Hepatocyte growth factor and c-MET in benign and malignant peripheral nerve sheath tumors

Hepatocyte growth factor and c-MET in benign and malignant peripheral nerve sheath tumors

Hepatocyte Growth Factor and c-MET in Benign and Malignant Peripheral Nerve Sheath Tumors UMA N.M. RAO, MD, ENI SONMEZ-ALPAN, MS, AND GEORGE K. MICHAL...

20MB Sizes 0 Downloads 29 Views

Hepatocyte Growth Factor and c-MET in Benign and Malignant Peripheral Nerve Sheath Tumors UMA N.M. RAO, MD, ENI SONMEZ-ALPAN, MS, AND GEORGE K. MICHALOPOULOS, MD, PHD Hepatocyte growth factor (HGF), secreted by mesenchymal cells, has pleiotropic biological activities on several cell types. HGF and

its receptor, the c-met proto-oncogene product (c-MET) have been implicated in the genesis and progression of several carcinomas and sarcomas. It has been suggested that MET/HGF autocrine signaling m a y contribute to tumorigenesis in sarcomas. HGF has been recently f o u n d to be a mitogen for rat Schwann cells and to be present in n e u r o f i b r o m a s in NF1 patients. In this investigation, we assessed the immunoreactive patterns of HGF and MET in benign and malignant peripheral nerve sheath tumors (PNST) using archival formalin-fixed tissue. The standard avidin-biotln-peroxidase method was used. All benign tumors were negative with HGF. Eight cases of MPNST were

H G F (hepatocyte growth factor), originally called h e p a t o p o i e t i n , 1 is p r o d u c e d by m e s e n c h y m a l cells a n d 9 is identical to the scatter factor. Initiation o f H G F action is d e p e n d e n t o n its b i n d i n g to the t r a n s m e m b r a n e tyrosine kinase r e c e p t o r e n c o d e d by the p r o t o - o n c o gene c-met, s'4 which is expressed p r e d o m i n a n t l y in epithelial cells. 5 T h e genes for b o t h H G F a n d c-met are located o n l o n g a r m o f c h r o m o s o m e 7. 6 H G F a n d its r e c e p t o r are t h o u g h t to constitute a m o d e l for paracrine signaling system w h e r e the ligand is expressed by m e s e n c h y m a l cells a n d binds to the r e c e p t o r expressed in epithelial cells. H G F has proliferative, m o t o g e n i c , m o r p h o g e n i c , a n d a n g i o g e n i c activities a n d p r o m o t e s cellular invasiveness. H G F / m e t pathway m a y transduce e p i t h e l i a l / m e s e n c h y m a l signaling a n d conversion during varied physiological a n d pathological processes such as w o u n d healing, embryogenesis, regeneration, a n d t u m o r progression. 7-1 1 '24 Co-transfection o f N I H 3 T 3 cells with c-MET a n d H G F / S F resulted in t u m o r i g e n e sis. 12 In the c-met t r a n s f o r m e d N I H / 3 T 3 cells, tumorigenicity was a c c o m p a n i e d by high levels o f b o t h ligand a n d r e c e p t o r in t u m o r explants. 13 T h e r e is considerable evidence indicating involvement o f H G F a n d c-MET in d e v e l o p m e n t a n d progression o f certain solid tumors, including some sarcomas, by either autocrine or paraFrom the Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA. Accepted for publication January 13, 1997. Supported by the Pathology Education and Research Foundation. Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA. Presented in part at the annual meeting of the United States and Canadian Academy of Pathology in Washington, DC, March, 1996. Address correspondence and reprint requests to Uma N.M. Rao, MD, Montifiore University Hospital, Rm NW 628, University of Pittsburgh Medical Center, 200 Lothrop St, Pittsburgh, PA 15213-2582. Copyright © 1997 by W.B. Saunders Company 0046-8177/97/28094)01155.00/0

positive with both HGF and MET. In some malignant PNST, positivity with both llgand and the receptor may be indicative o f an autocrine mediated signal transduction and m a y implicate HGF/MET in t u m o r progression. Immunoreactivity with MET was strikingly greater in MPNST in contrast to benign PNST; this finding may prove to be helpful in distinguishing some histologically low-grade MPNST f r o m cellular and atypical benign PNST. HUM PATHOL 28:1066-1070. Copyright © 1997 by W.B. Saunders C o m p a n y Key words: hepatocyte growth factor, MET, MIB1, malignant peripheral nerve sheath tumors. Abbreviations: HGF, hepatocyte growth factor; MPNST, malignant peripheral nerve sheath tumor; NF1, neurofibromatosis type 1.

crine mechanisms. TM H G F has b e e n f o u n d to be a p o t e n t m i t o g e n for rat schwann cells a n d has b e e n shown in schwann cell tumors, 15 which p r o m p t e d us to e x a m i n e the p a t t e r n o f distribution o f MET a n d H G F in b e n i g n a n d m a l i g n a n t nerve sheath tumors.

MATERIALS AND METHODS Ten benign and 14 malignant nerve sheath tumors in 22 patients were culled from the combined files of Montefiore and Presbyterian hospitals (University of Pittsburgh Medical Center) between 1990 and 1994. Histological diagnosis was established using previously described criteria. 16'17The diagnosis and grade of malignant peripheral nerve sheath tumor (MPNST) was based on presence of increased cellularity, mitotic activity, and presence of necrosis. All but one malignant tumor were histologically high grade. The malignant tumors ranged in size from 9 to 20 cm in maximum diameter and occurred in patients between the ages of 25 and 80 years and included one case in which the primary tumor, as well as liver metastases, were available for study. Three patients had documented type 1 neurofibromatosis (NF1). In two cases, although the tumors involved the brachial plexus and the sciatic nerve, respectively, there was no clinical evidence of NF1. The benign tumors were sporadic and included three cellular schwannomas, five classic schwannomas, one plexiform schwannoma, and one neurofibroma. They ranged from 2.9 to 5.5 cm in maximum diameter. Formalin-fixed paraffinembedded serial sections were studied using polyclonal antibodies to c-met (Santa Cruz Biotech, Santa Cruz, CA) and HGF (R&D Systems, Minneapolis, MN). A section of breast with fibroadenoma was used as a positive control for both cmet and HGF. In addition, a section of normal placenta was used as a positive control for HGF. Negative control slides were prepared by substituting a buffer for the primary antibody. All tumor sections were also examined for expression of cytokeratins (AE1, 3 Boehringer Mannheim, Indianapolis, IN: dil 1:900) and for proliferative activity using MIB1 (AMAC, Westbrook, ME; dil 1:100) using standard avidin-biotin complex techniques, is Four-micron-thick serial sections were

1066

HGF/MET IN NERVE SHEATH TUMORS (Rao et al)

TABLE 1.

HGF Positive Cells* (%) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.$

tumors, including one of three cellular schwannomas, showed rare focal (2+) positivity with MET, but the latter did not show abnormalities of chromosome 7 by traditional cytogenetic examination. Nonspecific weak stromal positivity was rarely noted. All benign tumors were negative with HGF. Endothelial cells and smooth and skeletal muscle were positive with MET and HGF. Schwann cells in normal nerve fibers were positive for MET only. The total n u m b e r of MIB-l-positive nuclei in three highpower fields ranged from 79 to 300 in malignant tumors, in contrast to benign tumors, where the range was 2 to 26. One cellular schwannoma had 54 cells with MIB1positive nuclei, and this case also showed focal (2+) positivity with MET. Clinical follow-up is available in nine cases of MPNST. Four of five patients who died h a d tumors positive with b o t h H G F and MET, and of these, two patients had NF1. One NF1 patient with this i m m u n o r e active pattern is alive 18 m o n t h s after surgical resection of the tumor, and one 80-year-old patient h a d no evidence of disease at last follow-up 2 years after a locally recurrent t u m o r was excised. O f the six patients whose tumors were not immunoreactive for HGF, follow-up is available on three. O n e patient died with lung metastasis 24 m o n t h s after diagnosis. Two are alive and free of disease at last follow-up at 16 and 18 months. T h e patients with cellular s c h w a n n o m a are free of recurrence at 3 and 4 years, respectively.

Immunoreactive Patterns of MPNST

20 20 10 5-10 30 50 50 Diffuse and weak Neg Neg Neg Neg Neg Neg

MET Positive Cells* (%)

MIB1 Positive Cells'~

50 100 100 100 80 75 100 50 100 70 100 50 5 Few positive

100 300 120 100 200 300 210 100 200 54 300 150 79 250

* Represents percentage of positive tumor cells in relation to total area of tumor on the section. Number of cells wkh MIBl-positive nuclei counted in three consecutive high-power fields (40X). ++Represents liver metastasis from case 5.

used. In a given case, staining reactions were recorded in approximately similar locations on the tumor sections with both antibodies. Percent of positive areas of tumor in relation to the total tumor area in the section was estimated, aided by an ocular micrometer. The intensity of staining reaction per cell was graded on a four-tiered score as follows: 1+ = focal weakly positive to negative, 2+ = weak positivity, 3+ = moderate, and 4+ = moderate, and 4+ = strong positive reactions. Tumors with diffuse 2+, focal, or diffuse 3+ and 4+ were considered positive. Tumor cells with MIBl-positive nuclei were counted in three consecutive high-power fields (40×), in sites of maximum immunoreactivity. Five cases (two cellular schwannomas and three MPNST) included in this study also underwent cytogenetic analysis after short-term culture (reported elsewhereIg).

RESULTS All 14 malignant nerve sheath tumors displayed an intense cytoplasmic and occasionally nuclear positivity with MET. Eight cases were positive with both MET and HGF (Table 1). Six MPNSTs, including the liver metastasis, were negative with HGF. There was intratumoral heterogeneity noted with MET. Compared with c-met, the staining reaction with HGF was always less intense. The positive staining reactions with MET and HGF occurred in approximately similar geographic locations and probably not in the same cells (Fig 1). In two instances, HGF and MET decorated epithelioid-appearing neoplastic cells (Fig 2). None of the tumors were positive with cytokeratins. One case of liver metastases was negative with HGF, and only a rare cell was positive with MET. Percent of positive cells as well as intensity of staining appeared to be greater in histologically high-grade tumors. There was only one lowgrade tumor for comparison, however. None of the tumors were positive with HGF alone. O f the two cases that displayed mostly numerical abnormalities of chromosome 7, only one tumor showed positivity with MET. Five benign

DISCUSSION Soft tissue sarcomas are rare tumors, and MPNST constitute approximately 10% of all sarcomas, and 4.6% of these tumors occur in patients with type 1 neurofibromatosis (NF1). Malignant transformation in a benign sporadic solitary s c h w a n n o m a is rare. Most MPNSTs, especially those that occur in patients with NF1, have b e e n n o t e d to have a high metastatic potential and p o o r 5-year survival. 16'~7 Data f r o m a recent study showed that HGF might act as a paracrine Schwann cell growth factor in rats. It was also shown to be present in neurofibromas f r o m NF1 patients, where it constituted 60% to 97% of the Schwann cell mitogenic activity extracted f r o m these tumors. This indicated that HGF may have a role in formation of schwann cell neoplasms, although other mitogens such as insulin-like growth factor-1 and basic fibroblast growth factor also may have a role. 15 H G F and c-met transcripts have not b e e n f o u n d to occur in the same cell in embryonic tissue2°; however, some h u m a n sarc o m a cell lines have b e e n f o u n d to have b o t h c-MET and H G F expression, with high levels of c-MET expression. In the same study, the immunoreactive pattern of some sarcomas was f o u n d to be h e t e r o g e n o u s on paraffin sections. This suggested b o t h autocrine and paracrine modes of stimulation. 21 In other experimental studies, co-expression of c - M E T / H G F in N I H 3T3 cells increased cell motility, collagenase activity, and invasiveness in vitro, and metastatic activity in vivo. 22'2~ In o u r series, HGF and MET decorated epithelioid-appearing

1067

0

ml

HGF/MET iN NERVE SHEATH TUMORS (Rao et al)

FIGURE 1. Immunohistochemical staining patterns of Met (A) and HGF (B) in a case of MPNST. The intensity of staining reaction with MET is graded as 4+ and that of HGF as 3+. in another case of MPNST, MET (C) is strongly positive (4+), whereas HGF (D) in a similar area is weakly positive (2+). Areas of connective tissue ~resent are negative; in addition to the tumor cells, some stromal ceils are positive for HGF. (Original magnification × 115.)

cells in two tumors. The epithelioid appearance is predicated on overall cell morphology, but it should be pointed out that the cells did not show keratin positivity. Thus, it is unlikely that this pattern is related to induction of epithelial mesenchymal transitions. 24 Six other MPNSTs were positive, with both HGF and MET in approximately similar geographic areas on sections of a given tumor. This suggests co-localization of HGF and MET and therefore possible autocrine m o d e of stimulation in some areas. The results of our study in the context of transformation experiments coexpressing HGF and c-MET suggest that co-localization of HGF and MET might identify a biologically aggressive subset of tumor. The clinical relevance and as well as the genetic events underlying this p h e n o m e n o n need to be defined in large numbers of cases.

FIGURE 2. ×230.)

In our series of MPNSTs, the differences in staining patterns with c-MET in benign and malignant tumors are quantitative as well as qualitative. The outcome of ligand stimulation is probably d e p e n d e n t on receptor level and activation in the tumor tissue, which in turn may be related to tumor differentiation. The staining intensity and percentage of positive cells was strikingly greater in malignant PNSTs, c o m p a r e d with benign PNSTs and correlated with the degree of proliferative activity seen with MIB1. This finding may prove to be useful in histological diagnosis of atypical neurofibromas and cellular schwannomas, which can be difficult to distinguish from MPNSTs.

Acknowledgment. The authors thank Wendy Grace for her help with typing and Linda Shaab for photography.

Immunostaining with Met (A) and HGF (B) in epithelioid-appearing cells in an area of MPNST. (Original magnification

1069

HUMAN PATHOLOGY

Volume 28, No. 9 (September 1997)

REFERENCES 1. Michalopoulos GK, Houck KA, Dolan ML, et al: Control of hepatocyte replication by two serum factors. Cancer Res 44:44144419, 1984 2. Naldini L, Weidner M, Vigna E, et al: Scatter factor and HGF are indistinguishable ligands for the MET receptor. Eur Mol Biol O r g a n J 10:286%2878, 1991 3. Cooper SC, Park M, Blair DG, et al: Molecular cloning of a new transforming gene from a chemically transformed human cell line. Nature 311:29-33, 1984 4. Naldini L, Vigna E, Narsimhan RP, et al: Hepatocyte growth factor (HGF) stimulates the tyrosine kinase activity of the receptor encoded by the proto-oncogene c-met. Oncogene 6:501-504, 1991 5. Comoglio PM: Structure, biosynthesis and biochemical properties of the HGF receptor in normal and malignant cells, in Goldberg ID, Rosen EM (eds): Hepatocyte Growth Factor-Scatter Factor (HGFSF) and the C-Met receptor. EXS 65. Basel, Switzerland, Birkhauser Verlag, 1993, pp 131-165 6. Saccone S, Narsimhan RP, Gaudino G, et al: Regional mapping of the human hepatocyte growth factor (HGF)-scatter factor gene to chromosome 7q21.1. Genomics 13:912-914, 1992 7. Weidner KM, Hartmann G, Naldini L, et al: Molecular characteristics of HGF/SF and its role in cell motility and invasion. EXS 65:311-328, 1993 8. Michalopoulos G: Hepatocyte growth factor. Hepatology 15:149-155, 1992 9. Strain AJ: Hepatocyte growth factor another ubiquitous cytokine. J Endocrinol 137:1-5, 1993 10. Rosen EM, Nigam SK, Goldberg ID: Scatter factor and the c-met receptor: A paradigm for mesenchymal/epithelial interaction. J Cell Biol 127:1783-1787, 1994 11. Wang Y, Seldon C, Farnaud S, et al: HGF/SF is expressed in human epithelial cells during embryonic devolopment: Studies by in situ hybridization and Northern blot analysis. J A_nat 185:543-551, 1994 12. Rong S, Bodescot M, Blair D, et al: Tumorigenesis of the

met protooncogene and the gene for hepatocyte growth factor. Mol Cell Biol 12:5152-5158, 1992 13. Rong S, Oskarsson M, Faletto D, et al: Tumorigenesis induced by coexpression of human hepatocyte growth factor and the human met protooncogene leads to high levels of expression of the ligand and receptor. Cell Growth Differ 4:563-569, 1993 14. CortnerJ, Vande Woude GF, Rong S: The Met-HG/SF autocrine signalling mechanism is involved in sacomagenesis, in Goldberg ID, Rosen EN (eds): Epithelial Mesenchymal Interactions in Cancer. Basel, Switzerland, Birkhauser Verlag, 1995, pp 89-121 15. Krasnoselsky A, Massay MJ, DeFrances MC, et al: Hepatocyte growth factor is a mitogen for schwann cells and is present in neurofibromas. J Neurosci 14:7284-7290, 1994 16. Enzinger FM, Weiss SW: Malignant tumors of peripheral nerves, in Soft Tumors, (ed 3). St Louis, Mosby, 1995, pp 889-928 17. Ducatman BS, Scheithauer BW; Piepgras DG: Malignant peripheral nerve sheath tumors: A clinicopathologic study of 120 cases. Cancer 57:2006-2021, 1986 18. Hsu SM, Raine L, Fager H: Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: A comparison between ABC and unlabeled anti-body (PAP) Procedures. J Histochem Cytochem 29:577-580, 1981 19. Rao UNM, Surti U, Hoffner L et al: Cytogenetic and histologic correlation of peripheral nerve sheath tumors of soft tissue. Cancer Genet Cytogenet 88:17-25, 1996 20. Sonnenberg E, Wiedner KM, Birchmeier C: Expression of the met-receptor and its ligand, HGF-SF during mouse embryogenesis. Experientia 65:381-394, 1993 (suppl) 21. Rong S, Jeffers M, ResauJH, et ah Met expression and sarcoma tumorigenicity. Cancer Res 53:5355-5360, 1993 22. Rong S, Segal S, Anver M, et al: Invasiveness and metastasis of NIH 3T3 cells induced by Met-hepatocyte growth factor/scatter factor autocrine stimulation. Proc Natl Acad Sci U S A 91:4731-4735, 1994 23. Bellusci S, Moens G, Gaudino G, et al: Creation of a hepatocyte growth factor/scatter factor autocrine loop in carcinoma cells induces invasive properties associated with increased tumorigenicity. Oncogene 9:1091-1099, 1994 24. Tsarfaty I, Rong S, ResauJH, et al: The Met proto-oncogene mesenchymal to epithelial cell conversion. Science 263:98-101, 1994

1070