Colorectal cancer (CRC) is the third most commonly diagnosed cancer and the second leading cause of cancer-related death in the USA (Calongi et al., 2008). In Egypt, it occupied the first rank among digestive system malignancies, representing 15.78%, and the fifth rank among total cancers, representing 4.34% (Mokhtar et al., 2007).
The metastatic potential in cancer depends on the invasion of the cancer cells to the surrounding matrix and penetration of basement membrane to gain access to the systemic circulation. These steps are basically influenced by tumor-associated proteolytic enzymes, which include endopeptidases, matrix metalloproteinase, and cathepsins (Tetu et al., 2002).
Cathepsin D (CD) can act directly by proteolytic cleavage of extracellular matrix (ECM) components or indirectly by coactivation of a cascade of other proteases such as metalloproteases or elastase, which degrade the ECM and basal membranes. In addition to its catalytic activity, human CD can also act as a direct or indirect mitogenic factor for cancer cells, thus promoting tumor growth (Berchem et al., 2002).
Several investigators have reported a wide range of CD and their antigen expression patterns in colorectal tumors in association with advanced disease stage, suggesting that CD may be a potential prognostic tumor marker and a target for future therapy (Sebzda et al., 2005).
The BCL-2 (B-cell lymphoma-2) is a gene coding for a 26 kDa protein. It specifically blocks programmed cell death without promoting cell proliferation (Scopa et al., 2003). BCL-2 has emerged as an important regulator of apoptosis owing to its role in protecting cells from death induced by several agents, including radiation, chemotherapy, or growth factor deprivation (Scopa et al., 2001), by facilitating G0, thus suggesting that maintenance of cell survival is at the expense of proliferation (Ruggiero, 2009). However, some studies have reported that BCL-2 expression in CRC is a prognostic marker associated with favorable outcome (Krajewska et al., 2005), owing to the ability of BCL-2 to be converted from a protector to a killer through interactions with a variety of proteins (Lin et al., 2004).
It has been suggested that targeting BCL-2 prosurvival proteins will have significant impact in cancer treatment (Adams and Cory, 2007). Direct inhibition of the antiapoptotic BCL-2 family members with GX15-070 (GX; obatoclax), a BH3-mimetic recently under investigations in clinical trials, is an attractive effective strategy to overcome antiestrogen resistance in breast cancers. GX has a dual effect in promoting cell death: (a) it induces apoptosis through direct inhibition of antiapoptotic BCL-2 family members; and (b) it inhibits cathepsins D and L protein expression to minimize the ability of cells to consume degraded material to induce cellular metabolism and restore homeostasis (Schwartz-Roberts et al., 2013).
The aim of this study was to evaluate the immunohistochemical expression of CD and BCL-2 in CRC and their relation with Ki-67 proliferation index and various clinicopathologic variables to verify their role in cancer progression, possible prognostic impact, and the feasibility of therapeutic targeting of both CD with BCL-2 in CRC.
Materials and methods
This retrospective study included 40 cases of primary CRCs obtained through a collection of archived paraffin blocks of colectomy specimens from the Pathology Department, Faculty of Medicine, Cairo University.
A written consent has been taken from each patient, or his relatives, to approve using clinico-pathological data without disclosing personal data, after explaining the nature of research.
From each paraffin block, 5-μm-thick sections were mounted on three charged slides. They were then deparaffinized and rehydrated, and subsequently microwave-treated in sodium citrate buffer (pH 6.0) twice. Endogenous peroxidase activity was blocked with 3% H2O2 for 15 min, followed by washing with Tris-buffered saline.
As regards CD immunostaining, the sections were incubated with CD antibody (A00032-IFU-IVD; Scy Tek laboratories, South West Logan, USA), which was supplied in liquid form ready to use without dilution for 1 h at room temperature.
As for BCL-2, two drops of BCL-2 antibody (clone 100/D5) (Cat. #MS-123-P; Lab vision, USA), diluted at 1 : 50, were placed on each slide.
The immunostaining for Ki-67 was done using Ki-67 antibody (Cat. #RB-9043-P; Lab Vision) diluted at 1 : 300.
The sections were then refrigerated at 4°C overnight in a closed humid chamber. The sections were again washed in Tris-buffered saline and incubated with the avidin–biotin–peroxidase system (DAKO, Glostrup, Denmark) for 30 min. Diaminobenzidine was used as a chromogen and hematoxylin as a counterstain. Positive controls for both CD and Ki-67 were obtained from invasive duct carcinoma breast sections. As for BCL-2, paraffin sections from BCL-2-positive follicular lymphoma were used. Negative controls were sections from the same cases immunohistochemically processed omitting the step of incubation with the primary antibodies.
Evaluation of immunostaining
Sections were examined microscopically at high-power magnification for detection of cytoplasmic staining of CD in tumor and stromal cells. A semiquantitative score was used to assess CD immunoreactivity in tumor cells: 0, negative (no stained tumor cells); +1, mild (<25% stained tumor cells); +2, moderate (25–50% stained tumor cells); and +3, marked (>50% stained tumor cells). Stromal cells were considered positive if more than 15% of cells were immunostained (Sis et al., 2004).
Cytoplasmic staining of BCL-2 was evaluated and semiquantitatively scored as a percentage of positively stained tumor cells. The extent of immunoreactivity was graded as negative (0 to <25%), mild (25 to <50%), moderate (50 to <75%), or marked (75–100%) (Qiu et al., 2006).
Homogeneous or granular nuclear staining for Ki-67 was considered positive. The area of highest marking (hot spot) was selected for evaluation of Ki-67 immunostaining. In each section at least five fields at high-power magnification were examined and 500 cells were counted. The Ki-67 proliferative index was defined as the percentage of positively stained tumor nuclei among the total number of malignant cells assessed (Schwartz-Roberts et al., 2013).
For numeric variables the measures for central tendency (mean, median) were calculated. Correlation for nominal variables was determined by χ2-tests. For multivariate analysis (more than two groups) Analysis of variance was used. P values less than 0.05 were considered statistically significant.
A total of 40 randomly collected CRC cases were studied. The age of the patients ranged from 26 to 73 years, with a mean age of 49.75 years and a median age of 50 years.
Cytoplasmic CD immunostaining was detected in both tumor and stromal cells (Figs 1 and 2). Tumor cells were positively stained for CD in 36 (90%) out of 40 cases: score +1 in two (5%) cases, score +2 in 15 (37.5%) cases, and score +3 in 19 (47.5%) cases. CD immunostaining was positive in peritumoral stromal cells in 37 (92.5%) cases. A significant correlation was detected between CD expression in stromal cells and depth of tumor invasion (T stage) (P=0.004) and between CD expression in tumor cells and clinical presentation (P=0.04), as most of the cases with +3 score presented clinically with bleeding per rectum. The clinicopathological variables and their correlation with CD immunostaining in tumor and stromal cells are summarized in Table 1.
BCL-2 cytoplasmic immunostaining was observed in tumor cells in 18 (45%) out of 40 cases (Fig. 3). Among positive cases mild staining was observed in 13 (32.5%) cases, moderate staining in four (10%) cases, and marked staining in one (2.5%) case. The correlation between BCL-2 immunostaining and clinicopathologic variables was nonsignificant (Table 2). No significant correlation was detected between BCL-2 immunohistochemical expression in CRC cases and CD expression in both tumor and stromal cells (P>0.05) (Table 3).
Regarding Ki-67 expression, Ki-67 proliferation index values ranged from 0 to 100%, with mean 64.88±32.375% and median 75% (Fig. 4). Level of proliferation showed nonsignificant correlation with all clinicopathological parameters included in the study (P>0.05). Table 4 shows Ki-67 proliferation indices in relation to immunohistochemical CD and BCL-2 expression. We found variation in the relationship of the extent of BCL-2 expression with mean Ki-67 score, and this difference nearly reached statistical significance (P=0.053).
Determining prognostic markers for CRC is essential for determining potential targets for therapeutic intervention (Kuester et al., 2008). In the current study we examined CD immunoreactivity in both cancer and stromal cells in 40 CRC cases. Tumor cells were positively stained for CD in 36 (90%) out of 40 cases. CD immunostaining was positive in peritumoral stromal cells in 37 (92.5%) out of 40 cases. A statistically significant relationship was detected between scoring of CD expression in tumor cells and clinical presentation, as most of the cases with +3 score presented clinically with bleeding per rectum. This may be explained by the fact that CD acts by degradation of the ECM and basal membranes (Berchem et al., 2002), thus causing vascular injury and bleeding. Also statistical significance was attained between CD expression in stromal cells and depth of invasiveness and thus tumor (T) stage. Sis et al. (2004) similarly reported a significant relationship in terms of correlation with depth of invasion.
This means that CD overactivity is implicated in the process of deeper invasion of CRC, and subsequently metastasis, which explains the clinical presentations and advanced T stage of tumor associated with its expression.
The level of CD expression in tumor cells in our study was 90%, which is higher than that reported by others. Kirana et al. (2012) reported an expression of 46.3% in the main tumor bulk and Shin et al. (2014) reported an expression of 38.7%. This may be attributed to genetic and geographic variability, differences in tissue processing and immunohistochemical techniques, different primary antibodies used, or the incorporation of score (+1) in our study, which was disregarded from positive cases in some studies, such as that of Kirana et al. (2012).
Our results are in concordance with those of Kirana et al. (2012), who found that CD expression in tumor cells tended to correlate with higher T (as well as N&M) stage, although this did not reach statistical significance (P=0.06), and with those of Kun-Lun et al. (2010), who found the expression of CD in CRC was significantly higher in Dukes C stage than in Dukes A or B stage (P=0.05).
Our study showed that 100% of signet ring cell carcinomas (which are poorly differentiated) have more intense CD tumor cell expression (score +3) compared with 50% of mucinous adenocarcinomas and 41.2% of conventional adenocarcinomas. However, no statistical significance was reported between different histopathological types of CRC and CD expression in tumor cells. This result was in contrast to that reported by Oh-e et al. (2001) and was in concordance with the findings of Kun-Lun et al. (2010).
The present work demonstrated that 45% of the studied CRC cases were BCL-2 positive. Most showed mild BCL-2 positivity. These results were nearly similar to those of Poincloux et al. (2009) in whose study 36% of CRCs were BCL-2 positive, mostly mildly positive. Watson et al. (2005) also noted BCL-2 cytoplasmic staining in 45% of their cases. Our results were lower than those of Hilska et al. (2005), Qiu et al. (2006), and Shanmugam et al. (2008), who found positive BCL-2 immunoreactivity in 69, 52.8, and 69% of their cases, respectively. On the other hand, Menezes et al. (2010) reported a somewhat slightly lower figure (31.7%) than the one recorded in the present work. Such differences may be attributed to genetic and geographic variability, differences in tissue processing and immunohistochemical techniques, different primary antibodies used, and scoring with a different setting of threshold scores among different studies.
BCL-2 expression in CRC cases in the present study did not statistically correlate with the studied clinicopathological factors. This corresponds with published results from a study conducted by Hilska et al. (2005), Poincloux et al. (2009), and Menezes et al. (2010). However, Watson et al. (2005) and Nehls et al. (2007) showed a statistically significant correlation between BCL-2 and high grades of cell differentiation. Torsello et al. (2008) reported BCL-2 expression as an independent prognostic parameter of Dukes’ classification or TNM stage following multivariate analysis.
Regarding Ki-67 expression, Ki-67 proliferation index values ranged from 0 to 100%, with a mean of 64.88±32.375% and median 75%. These results were similar to some extent to those reported by Terzi et al. (2008), who stated that the percentage of Ki-67-positive cells in the tumor ranged from 20 to 98%, with a mean of 84% and a median of 89%. Ihmann et al. (2004) and Hilska et al. (2005) detected much lower figures for mean and median Ki-67 proliferation index. Such discrepancy was explained by Watson et al. (2005), who stated that quantitative scoring of immunostaining for practical/diagnostic purposes is particularly a very difficult affair and liable to much subjectivity.
The immunohistochemical expression of Ki-67 did not present statistically significant correlation with any of the studied clinicohistopathological parameters. These results were in agreement with the findings of Ihmann et al. (2004) and Menezes et al. (2010). However, Wei-Mei et al. (2009) reported significantly higher Ki-67 expression in patients with lymph node metastasis than in those with non-lymph-node metastasis.
We found variation in the relation of the extent of BCL-2 expression with mean Ki-67 score, and this difference nearly reached statistical significance (P=0.053). This agrees with the findings of Hilska et al. (2005), who found significant correlation, and disagrees with those of Menezes et al. (2010), who did not.
Ki-67 did not correlate with either cancer cell or stromal expression of CD. Shin et al. (2014) reached the same results in CRC. This was not the case with Ioachim et al. (1999), who found that CD expression in cancer cells was positively correlated with proliferation indices (proliferating cell nuclear antigen score).
Our study showed no significant correlation between BCL-2 immunohistochemical expression in CRC cases and CD expression in both tumor and stromal cells. This agrees with Ioachim et al. (1999). However, this point needs thorough evaluation as direct targeting of the antiapoptotic BCL-2 members with GX15-070 (GX; obatoclax) inhibits downstream CD protein expression to limit the ability of cells to use degraded material to fuel cellular metabolism. Thus, the feasibility of this drug for CRC is worth studying.
CD immunoreactivity was detected in a high percentage of tumor cells and stromal cells of CRCs and there were statistically significant relationships regarding the association of CD expression in stromal cells with deeply invasive tumors and CD expression in tumors with clinical presentations, which confirms its role in the invasiveness and dissemination of cancer cells. BCL-2 expression was detected at a relatively lower percentage in tumor cells; however, it nearly correlated with high proliferation indices, suggesting its role in the progression of CRC. Further studies are required to rule out a link between CD and BCL-2 in CRC for possible therapeutic targeting by GX15-070 (inhibits both CD and BCL-2), which is under investigations in clinical trials for other tumors.
Conflicts of interest
There are no conflicts of interest.
Adams JM, Cory S (2007). The Bcl-2 apoptotic switch in cancer development and therapy. Oncogene 26:1324–1337.
Berchem G, Glondu M, Gleizes M, Brouillet JB, Vignof F, Garcia M, et al (2002). Cathepsin-D affects multiple tumor progression steps in vivo: proliferation, angiogenesis and apoptosis. Oncogene 21:5951–5955.
Calongi N, Pettiti D, Dewwih T (2008). Screening for colorectal cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 149:627–637.
Hilska M, Collan YU, Laine VJO, Kössi J, Hirsimäki P, Laato M, et al (2005). The significance of tumor markers for proliferation and apoptosis in predicting survival in colorectal cancer. Dis Colon Rectum 48:2197–2208.
Ihmann T, Liu J, Schwabe W, Hausler P, Behnke D, Bruch HP, et al (2004). High-level mRNA quantification of proliferation marker pKi-67 is correlated with favourable prognosis in colorectal carcinoma. J Cancer Res Clin Oncol 130:749–756.
Ioachim EE, Goussia AC, Machera M, Tsianos EV, Kappas AM, Agnantis NJ (1999). Immunohistochemical evaluation of cathepsin D expression in colorectal tumours: a correlation with extracellular matrix components, p53, pRb, bcl-2, c-erbB-2, EGFR and proliferation indices. Anticancer Res 19:2147–2155.
Kirana C, Shi H, Laing E, Hood K, Miller R, Bethwaite P, et al (2012). Cathepsin D expression in colorectal cancer: from proteomic discovery through validation using western blotting, immunohistochemistry, and tissue microarrays. Int J Proteomics 2012:245819.
Krajewska M, Kim H, Kim C, Kang H, Welsh K, Matsuzawa S, et al (2005). Analysis of apoptosis protein expression in early-stage colorectal cancer suggests opportunities for new prognostic biomarkers. Clin Cancer Res 11:5451–5461.
Kuester D, Lippert H, Roessner A, Krueger S (2008). The cathepsin family and their role in colorectal cancer. Pathol Res Pract 204:491–500.
Kun-Lun Y, Zhi-Jian D, Hui-Jie H, Young L (2010). Expression of cathepsin D in colorectal cancer. Inner Mongolia Med J 01:22–25.
Lin B, Kolluri SK, Lin F, Liu W, Han YH, Cao X, et al (2004). Conversion of Bcl-2 from protector to killer by interaction with nuclear orphan receptor Nur77/TR3. Cell 116:527–540.
Menezes HL, Jucá MJ, Gomes EGA, Nunes BLBBP, Costa HO, Matos D (2010). Analysis of the immunohistochemical expressions of p53, bcl-2 and Ki-67 in colorectal adenocarcinoma and their correlations with the prognostic factors. Arq Gastroenterol 47:141–147.
Mokhtar N, Gouda I, Adel I (2007). Cancer Pathology Registry 2003–2004 and time trend analysi. Cairo, Egypt: Department of Pathology, National Cancer Institute, Cairo University. 56.
Nehls O, Okech T, Hsieh CJ, Enzinger T, Sarbia M, Borchard F, et al (2007). Studies on p53, BAX and Bcl-2 protein expression and microsatellite instability in stage III (UICC) colon cancer treated by adjuvant chemotherapy: major prognostic impact of proapoptotic BAX. Br J Cancer 96:1409–1418.
Oh-e H, Tanaka S, Kitadai Y, Shimamoto F, Yoshihara M, Haruma K (2001). Cathepsin D expression as a possible predictor of lymph node metastasis in submucosal colorectal cancer. Eur J Cancer 37:180–188.
Poincloux L, Durando X, Seitz JF, Thivat E, Bardou VJ, Giovannini MH, et al (2009). Loss of Bcl-2 expression in colon cancer: a prognostic factor for recurrence in stage II colon cancer. Surg Oncol 18:357–365.
Qiu CZ, Wang C, Huang ZX, Zhu SZ, Wu YY, Qiu JL (2006). Relationship between somatostatin receptor subtype expression and clinicopathology, Ki-67, Bcl-2 and p53 in colorectal cancer. World J Gastroenterol 12:2011–2015.
Ruggiero MSchwab M (2009). Bcl-2. Encyclopedia of cancer. 2nd ed. Berlin; Heidelberg; New York: Springer-Verlag. 309–312.
Schwartz-Roberts JL, Shajahan AN, Cook KL, Wärri A, Abu-Asab M, Clarke R (2013). GX15-070 (obatoclax) induces apoptosis and inhibits cathepsin D and L mediated autophagosomallysis in antiestrogen resistant breast cancer cells. Mol Cancer Ther 12:448–459.
Scopa CD, Vagianos C, Kardamakis D, Kourelis TG, Kalofonos HP, Tsamandas AC (2001). Bcl-2/bax ratio as a predictive marker for therapeutic response to radiotherapy in patients with rectal cancer. Appl Immunohistochem Mol Morphol 9:329–334.
Scopa CD, Tsamandas AC, Zolota V, Kalofonos HP, Batistatou A, Vagianos C (2003). Potential role of bcl-2 and Ki-67 expression and apoptosis in colorectal carcinoma: a clinicopathologic study. Dig Dis Sci 48:1990–1997.
Sebzda T, Saleh Y, Gburek J, Andrzejak R, Gnus J, Siewinski M, et al (2005). Cathepsin D expression in human colorectal cancer: relationship with tumour type and tissue differentiation grade. J Exp Ther Oncol 5:145–150.
Shanmugam C, Katkoori VR, Jhala NC, Grizzle WE, Siegal GP, Manne U (2008). p53 nuclear accumulation and Bcl-2 expression in contiguous adenomatous components of colorectal adenocarcinomas predict aggressive tumor behavior. J Histochem Cytochem 56:305–312.
Shin IY, Sung NY, Lee YS, Kwon TS, Si Y, Lee YS, et al (2014). The expression of multiple proteins as prognostic factors in colorectal cancer: cathepsin D, p53, COX-2, epidermal growth factor receptor, C-erbB-2, and Ki-67. Gut Liver 8:13–23.
Sis B, Sagol O, Küpelioğlu A, Sokmen S, Terzi C, Fuzun M, et al (2004). Prognostic significance of matrix metalloproteinase-2, cathepsin D, and tenascin-C expression in colorectal carcinoma. Pathol Res Pract 200:379–387.
Terzi C, Canda AE, Sagol O, Atila K, Sonmez D, Fuzun M, et al (2008). Survivin, p53, and Ki-67 as predictors of histopathologic response in locally advanced rectal cancer treated with preoperative chemoradiotherapy. Int J Colorectal Dis 23:37–45.
Tetu B, Brisson J, Wang G (2002). Expression of cathepsin D,stromoyelin-3, and urokinase by reactive stromal cells on breast carcinoma prognosis. Cancer 92:2957–2964.
Torsello A, Garufi C, Cosimelli M, Diodoroc MG, Zeuli M, Vanni B, et al (2008). P53 and bcl-2 in colorectal cancer arising in patients under 40 years of age: distribution and prognostic relevance. Eur J Cancer 44:1217–1222.
Watson NFS, Madjd Z, Scrimegour D, Spendlove I, Ellis IO, Scholefield JH, et al (2005). Evidence that the p53 negative/Bcl-2 positive phenotype is an independent indicator of good prognosis in colorectal cancer: a tissue microarray study of 460 patients. World J Surg Oncol 3:47.
Wei-Mei L, Jian-Min L, Lin F (2009). The expression of bcl-2 protein and ki-67 protein in colon cancer and precancerous lesions and its clinical significance. J Chin Pract Med 2.