Expression of smooth muscle calponin in tumor vessels of human hepatocellular carcinoma and its possible association with prognosis
Version of Record online: 15 MAR 2002
Copyright © 2002 American Cancer Society
Volume 94, Issue 6, pages 1777–1786, 15 March 2002
How to Cite
Sasaki, Y., Yamamura, H., Kawakami, Y., Yamada, T., Hiratsuka, M., Kameyama, M., Ohigashi, H., Ishikawa, O., Imaoka, S., Ishiguro, S. and Takahashi, K. (2002), Expression of smooth muscle calponin in tumor vessels of human hepatocellular carcinoma and its possible association with prognosis. Cancer, 94: 1777–1786. doi: 10.1002/cncr.10402
- Issue online: 15 MAR 2002
- Version of Record online: 15 MAR 2002
- Manuscript Accepted: 26 NOV 2001
- Manuscript Revised: 8 OCT 2001
- Manuscript Received: 22 MAY 2001
- hepatocellular carcinoma;
- tumor vessel;
- prognostic factor;
- prognosis after hepatic resection
Hepatocellular carcinoma (HCC) is a vascular-rich tumor. The tumor vessels in HCC were demonstrated to have α-smooth muscle actin positive smooth muscle cells (SMCs). However, it is unclear whether the SMCs in the wall of the tumor vessels are differentiated or undifferentiated. Basic calponin is an actin-, tropomyosin-, and calmodulin-binding protein, and expression of the calponin gene in SMCs has been recognized as one of the late stage differentiation markers of SMCs. The authors investigated the differentiation state of SMCs in tumor vessels by immunohistochemical examination of calponin in patients with HCC, and whether it is associated with the patients' prognosis.
Tumor and nontumor tissues were obtained from 75 patients with HCC who underwent radical hepatic resection. The differentiation state of the smooth muscle cells were evaluated based on the expression level of calponin, an actin-binding protein, using immunohistochemistry and reverse transcription–polymerase chain reaction analysis. The disease free survival (DFS) rates were estimated according to the Kaplan–Meier method comparing groups of patients with calponin positive and negative tumor vessels. A multivariate analysis based on the Cox proportional hazards regression model was performed to estimate whether the expression of calponin is an independent prognostic factor.
In the 75 patients with HCC examined, 36 patients (48%) possessed calponin positive SMCs, and the remaining 39 (52%) did not. There were no significant differences in either clinical or pathologic factors between the two groups of patients. The 5- and 8-year DFS rate of the patients with calponin positive vessels were 37% and 26%, respectively. These values were significantly higher (11% and 5%) than those of patients with calponin negative vessels. Gender, TNM classification, perioperative transfusion, and calponin expression were found to be independent prognostic factors for DFS.
Immunohistochemical examination of the calponin expression in the tumor vessels is a new and useful means to predict the prognosis of HCC patients after hepatic resection. Cancer 2002;94:1777–86. © 2002 American Cancer Society.
Generally, tumor vessels are morphologically and physiologically different from normal vessels. On the basis of histologic observations of various experimental tumors, Krylova1 reported that tumor vessels in the sprout formation were composed of only epithelium, expressing the markers of endothelial cells such as CD31, ulex europaeus 1 lectin (UEA1), and factor VIII.2–6 Afterward, these tumor vessels were immunohistochemically proven to have smooth muscle cells (SMCs) that are positive for α- smooth muscle actin (SMA).7
Hepatocellular carcinoma (HCC) is a vascular-rich tumor neoplasm, and the appearance of abundant new tumor vessels is one of the important characteristics of HCC.4 Terada and Nakanuma8 showed in differentiated HCC that the intratumor arteries or arterioles, without accompanying obvious fibrous tissue or portal veins, were immunohistochemically stained with an anti-α-SMA antibody. However, it is unclear whether the tumor vessels that have α-SMA positive SMCs are undifferentiated or differentiated. The relation between microvessel density (MVD) in HCC and its prognosis has been elucidated. Tanigawa et al.,9 and El-Assal et al.10 reported that the intratumor MVD was the independent prognostic factor for overall survival or DFS by using the markers of endothelial cells. To our knowledge, however, there have been no reports that have investigated the relation between the degree of phenotypic differentiation of SMCs in tumor vessels and the malignant potential of HCC.
Basic or h1 calponin, herein referred to as calponin, is an actin-, tropomyosin-, and calmodulin-binding protein that originally was isolated from the smooth muscles of bovine aortas and chicken gizzards.11, 12 In normal tissues, calponin is most highly expressed in terminally differentiated and nonreplicating SMCs, whereas proliferation of SMC down-regulates its expression.13 Expression of the calponin gene in SMCs during embryonic and postnatal development has been recognized as one of the differentiation markers of SMCs.14
In this study, we examined expression of the calponin protein and mRNA in α-SMA positive SMC in tumor vessels of patients with HCCs by immunohistochemistry and reverse transcription–polymerase chain reaction (RT-PCR) analyses. We also investigated whether the expression of the calponin protein in tumor vessels is correlated with the clinical prognosis of patients with HCC.
PATIENTS AND METHODS
In patients with HCC, who underwent radical hepatic resection between April 1990 and March 1993 at the Osaka Medical Center for Cancer and Cardiovascular Diseases, 75 patients were randomly selected for this study. After an initial review of all available hematoxilin and eosin–stained slides of surgical specimens, one representative paraffin block from each case was taken for further study. Consecutive 3-μm sections were resected from each block. One section from each case was stained with hematoxilin and eosin, and other sections were immunostained for α-SMA and calponin.
Clinical data were picked up from each chart, and prognostic factors were collected retrospectively. Binary covariates were obtained from available data on the patients and evaluated for disease free survival (DFS).
Immunohistochemical staining was performed by the immunoperoxidase technique following predigestion and trypsinization. Antibodies used were a monoclonal antibody at a 1:1000 dilution for calponin (Sigma Chemical Co., St. Louis, MO) and a monoclonal antibody at a 1:50 dilution for α-SMA (Dako, Glostrup, Denmark). The sections were mounted on poly-L-lysin coated microslides, deparaffinized in xylene, dehydrated through graded alcohol and immersed in 70% methanol with H2O2 to block endogeneous peroxidase. Antigen retrieval for calponin was performed using a 400-W microwave oven (Toshiba ERT 330, Tokyo, Japan) for 5 minutes (4 times) in a 10-mM citrate buffer (pH 7.0). The sections were incubated with 1% (v/v) goat serum/PBS for 1 hour at room temperature, washed in PBS, and incubated with the antibody in 2% (w/v) bovine serum albumin (BSA)/phosphate-buffered saline (PBS) overnight at 4 °C. They then were washed 5 times with 0.005% (v/v) Tween 20/PBS and subsequently followed by incubation with biotinylated goat anti-mouse immunoglobulin (Ig) G (TAGO Immunologicals, Camarillo, CA) in 2% (w/v) BSA/PBS for 1 hour at room temperature and then avidin-biotin-horseradish peroxidase complex (Vector Laboratories, Burlingame, CA) for 30 minutes at room temperature. After being washed in 0.005% (v/v) Tween 20/PBS, the final reaction products were visualized with diaminobenzidine (WAKO Chemicals, Osaka, Japan), and the sections were counterstained with hematoxylin. A negative control slide with mouse nonimmune IgG was included to assess nonspecific staining.
Evaluation of Immunostaining
Walls of blood vessels were defined by the presence of α-SMA positive cells. Presence or absence of calponin was evaluated on the walls of blood vessels in the parenchyma of tumor tissues. Ito cells, which also could be stained for α-SMA and calponin15 in the metaplastic state, were distinguishable from SMCs surrounding blood vessels, because Ito cells are predominantly located in the perisinusoidal region, and their distribution is not as uniform and concentric as the walls of the blood vessels. The blood vessels surrounded by connective tissues existing near bile ducts were excluded from being the subjects of evaluation because these vessels are considered to be normal pre-existing vessels of portal triads. In every case, using a ×100 magnification and without referring to the patient's clinical data, the number of calponin-stained blood vessels per square millimeter was counted in the five fields with the highest concentration of tumor vessels. When total number of the calponin-stained vessels in these 5 fields exceeded 10, the HCC was assessed as calponin positive. The representative histologic and immunohistochemical findings of the cancerous area and the adjacent noncancerous area in HCC with calponin positive tumor vessels are shown in Fig. 1. Those in HCC with calponin negative tumor vessels are shown in Figure 2.
Detection of Calponin mRNA
Total RNA was extracted from fresh frozen tissues of both the cancerous area and the surrounding noncancerous area using the Isogen RNA extraction kit (Nippon Gene, Toyama, Japan) in each representative case having either calponin positive or calponin negative tumor vessels. Reverse transcription–PCR analysis was conducted by the previously reported method.16
Statistical analysis was performed using a Power Macintosh G4 and StatView 5.0 software (SAS Institute, Berkley, CA) and finally reviewed by a statistician from our institute. The statistical analysis was performed with Student t test for unpaired observations and the chi-square test for the frequency of various attributes between the groups. Survival and DFS curves were analyzed using the Kaplan–Meier method. Differences between curves were assessed according to the log-rank test. In univariate analysis, statistical comparisons between the subgroups of patients were made using the Mantel–Cox test. The Cox proportional hazards regression model was used for multivariate analysis. Differences with a P value less than 0.05 were considered significant. All P values were two-tailed.
Of the examined 75 patients, 36 patients (48%) were assessed as calponin positive (the number of the calponin-stained blood vessels ranged from 16 to 92; median, 38), and the remaining 39 (52%) were assessed as negative (the number of the calponin-stained blood vessels was ≤ 10). Clinicopathologic factors were compared between the HCC patients with calponin positive and negative tumor vessels. There were no significant differences in both clinical and pathologic factors between the two groups (Table 1).
|Variable||Calponin-positive tumor vessels||Calponin-negative tumor vessels||P value|
|Clinical or laboratory factor|
|Age (yrs)||60.3 ± 9.9||62.9 ± 8.4||0.23|
|Albumin (g/dL)||3.74 ± 0.32||3.64 ± 0.31||0.16|
|T. bilirubin (mg/dL)||0.72 ± 0.45||0.92 ± 0.42||0.05|
|ICG-R15 (%)||15.0 ± 7.3||20.4 ± 8.3||0.19|
|Prothronbin time (%)||81.5 ± 11.9||83.8 ± 15.7||0.48|
|AFP (ng/mL)||761 ± 2670||634 ± 2540||0.83|
|Child's classification (A/B)||24/12||22/17||0.36|
|No. of nodules (single/multiple)||26/10||28/11||0.97|
|Dimension (cm)||2.9 ± 1.7||3.6 ± 1.9||0.10|
|Width of margin (≥ 10 mm/< 10 mm)||14/22||12/27||0.46|
|Volume of resection (g)||170 ± 139||179 ± 140||0.79|
|Intraoperative blood loss (mL)||1240 ± 970||1150 ± 890||0.66|
|Operation time (min)||220 ± 60||230 ± 60||0.80|
|Intraoperative transfusion (mL)||420 ± 780||390 ± 660||0.84|
|Postoperative transfusion (+/−)||6/30||5/34||0.64|
|Edmondson–Steiner classification (I/II/III)||11/19/6||11/25/3||0.43|
|Growth type (expansive/infiltrative)||28/8||28/11||0.55|
|Intracapsular invasion (+/−)||22/14||23/16||0.78|
|Extracapsular invasion (+/−)||20/16||16/23||0.21|
|Satellite nodule (+/−)||9/27||13/26||0.39|
|Vascular invasion (+/−)||14/22||15/24||0.97|
The survival and the DFS were compared between the HCC patients with calponin positive and negative tumor vessels. The 5- and 8-year survival rates of the patients with the calponin positive tumor vessels were 72% and 51%, respectively, and those of the patients with calponin negative vessels were 62% and 42%, respectively. The survival of the former tended to be better than that of the latter (P = 0.052; Fig. 3). Conversely, the 5- and 8-year DFS rates of the patients with calponin positive tumor vessels, 37% and 26%, respectively, were significantly better than those with calponin negative tumor vessels, 11% and 5%, respectively (P = 0.02; Fig. 4). The DFS rates between the calponin positive group and the negative group were separately compared in TNM Stage I and II.16 (Liver Cancer Stage I [T1N0M0]: solitary tumor 2 cm or less in greatest dimension without vascular invasion (T1), no regional lymph node metastasis [N0], no distant metastasis [M0], and Stage II [T2N0M0]: solitary tumor 2 cm or less in greatest dimension with vascular invasion, or multiple tumors limited to one lobe, none more than 2 cm in greatest dimension without vascular invasion.17) In Stage I, the DFS of calponin positive group had a better tendency than that of the negative group (P = 0.08). Conversely, in Stage II, there was no significant difference between the DFS of the two groups (Fig. 5).
In univariate analysis for the DFS, gender, hepatitis B surface antigen, number of nodules, TNM classification, operation time, perioperative transfusion, Edmondson–Steiner classification, and calponin-antigen in tumor vessels were considered as prognostic factors (Table 2). Multivariate analysis was performed for DFS using the variables of which the P values were less than 0.05 in univariate analysis. The calponin antigen in tumor vessels was one of the independent prognostic factors for DFS together with gender, TNM classification, and perioperative transfusion (Table 3).
|3-yr (%)||5-yr (%)||P value|
|Age (yrs)||≤ 62||35||46||33||0.2|
|Albumin (g/dL)||< 3.7||40||40||31||0.68|
|T. bilirubin (mg/dL)||≤ 0.8||42||38||20||0.14|
|ICG-R15 (%)||< 20||37||49||29||0.35|
|Prothrombin time (%)||≤ 83||38||50||32||0.17|
|AFP (ng/mL)||< 40||38||42||18||0.69|
|Type of resection||Atypical||37||41||19||0.62|
|No. of nodules||Single||54||57||32||< 0.0001|
|Width of margin (mm)||≥ 10||26||62||31||0.061|
|Stage||I + II||55||56||32||< 0.0001|
|III + IV||20||10||0|
|Dimension (cm)||≤ 2.5||38||50||27||0.52|
|Volume of resection||≤ 120||37||43||27||0.83|
|Intraoperative blood loss (mL)||< 850||37||54||29||0.24|
|Operation time (min)||< 220||39||51||31||0.024|
|II + III||53||38||17|
|Calponin in tumor vessels||Positive||36||56||37||0.02|
|Variable||Coefficient||SE||Coefficient/SE||Chi-square||P value||Odds ratio||95% CI|
|No. of nodules (multiple)||0.002||0.596||0.003||1.115||0.9973||1.002||0.311–3.226|
|Operation time >220 min||0.139||0.349||0.398||0.158||0.6906||1.149||0.580–2.276|
|Perioperative transfusion (yes)||0.993||0.401||2.474||6.123||0.0133||2.699||1.229–5.927|
|Edmondson–Steiner classification II + III||0.439||0.376||1.167||1.362||0.2431||1.551||0.742–3.240|
|Calponin in tumor vessels (negative)||0.892||0.297||3.008||9.047||0.0026||2.441||1.364–4.365|
RT-PCR analysis of the cancerous and the noncancerous area was conducted in each representative case of the tumor vessels showed high or low expressions of calponin in the immunohistochemical analysis. The calponin mRNA was expressed at higher levels in cancerous tissues with calponin positive tumor vessels than in those with calponin negative vessels, as well as in the noncancerous tissues of HCCs with either calponin positive or negative tumor vessels (Fig. 6).
Various kinds of indicators have been reported as prognostic factors for patients with HCC after hepatic resection, including clinical factors such as age, gender, and liver function of the patients, tumor factors such as size, number and TNM classification of the tumors, and the concentrations of serum α-fetoprotein, and surgical factors such as surgical margin and perioperative transfusion.18–26 Moreover, the degree of histologic differentiation,25 the presence of satellite nodules23 or vascular invasion,18, 23, 26 the existence of capsular formation,21, 27 and the abnormal presence of nuclear DNA contents24, 26, 28 are known to or suspected to predict the prognosis as pathologic factors. In this study, we reported that the expression of calponin in tumor vessels could be one of the prognostic factors that affect DFS independently.
Immunohistochemical staining with anti-α-SMA antibody clearly showed that some of the HCC vessels consisted not only of endothelial cells but also of surrounding SMCs. Judging from the weakness of response to catecholamine29, 30 or angiotensin II,31 however, they may be in an immature or undifferentiated state. These findings suggest that the existence of SMC does not always show well differentiated tumor vessels. To distinguish the degree of differentiation of tumor vessels, we examined the expression of calponin, which is one of the late differentiation markers of SMCs. In the current study, for the first time to our knowledge, we showed that there exist two types of HCC tumor vessels, i.e., differentiated or undifferentiated, as judged from tumors vessels surrounded by calponin positive or negative SMC.
The HCC patients with mature tumor vasculatures manifested by calponin positive SMC had better prognosis in DFS than those with immature vessels containing calponin negative SMC. Of note, the difference was recognized more clearly in Stage I than in Stage II. This result indicates that the influence of tumor vessel maturity on prognosis is diminished in more advanced stages. We speculate that the other prognostic factors become stronger in relation to tumor progression. In overall survival, patients with mature tumor vasculatures tended to show a better prognosis than those with immature vessels. However, there was no significant difference between the two groups. The active therapy for recurrent foci such as repeated surgery or transcatheter arterial chemoembolization32 may effectively prolong the survival of patients and consequently may hide the influence of the expression of calponen in tumor vessels of the initial lesions in regards to survival. On this basis, we suggest that immunohistochemical examination of the expression of calponin in the tumor vessels is a new means for evaluating the prognosis of the patients with HCC.
A fundamental question is the mechanism of association between calponin expression in the tumor vessels and long-term prognosis of the patients. There are several possibilities. For example, platelet derived growth factor (PDGF)–induced chemotaxis of the calponin-deficient SMC is greatly reduced (H. Koyama, H. Yamamura, and K. Takahashi, unpublished observation), and thus the integrity of the vessels with calponin negative SMC may be impaired. Also, it has been shown that PDGF-β and PDGF receptor-β play a crucial role in the recruitment of SMCs and pericytes during embryonic blood vessel formation.33 In addition, we have reported recently that loss of calponin in SMC altered the contractile properties of SMC with increased shortening velocity.15, 33 These properties also may influence the structural integrities of the blood vessels and consequently may increase the permeability of local microvessels, which could facilitate the extravasation and migration of cancer cells.
To the best of our knowledge, there have been no previous reports on calponin expression in tumor vessels. Ono et al.35 demonstrated that calponin was uniformly down-regulated in SMC surrounding the tumor neovasculature of synovial sarcomas. It has been reported that the calponin mRNA is up-regulated by transforming growth factor (TGF)-β.36 Bedossa et al.37 reported that the cytoplasm of HCC tumor cells showed more intense staining for TGF-β1 mRNA and the protein than that of normal hepatocytes or hepatocytes in cirrhotic nodules. Moreover, Shirai et al.38 reported that the levels of plasma TGF-β1 are significantly higher in patients with HCC than in patients with normal liver or chronic hepatitis. In the future, it will be worthwhile to examine the relation between the expression of calponin and the levels of TGF-β1 expression in HCC tumor cells.
In conclusion, the expression of basic calponin in the tumor vessels is a good prognostic marker for patients with HCC. Immunohistochemical examination of the expression of calponin protein in resected specimens is a new diagnostic strategy useful in predicting the prognosis of a patient with HCC after hepatic resection. This procedure, if conducted with the biopsy specimens before surgery, may also be useful in choosing the most appropriate surgical methods or perioperative adjuvant therapies.
- 6An immunohistochemical study of the blood vessels within primary hepatocellular tumors. Liver. 1992; 79: 245–250., , , .
- 10Clinical significance of microvessel density and vascular endothelial growth factor expression in hepatocelluilar carcinoma and surrounding liver: possible involvement of vascular endothelial growth factor in the angiogenesis of cirrhotic liver. Hepatology. 1998; 27: 1554–1562., , , et al.
- 17TNM classification of malignant tumors. 4th, fully revised ed. International Union Against Cancer. Berlin: Springer-Verlag; 1987.
- 18The Liver Cancer Study Group of Japan. Predictive factors for long-term prognosis after partial hepatectomy for patients with hepatocellular carcinoma in Japan. Cancer. 1994; 74: 1772–1780.