
	Version PDF

Microvessel density and HIF-1α expression correlate with malignant potential in fibrohistiocytic tumors

European Journal of Dermatology. Volume 15, Number 6, 465-9, November-December 2005, Investigative report

Summary

Author(s) : Kaori Koga, Kazuki Nabeshima, Noriko Nishimura, Mikiko Shishime, Juichiro Nakayama, Hiroshi Iwasaki , Department of Dermatology, School of Medicine, Fukuoka University, Nanakuma 7-45-1, Jonan-ku, Fukuoka 814-0180, Japan., Department of Pathology, School of Medicine, Fukuoka University, Nanakuma 7-45-1, Jonan-ku, Fukuoka 814-0180, Japan.

Summary : Angiogenesis is a central process in the growth of solid tumors. Hypoxia-inducible factor-1α (HIF-1α) is an oxygen-dependent transcriptional activator, which plays a crucial role in tumor angiogenesis. However, involvement of HIF-1α has never been studied in so-called fibrohistiocytic tumors, such as dermatofibroma (DF), dermatofibrosarcoma protuberans (DFSP) and malignant fibrous histiocytoma (MFH). We analyzed the extents of angiogenesis in relation to the expression levels of HIF-1α in 26 DF, 13 DFSP and 23 MFH cases. MFH showed significantly higher microvessel density (MVD) compared with DF and DFSP. Immunohistochemically, HIF-1α-positive cases constituted 31%, 15% and 98% of DF, DFSP and MFH, respectively, indicating significantly higher HIF-1α expression in MFH compared with DF and DFSP. Furthermore, MFH cases expressing high levels of HIF-1α showed significantly higher MVD than those with low levels of HIF-1α. Thus, higher levels of angiogenesis and HIF-1α expression are both closely associated with the malignant potential in so-called fibrohistiocytic tumors, and HIF-1α is possibly involved in angiogenesis in MFH.

Keywords : angiogenesis, dermatofibroma, dermatofibrosarcoma protuberance, malignant fibrous histiocytoma, HIF-1α

Pictures

ARTICLE

Auteur(s) : Kaori Koga^1,2, Kazuki Nabeshima², Noriko Nishimura², Mikiko Shishime², Juichiro Nakayama¹, Hiroshi Iwasaki²

¹Department of Dermatology, School of Medicine, Fukuoka University, Nanakuma 7-45-1, Jonan-ku, Fukuoka 814-0180, Japan.
²Department of Pathology, School of Medicine, Fukuoka University, Nanakuma 7-45-1, Jonan-ku, Fukuoka 814-0180, Japan

accepté le 31 Août 2005

Intra-tumoral angiogenesis is an important process in the progression of solid tumours [1]. Angiogenesis depends on the production of angiogenic factors by neoplastic and normal cells. A number of angiogenic factors, such as vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF), have been identified.Hypoxia-inducible factor-1 (HIF-1) is an oxygen-dependent transcriptional activator, which induces the transcription of more than 60 proteins, including VEGF and erythropoietin that promote angiogenesis [2]. HIF-1 consists of α and β subunits, both are basic helix-loop-helix proteins. The expression of the α subunit is remarkably high during hypoxia and is maintained at low levels in most cells under normoxic conditions [2, 3]. Up-regulation of HIF-1α expression has recently been reported in human cancers in association with tumor progression and angiogenesis [2, 4].Little information is currently available on the prevalence and distribution of HIF-1α in soft tissue tumors. To our knowledge, there is no study on the expression of HIF-1α in so-called fibrohistiocytic tumors such as dermatofibroma (DF), dermatofibrosarcoma protuberans (DFSP) and malignant fibrous histiocytoma (MFH). MFH (undifferentiated high grade pleomorphic sarcoma [5]) is one of the most common soft tissue sarcomas in adult life having an aggressive behavior and a high metastatic potential, although its histogenesis has been a controversial and unresolved issue [6]. DF is known as benign fibrohistiocytic tumors, and DFSP as of intermediate malignancy. In this study, to investigate whether microvessel density (MVD) is closely associated with tumor grades or tumor malignancy in fibrohistiocytic tumors, we studied cases of DF, a benign fibrohistiocytic tumor; DFSP, an intermediately malignant one; and MFH, a highly malignant one. Furthermore, we examined a correlation between expression levels of HIF-1α and extents of angiogenesis in the tumors.

Materials and methods

Tissue samples

This study was performed on the tumor tissues from 62 patients with fibrohistiocytic tumors, including 26 DF (4 males, 22 females; age range, 13 – 60 [mean = 35] years), 13 DFSP (5 males, 8 females; age range, 6 – 68 [mean = 36] years) and 23 MFH (9 males, 14 females; age range, 23 – 85 [mean = 60] years), diagnosed at the Department of Pathology, Fukuoka University, Japan in accordance with Local Ethical Guidelines. Anonymous use of redundant tissue is part of the standard treatment agreement with patients in our hospital when no objection has been made. Each specimen obtained at surgery was fixed in 20% formalin and embedded in paraffin. Histological diagnosis of DF, DFSP and MFH was made according to the widely accepted criteria [6]. All MFH cases showed the storiform-pleomorphic subtype.

Immunohistochemisrty

Antibodies used in this study included monoclonal antibody (MAb) to CD31 (JC/70A, DAKO, Carpinteria, CA) and rabbit polyclonal antibodies to HIF-1α (Santa Cruz Biotechnology, Santa Cruz, CA).

Immunostaining of formalin-fixed, paraffin-embedded tissue sections was performed using a biotin-streptavidin method (for CD31) or Envision labelled polymer reagent (DAKO, Carpinteria, CA) (for HIF-1α) as described before with some modifications [7]. Briefly, sections were deparaffinized, rehydrated in descending alcohol dilutions, and washed in Tris-buffered saline, pH7.6 (TBS). The slides for HIF-1α were immersed in 0.3% hydrogen peroxide in methanol for 30 min at room temperature (RT) to block endogenous peroxidase activity, and placed with citrate buffer (0.01M pH6.0) in a microwave oven (750 W) at 95 °C for 10 minutes for the purpose of antigen retrieval. After non-specific sites were blocked with 3% bovine serum albumin and 1% non-fat dry milk in TBS for 30 min RT, the sections were incubated with the primary antibody overnight at 4 °C. For CD31, the sections were then washed in TBS, and incubated with biotinylated horse anti-mouse IgG (Vector Laboratories, Burlingame, CA) for 30 min RT, followed by streptavidin conjugated to alkaline phosphatase (DAKO) for another 30 min. The reaction was revealed with naphthol AS-BI phosphate (Sigma Chemical Co., St. Louis, MO) in 100 ml of 0.2 M TBS (pH 8.2) containing 4% hydrochloric acid and 4% nitric acid and counterstained with methylgreen. For HIF-1α, the sections were incubated with Envision reagent for 30 min at RT, and the reaction was revealed with 3,3′-diaminobenzidine (DAB) (Sigma Chemical Co., St. Louis, MO), followed by counterstaining with Mayer’s hematoxylin.

The immunohistochemical specificity of the antibody was confirmed by negative control: substituting rabbit non-immune IgG for the primary antibody and omitting the primary antibody in the staining protocol.

Immunostaining was considered negative if less than 10% of the tumor cells failed to stain. In specimens considered positive, staining of the tumor was quantitated on a scale from 1 to 4 based on the percentage of positive tumor cells. The scale was structured as follows: 1+ = 10% to 25%; 2+ = 25% to 50%; 3+ = 50% to 75%; and 4+ = >75%. Furthermore, for HIF-1α immunostaining, two groups of low (<1+, 0-25%) vs. high (>2+, 25-100%) expression were defined as described [8]. All slides were reviewed by the same pathologists.

Microvessel detection and counting

The method for microvessel detection and counting has been described previously [7, 9]. Briefly, intratumor microvessels were highlighted with anti-CD31 MAb as described above. Each sample was then examined under low power magnification (× 40) to identify the region of the section with the highest number of microvessels (“hot spot”). The selected areas were scanned and individual microvessel counts were made on a × 200 field (× 10 objective and × 20 ocular, equivalent to 0.998 mm²). Any endothelial cell or endothelial cell cluster, positive for CD31, with or without a lumen and clearly separated from adjacent microvessels, was considered as an individual vessel.

Statistical analysis

Statistical significance was evaluated with the χ² and Mann-Whitney U-tests. Curves for overall survival were drawn according to the Kaplan-Meier methods, and differences between the curves were analyzed by applying the log-rank test. The significance level was set at 5% for each analysis.

Results

Microvessel density

Staining with anti-CD31 antibody showed intense and clear staining of the endothelium of intra-tumoural vessels, with a low background staining ( (figure 1A-C) ). The extent of MVD (means and standard errors of the mean) in DF, DFSP and MFH are shown in table 1( Table 1 ). MFH showed significantly higher MVD compared with DF (p < 0.0001) and DFSP (p = 0.0157). MVD in DF and DFSP showed no significant difference.
Table 1 MVD and expression of HIF-1α in fibrohistiocytic tumors

Tumors	*MVD (Mean ± SE)**	No of positive cases (%)
HIF-1α
DF
(N = 26)	36.98 ± 5.13	8.13
DFSP
N = 13	48.77 ± 6.11	2(15)
MFH
(N = 23)	72.66 ± 5.98**	21(91)***

HIF1-α expression

HIF-1α immunoreactivity was located in both nuclei and cytoplasm of tumor cells ( (figure 2) ): although cytoplasmic staining was more frequently observed, some tumor cells show intranuclear reactivity (arrows in ( figure 2A-C ), inset in ( figure 2B )). MFH ( (figure 2C) ) showed increased reactivity compared with DF ( (figure 2A) ) and DFSP ( (figure 2B) ). The immunohistochemical staining results are summarized in table 1.

Eight (31%) of 26 DF, two (15%) of 13 DFSP and 21(91%) of 23 MFH showed positive (> 10%) HIF-1α reactivity. This positive rate of HIF-1α expression in MFH was significantly greater than those in DF and DFSP (p < 0.0001), while no significant difference was demonstrated between those in DF and DFSP. The expression levels in MFH were also significantly higher than those in DF and DFSP (> 2+ cases; 1 in DF vs 1 in DFSP vs 9 in MFH, p < 0.0001).

Moreover, within MFH, HIF-1α expression levels correlated with the extents of MVD. Low vs. high HIF-1α expression groups were defined as described in the Materials and Methods section. The high expression group (n = 9) showed significantly higher MVD compared with that in the low expression group (n = 14) (p = 0.0167, table 2( Table 2 )). However, the high and low expression groups showed no significant difference in their clinical outcome determined as overall survival (n = 20) (data not shown).
Table 2 Relationship between MVD and HIF-1α expression in MFH

	HIF-1α expression
Low expression	High expression
N = 14	N = 9
MVD* (Mean ± SE)	58.46 ± 7.33	86.39 ± 6.66

Discussion

This is, to our best knowledge, the first study that explored the expression of HIF-1α in human so-called fibrohistiocytic tumors in association with extents of angiogenesis and tumor malignancy. MVD and HIF-1α protein expression levels correlated with the malignant potentials of the tumors. Furthermore, higher levels of HIF-1α expression were associated with increased MVD in MFH. These lines of evidence suggest a role of HIF-1α in angiogenesis and malignant transformation in so-called fibrohistiocytic tumors.

Immunohistochemical analyses have revealed that HIF-1α is overexpressed in many human cancers [10-17]. Although HIF-1α is a transcriptional activator, mixed nuclear and cytoplasmic staining patterns were observed in these studies. In our study, HIF-1α immunoreactivity was also present in both nuclei and cytoplasm of tumor cells, with the latter being predominant. Cytoplasmic reactivity was often localized in perinuclear Golgi areas. Similar results with predominant cytoplasmic reactivity were also reported in oesophageal squamous cell carcinoma [11]: the mean percentage of cells with nuclear localisation was 4.2% while that of cells with cytoplasmic reactivity was 20.7%. However, the accurate reason for these differences in staining patterns is currently unknown.

HIF-1α expression correlates with VEGF expression and MVD in several tumor types [11-13]. Moreover, significant associations between HIF-1α overexpression and adverse clinical outcome have been shown in many cancers [8, 11, 17]. However, this association is not universal, rather depending on the cancer type. For example, in squamous cell carcinoma of the oropharynx, the degree of HIF-1α expression correlated inversely with the rate of complete remission of the primary tumor, local failure-free survival, disease-free survival and overall survival [14]. In human gliomas, HIF-1α expression levels correlated with induction of angiogenesis and tumor grade [12]. In contrast, HIF-1α protein overexpression alone did not influence the prognosis in ovarian cancer although HIF-1α expression correlated with MVD [13]. This was explained by the fact that MVD itself was not an independent prognostic factor in ovarian cancer. Moreover, expression of HIF-1α in surgically treated patients with head and neck cancer was associated with improved disease-free and overall survival [15]. It was suggested that this might reflect the potentially aggressive nature of HIF-1^–/– tumor cells, which lose their normal ability to undergo apoptosis at a distance from blood vessels reducing their dependence on vascular supply. In our study, HIF-1α expression correlated with extents of MVD and tumor malignancy in so-called fibrohistiocytic tumors, but not with clinical outcome in MFH as in ovarian cancer.

Despite the interests in the study of angiogenic factors in many epithelial tumors, only a few clinical reports have addressed their expression in soft tissue tumors. HIF-1α expression was detected in all grades of chondrosarcoma, but not in normal articular cartilage or benign cartilage tumors [16]. This HIF-1α expression was linked to increased VEGF expression. VEGF expression was diverse amongst the various histologic subtypes of soft tissue sarcomas [18-20], and leiomyosarcomas (LMS), carcinosarcomas and MFH were more likely to overexpress VEGF than the other histologic subtypes [18]. Tumor VEGF immunoreactivity correlated with the tumor grade in many soft tissue sarcomas [20]. However, VEGF overexpression did not correlate with clinical outcome in any subtype of soft tissue sarcomas, except LMS [18, 20]. It was also shown that MVD was not a key factor in the formation of metastasis in MFH [21]. Similarly, our study demonstrated a correlation between HIF-1α expression and increased MVD in MFH but no significant difference in overall survival between the high and low HIF-1α expression groups. However, the number of cases included in our study were too small for this lack of HIF-1α expression levels and overall survival to be meaningful. Further investigations with a large number of cases are needed.

HIF-1α has pluripotent functions in tumorigenesis. On the one hand, HIF-1α supports tumor growth by induction of angiogenesis via transactivation of the VEGF gene and VEGF-independent mechanisms [22]. On the other hand, HIF-1α activates the transcription of genes that are involved in crucial aspects of cancer biology, including cell survival and glucose metabolism [2]. Moreover, HIF-1α expression activates programs of gene expression controlling cancer cell invasion [23]. Expression of HIF-1α is enhanced by intratumoral hypoxia, genetic alternations in tumor suppressor genes and oncogenes, and by several growth factors [2, 23]. Significantly upregulated HIF-1α in MFH may be influenced by these many factors and involved in the various aspects of tumorigenesis.

The involvement of HIF-1α in tumor angiogenesis and progression makes this transcription factor an attractive target for cancer therapy. Many strategies to look for inhibitors of HIF-1α are ongoing [2]. Our study may support the therapy targeting HIF-1α for MFH.

References

1 Folkman J. Seminars in Medicine of the Beth Israel Hospital, Boston. Clinical applications of research on angiogenesis. N Engl J Med 1995; 333: 1757-63.

2 Semenza GL. Targeting HIF-1 for cancer therapy. Nat Rev Cancer 2003; 3: 721-32.

3 Kim KR, Moon HE, Kim KW. Hypoxia-induced angiogenesis in human hepatocellular carcinoma. J Mol Med 2002; 80: 703-14.

4 Semenza GL. HIF-1 and tumor progression: pathophysiology and therapeutics. Trends Mol Med 2002; 8: S62-S67.

5 Fletcher CDM, Van den Berg E, Molenaar WM. Pleomorphic malignant fibrous histiocytoma/Undifferentiated high grade pleomorphic sarcoma. In: Fletcher CDM, Kriahnan Unni K, Mertens F, eds. WHO classification Tumors of Soft tissue and Bone. Lyon: IARCPress, 2002: 120-2.

6 Weiss SW, Goldblum JR. Benign fibrohistiocytic tumors, fibrohistiocytic tumors of intermediate malignancy, and malignant fibrohistiocytic tumors. In: Weiss SW, Goldblum JR, eds. Enzinger and Weiss’s Sott Tissue Tumors. St. Louis: Mosby, Inc., 2001: 441-569.

7 Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastasis--correlation in invasive breast carcinoma. N Engl J Med 1991; 324: 1-8.

8 Kuwai T, Kitadai Y, Tanaka S, Onogawa S, Matsutani N, Kaio E, et al. Expression of hypoxia-inducible factor-1alpha is associated with tumor vascularization in human colorectal carcinoma. Int J Cancer 2003; 105: 176-81.

9 Stipa F, Lucandri G, Limiti MR, Bartolucci P, Cavallini M, Di Carlo V, et al. Angiogenesis as a prognostic indicator in pancreatic ductal adenocarcinoma. Anticancer Res 2002; 22: 445-9.

10 Zhong H, De Marzo AM, Laughner E, Lim M, Hilton DA, Zagzag D, et al. Overexpression of hypoxia-inducible factor 1alpha in common human cancers and their metastases. Cancer Res 1999; 59: 5830-5.

11 Kimura S, Kitadai Y, Tanaka S, Kuwai T, Hihara J, Yoshida K, et al. Expression of hypoxia-inducible factor (HIF)-1alpha is associated with vascular endothelial growth factor expression and tumour angiogenesis in human oesophageal squamous cell carcinoma. Eur J Cancer 2004; 40: 1904-12.

12 Zagzag D, Zhong H, Scalzitti JM, Laughner E, Simons JW, Semenza GL. Expression of hypoxia-inducible factor 1alpha in brain tumors: association with angiogenesis, invasion, and progression. Cancer 2000; 88: 2606-18.

13 Birner P, Schindl M, Obermair A, Breitenecker G, Oberhuber G. Expression of hypoxia-inducible factor 1alpha in epithelial ovarian tumors: its impact on prognosis and on response to chemotherapy. Clin Cancer Res 2001; 7: 1661-8.

14 Aebersold DM, Burri P, Beer KT, Laissue J, Djonov V, Greiner RH, et al. Expression of hypoxia-inducible factor-1alpha: a novel predictive and prognostic parameter in the radiotherapy of oropharyngeal cancer. Cancer Res 2001; 61: 2911-6.

15 Beasley NJ, Leek R, Alam M, Turley H, Cox GJ, Gatter K, et al. Hypoxia-inducible factors HIF-1alpha and HIF-2alpha in head and neck cancer: relationship to tumor biology and treatment outcome in surgically resected patients. Cancer Res 2002; 62: 2493-7.

16 Lin C, McGough R, Aswad B, Block JA, Terek R. Hypoxia induces HIF-1alpha and VEGF expression in chondrosarcoma cells and chondrocytes. J Orthop Res 2004; 22: 1175-81.

17 Kurokawa T, Miyamoto M, Kato K, Cho Y, Kawarada Y, Hida Y, et al. Overexpression of hypoxia-inducible-factor 1alpha(HIF-1alpha) in oesophageal squamous cell carcinoma correlates with lymph node metastasis and pathologic stage. Br J Cancer 2003; 89: 1042-7.

18 Potti A, Ganti AK, Tendulkar K, Sholes K, Chitajallu S, Koch M, et al. Determination of vascular endothelial growth factor (VEGF) overexpression in soft tissue sarcomas and the role of overexpression in leiomyosarcoma. J Cancer Res Clin Oncol 2004; 130: 52-6.

19 Potti A, Ganti AK, Foster H, Knox S, Hebert BJ, Tendulkar K, et al. Immunohistochemical detection of HER-2/neu, c-kit (CD117) and vascular endothelial growth factor (VEGF) overexpression in soft tissue sarcomas. Anticancer Res 2004; 24: 333-7.

20 Chao C, Al-Saleem T, Brooks JJ, Rogatko A, Kraybill WG, Eisenberg B. Vascular endothelial growth factor and soft tissue sarcomas: tumor expression correlates with grade. Ann Surg Oncol 2001; 8: 260-7.

21 Ohsawa M, Tomita Y, Kuratsu S, Kanno H, Aozasa K. Angiogenesis in malignant fibrous histiocytoma. Oncology 1995; 52: 51-4.

22 Ryan HE, Poloni M, McNulty W, Elson D, Gassmann M, Arbeit JM, et al. Hypoxia-inducible factor-1alpha is a positive factor in solid tumor growth. Cancer Res 2000; 60: 4010-5.

23 Krishnamachary B, Berg-Dixon S, Kelly B, Agani F, Feldser D, Ferreira G, et al. Regulation of colon carcinoma cell invasion by hypoxia-inducible factor 1. Cancer Res 2003; 63: 1138-43.