metastasis in vivo by anti-CD44 monoclonal antibody. Cancer Res 1994;54(6):1561–5. [23] Afify A, McNiel Maaya A, Braggin Jacquelyn, Bailey Helen. Expression of CD44s, CD44v6 and hyaluronan across the spectrum of normal-hyperplasia-carcinoma in breast. Appl Immunohistochem Mol Morphol 2008;16:121–7. [24] Afify AM, Tate S, Durbin-Johnson B, Rocke DM, Konia T. Expression of CD44s and CD44v6 in lung cancer and their correlation with prognostic factors. Int J Biol Markers 2011;26(1):50–7. [25] Li XD, Ji M, Wu J, Jiang JT, Wu CP. Clinical significance of CD44 variants expression in colorectal cancer. Tumori 2013;99(1):88–92. [26] Saito S, Okabe H, Watanabe M, et al. CD44v6 expression is related to mesenchymal phenotype and poor prognosis in patients with colorectal cancer. Oncol Rep 2013; 29(4):1570–8. [27] Delektorskaia VV, Perevoshchikov AG, Golovkov DA, Kushlinskiĭ NE. Immunohistochemical study of E-cadherin, beta-catenin and CD-44v6 expression in the cells of primary colon cancer and its metastases. Arkh Patol 2005;67(6):34–8. [28] Jüngling B, Menges M, Goebel R, Wittig BM, Weg-Remers S, Pistorius G. Expression of CD44v6 has no prognostic value in patients with colorectal cancer. Z Gastroenterol 2002;40(4):229–33. [29] Weg-Remers S, Anders M, Von Lampe B, et al. Decreased expression of CD44 splicing variants in advanced colorectal carcinomas. Eur J Cancer 1998;34(10):1607–11. [30] Zhao LH, Lin QL, Wei J, Huai YL, Wang KJ, Yan HY. CD44v6 expression in patients with stage II or stage III sporadic colorectal cancer is superior to CD44 expression for predicting progression. Int J Clin Exp Pathol 2015;8(1):692–701. [31] Köbel M, Weichert W, Cruwell K, Schmitt WD, Lautenschläger C, Hauptmann S. Epithelial hyaluronic acid and CD44v6 are mutually involved in invasion of colorectal adenocarcinomas and linked to patient prognosis. Virchows Arch 2004;445:456–64. [32] Coppola D, Hyacinthe M, Fu L, et al. CD44V6 expression in human colorectal carcinoma. Hum Pathol 1998;29(6):627–35. [33] Avoranta ST, Korkeila EA, Syrjänen KJ, Pyrhönen SO, Sundström JT. Lack of CD44 variant 6 expression in rectal cancer invasive front associates with early recurrence. World J Gastroenterol 2012;18(33):4549–56. [34] Okamoto I, Kawano Y, Tsuiki H, et al. CD44 cleavage induced by a membraneassociated metalloprotease plays a critical role in tumor cell migration. Oncogene 1999;18:1435–46. [35] Nagano O, Saya H. Mechanism and biological significance of CD44 cleavage. Cancer Sci 2004;95:930–5. [36] Okamoto I, Kawano Y, Matsumoto M, et al. Regulated CD44 cleavage under the control of protein kinase C, calcium influx, and the Rho family of small G proteins. J Biol Chem 1999;274:25525–34. [37] Kajita M, Itoh Y, Chiba T, et al. Membrane-type 1 matrix metalloproteinase cleaves CD44 and promotes cell migration. J Cell Biol 2001;153:893-04. [38] Okamoto I, Tsuiki H, Kenyon LC, Godwin AK, et al. Proteolytic cleavage of the CD44 adhesion molecule in multiple human tumors. Am J Pathol 2002;160(2):441–7.