Vla-3 Distribution in Normal and Neoplastic Non-lymphoid Human Tissues

Path. Res. Pract. 189,387-393 (1993)

Original Papers

Vla-3 Distribution in Normal and Neoplastic Non-lymphoid Human Tissues A. Bartolazzi, c. Cerboni, C. Full, C. Valentini and P. Giorgio Natali Department of Immunology, Immunology, Regina Elena Cancer Institute, Rome, Italy

I. Venturo Department of Immunology, Medical Oncology, Regina Elena Cancer Institute, Rome, Italy

A. Bigotti Department of Immunology, Pathology, Regina Elena Cancer Institute, Rome, Italy

SUMMARY Using monoclonal antibody (mAb) M-Kid 2 to the a3~1 heterodimer, we have evaluated immunohistochemically the in vivo expression of the Vla-3 integrin in normal and transformed non-lymphoid human tissues. In normal tissues the a3~1 complex displays a polarized distribution at the baso-lateral aspect of most keratinizing and glandular epithelia. In addition the integrin is detected in perineurium, basal lamina of smooth muscular fibers, vascular media, podocytes and Bowman's capsule, myoepithelial cells of the parotid and breast, and in pulmonary alveoli. Neoplastic transformation is associated with qualitative and quantitative changes in expression of this integrin. The loss of polarized distribution often occurs in various malignancies. Furthermore, a significant decrease in expression occurs in 13% of the colon-rectum carcinomas, 75% of the ductal invasive, and 40% of the lobular invasive breast carcinomas. Among the lung malignancies tested, the small cel/lung carcinomas (SCLC) were found to be consistently unreactive with mAb M-Kid 2. Analysis of Vla-3 expression in established tumor cell lines demonstrated that the integrin is almost invariably expressed by the plastic adherent cell subpopulations.

Introduction Cell-to-cell and cell-to-extracellular matrix interactions are of major relevance in morphogenesis, maintenance and repair of tissue architecture 13 , 19,32, as well as in pathological processes such as malignant transformation and metastatic dissemination 9, 11, 14, 15, 17,30, 33. © 1993 by Gustav Fischer Verlag, Stuttgart

These interactions are mediated by cell surface adhesion receptors of which the integrins are one of the largest groups, which is increasingly characterized at molecular and functional levels l , 13, 16,32. The use of monoclonal antibodies to integrin a-chain subunits has clearly pointed out that the antigenic repertory of cultured cells is far from being a mirror of their phenotype in vivo, thus suggesting 0344-0338/93/0189-0387$3.5010

388 . A. Bartolazzi et al.

that a critical evaluation of the physiopathological role of the integrin system can be more completely appreciated by the analysis of its in vivo distributionS, 7, 12, 18,22,28,29,34,36. We report in the present study the distribution of a3~1 in normal human tissues and their neoplastic counterparts, using the newly developed murine mAb M-Kid 2, which recognizes the a3~1 complex. Material and Methods

Monoclonal and Polyclonal Antisera As previously reported 3 the murine mAb M-KID 2 of the IgG lk isotype, recognizing the a3~1 heterodimer, was obtained using as immunogen the plastic adherent cell fraction of the cell line KJ 29. Fluorescein labelled F (abh fraction of a rabbit anti-mouse IgG (Sorin, Saluggia VC, Italy) and Vectastain avidin-biotin staining kit for murine IgG (Vector Laboratories Inc, Burlingame CA, USA) were used for indirect immunohistochemical procedures.

Cell Lines and Tissues The long-term renal carcinoma KJ 29 cell line, established from a primary clear cell carcinoma, was a gift from Dr. R. Gambari (Department of Biochemistry, Univ. of Ferrara, Italy)6. Cell lines WIL-2, Molt-4, K562, LS 174-T and A431 were obtained from American Type Culture Collection (Bethesda, MD, USA). The melanoma cell line 1007-mel, deriving from a primary tumor, was obtained from Dr. G. Fossati (Istituto Tumori di Milano, MI, Italy); the Colo-38 melanoma cell line has been previously described 10 . The microcitoma cell line (AE 2) was a gift from Dr. G. Zupi (Lab. of Experimental Chemoterapy, Regina Elena Cancer Institute)21. Cell lines SAN, CIAV, CAP, LOG, and TED, were established in our laboratory from neoplastic effusions. The cell line A 2 182 was provided by Dr. A. S. Aaronson (Laboratory of Cellular and Molecular Biology, National Institute of Health, NCI, Bethesda, Ma, USA). Cell lines were routinely maintained in RPMI 1640 medium (Gibco Lab. Paisley U.K.) supplemented with 10% heat inactivated fetal bovine serum and 1 % glutamine (200 mM). Cells were grown at 37°C in a humidified atmosphere containing 5% CO 2. Cytospins were obtained using a Shandon cytocentrifuge (Shandon, Runcorn, Cheshire, UK) and fixed for 10 min in absolute acetone as previously described 25 . Normal and neoplastic tissue specimens were obtained from patients free from therapy, undergoing surgical treatment at the Regina Elena Cancer Institute. Upon removal, tissues were divided into two parts; one was processed for routine histological examination, the other was snap frozen in liquid nitrogen and stored at - 20°C prior to use in immunohistochemical assays. Four micron cryostatic sections were obtained and fixed in absolute acetone for 10 min. The sections were either immediately used as a substrate for indirect immunoperoxidase or immunofluorescence, or stored at - 20°C up to six months with no appreciable changes in reactivity. The sections were also stained with 1 % toluidine blue for morphological evaluation.

Immunohistochemical and Cytochemical Assay Indirect immunofluorescence (IIF) was performed according to standard conditions employing a fluorescein conjugated antiserum at protein concentrations ranging from 200 f.tglml onwards with an FIP ratio of 3, as previously described26 .

Negative controls consisted of tissue sections incubated with an uncorrelated mAb of the same isotype as mAb M-KID 2. The cell lines used in flow cytometry assay were harvested from culture flasks, extensively washed with HBSS (Gibco Lab. Paisley U.K.) and resuspended to a concentration of 1 X 106 /ml. Thereafter cells were incubated with purified mAb M-Kid 2 at concentration ranging from 50 to 100 f.tglml or with the isotype-matched control mAb for 1 h at 4°C. Cells were then washed twice with PBS and incubated with F(abh rabbit-anti mouse Ig-FITC for 45 min at 4°C. This incubation was followed by three additional washes in PBS. Cells were finally analyzed using flow cytometer (Becton and Dickinson FACS) with an argon laser tuned to 488 nm.

Cell Labelling and Immunoprecipitation For metabolic labelling cell lines were washed with methioninefree RPMI1640 medium and starved in the same medium containing 10% dialyzed heat inactivated FCS, for 2 hours. Cells were then pulsed for 12 hours with 250 f.tCi/ml of [35S]-methionine (Amersham International pic, Amersham UK). After labelling, cells were washed in PBS and lysed at 4°C for 1 hour in lysis buffer containing 1 % NP 40 (BDH Italia Sri Milan, Italy), supplemented with aprotinin 100 Ulml (Lepetit, Milan, Italy), leupeptin 10 f.tglml (Sigma Chemical Co, St. Louis, Mo), and PMSF 10 f.tM (Bethesda Research Laboratories, Inc.) as proteinase inhibitors. Celllysates were centrifuged at 6000 g and supernatant precleared by incubation for 2 hours with proteinA-sepharose CL4B (Pharmacia, Uppsala, Sweden), coated with affinity purified rabbit-anti mouse IgG (Sigma) Chemical Co, St. Louis, Mo). After preclearing, celllysates were incubated for 1 hour at 4°C with protein-A-sepharose CL4B, previously conjugated with M-KID 2 mAb. Immunoprecipitates were analyzed by SDS polyacrylamide gel electrophoresis on 7.5% polyacrylamide resolving gels and 3.5% acrylamide stacking gels 20 . The gels were fixed, dried and fluorographed as described by Bonner and Laskey4.

Table 1. Expression of VLA-3 integrin in continuous cell lines and freshly isolated tumor cells Established Origin cell lines

Growth pattern VIa-3 1 expression in vitro

K562 WIL-2 Molt 4 KJ29 KJ29 AE2 A431 CIAV CAP LS 174T A2182 Colo-38 1007-mel

suspension suspension suspension suspension adherent unselected adherent adherent adherent adherent adherent unselected adherent

(Erytroleukemia) (B-lymphocytes) (LLA) (Renal carcinoma) (Renal carcinoma)

(Microcitoma)

(Vulva carcinoma) (Ovarian carcinoma) (Breast carcinoma) (Colon carcinoma) (Lung carcinoma) (Melanoma) (Melanoma)

+ + + + + +

Cells freshly collected from effusions

SAN TED LOG

(Lung carcinoma) (Breast carcinoma) (Ovarian carcinoma)

suspension adherent adherent

+ +

1 Detected by indirect immunofluorescence on cell suspension.

Vla-3 in Human Tissues . 389

Results In Vitro Modulation in Expression of Vla-3 in Cell Cultures

As the renal carcinoma cell line KJ 29, which was employed to generate mAb M-KID 2, is known to express different levels of Vla-3 according to its growth pattern (adhesion versus suspension), we have analyzed whether this modulation occurs in other transformed cells from tumours of various histological types propagated in cul-

o

o

800

400

A

400

800

96%

B

2.29%

ture, by employing indirect immunofluorescence. The results of this study are summarized in Table 1. With the exception of the primary melanoma cell line 1007-Mel, plastic adherent cells were almost consistently reactive with mAb M-KID 2, while cell lines growing in suspension invariably expressed no detectable levels of Vla-3 on the cell surface. When tumor cells freshly collected from metastatic effusions were tested, only those which could be propagated as adherent cell lines (TED and LOG) expressed the Vla-3 molecule. These findings were confirmed by flow cytometry on selected cell lines (Fig. 1) and immunochemically, as shown in Fig. 2. mAb M-KID 2 immunoprecipitates a 110-150 Kd complex in nonreducing conditions only from lysates of the plastic adherent cell lines KJ 29 (lane 7) and A 2182 (lane 4), but not from lysates of the lung adenocarcinoma SAN, of the microcitoma AE2, and of the melanoma cell line Colo-38 (lanes 6,5,2 respectively), growing in suspension, nor from the cell lysate of primary melanoma 1007-Mel (lane 3).

~1&OU

~ 110

ltd

en LL

o

+'E~~~~~~~

F

0.84%1

I

77.6%1I

I I I I

I

I

I I

I

2.1%

H

I

6.6%

I I I

I

I

I I I

I

\ 10°

10'

10

2

103

4

10

10°

FI 1

10'

10

2

10 3 10 4

Fig. 1. Expression of Vla-3 integrin on selected cell lines with various growth patterns in vitro, as monitored by FACS analysis. While the plastic adherent cell fraction of cell lines KJ 29 (B) and A 2182 (F) are strongly stained by mAb M-KID 2, the subpopulation of KJ 29 (0) and the cell line K 562 (H) growing in suspension, showed only a faint reactivity (9% and 4.5% respectively) with mAb M-KID 2. Panels A, C, E, G show the reactivity of adherent and non-adherent KJ 29 subpopulations, A 2182 and K 562 cell lines respectively, with FITC rabbit antimouse Ig, as negative control.

1

2

3

4

e

7

Fig. 2. Molecular species identified by mAb M-KID 2 in various cell lines biosynthetic labelled with 35S-Methionine. a3~1 heterodimer (110-150 kd) was immunoprecipitated in non-reducing conditions only from lysates of plastic adherent cell lines KJ 29 (lane 7) and A 2182 (lane 4), but not from celllysates of lung adenocarcinoma SAN, microcitoma AE2 and melanoma cell line Colo-38 (lanes 6, 5, 2, respectively) growing in suspension, as well as from cell lysates of primary melanoma adherent cell line 1007-mel (lane 3). Rabbit anti mouse IgG as negative control (lane 1) (7.5% SOS-PAGE).

390 . A. Bartolazzi et al.

Distribution of a3~1 Complex in Normal Human Tissues The staining of normal adult tissues with mAb M-KID 2 revealed that the a3~1 heterodimer is widely expressed in cells of different embryonic origin (Table 2). No detectable staining was seen in hepatocytes, skeletal muscle, brain cortex, meninges, endoneurium, myocardium, endocrine and exocrine pancreas and acinar epithelium of the parotid. In the epithelial tissues tested, three major staining patterns were observed. Almost as a general rule, mAb M-KID 2 stained the "basolateral" aspect of the cell layer abutting the basement membrane. Varying from this pattern, in the pharynx a weak stain was also detected in the cells of the suprabasal layer. A cell membrane staining of all the cell layers was detected in transitional epithelium of the urinary bladder. Finally staining limited to the "basal" cell aspect adjacent to the basement membrane (BM-like), was observed in the distal renal tubules, thyrocytes and seminiferous epithelium (Fig. 3A and B). Among the non-epithelial tissues tested, mAb M-KID 2 stained the perineurium, the basal lamina of the smooth muscular fibers, the vascular media, the myoepithelial cells of the parotid and of the mammary glands. Additional epithelial cells expressing the Vla-3 molecule were podocytes, those of the Bowman's capsule, and pulmonary alveoli. In the latter structures a linear staining of alveolar walls was detected. Expression of a3~1 Complex on Human Primary Tumors On the basis of the information obtained in the previous analysis, the expression of the Vla-3 molecule was comparatively evaluated on 221 specimens of primary tumors of various histological type and degree of differentiation. As a

Table 2. Patterns of VLA-3 expression tissues Pattern

In

normal human

Tissue

Baso-lateral Pharynx (basal and suprabasallayer) Esophagus (basal layer) Stomach (surface and glandular epithelium) Jejunum Colon Epidermis and adnexa (basal layer, sebaceous and sweat glands) Bronchial epithelium (including bronchial glands) Fallopian tubes Celomic epithelium Uterine cervix (basal layer) Endometrial glands and surface epithelium Breast (ducts and acini) Gall bladder and biliary ducts Pancreatic ducts Parotid ducts Prostate Syncytiotrophoblast and cytotrophoblast Choroidal plexus Whole PM

Transitional epithelium (all celllaye'rs)

Basal

Distal renal tubules Thyrocytes Seminiferous epithelium

No distinct pattern could be established in perineurium, basal lamina of smooth muscular fibers, vascular media, podocytes and Bowman's capsule, myoepithelial cells of the parotid and breast glands, pulmonary alveoli.

general feature, neoplastic transformation is associated with the expression ofVla-3 in most tumors. No detectable staining was in fact observed in about 25% of the specimens studied.

Fig. 3. Immunohistochemical reactivity of mAb MKID 2 on normal and neoplastic human tissues (Indirect immunofluorescence). - A: Gall bladder epithelium shows a baso-lateral stain.B: In distal renal tubules a stain BM-like is observed. C: Epidermoid skin carcinoma with variable stain of neoplastic cell nests and basal reinforcement. - D: Urinary bladder carcinoma homogenously stained with mAb M-KID 2 (A, B, C, D X 320; scale bar 1 cm = 31 ~m).

Vla-3 in Human Tissues . 391

Nevertheless, various changes in the distribution of the heterodimeric complex were observed. They consisted of three major staining patterns (Table 3), which can be summarized as follows: in most instances, the staining of the whole cell plasma membrane was associated with a more intense staining at the periphery of the tumor cell nests. In about 31 % of the cases a staining pattern outlining almost homogeneously the single cell boundaries was found. This was the case in tumors from colon, esophagus, stomach, breast, endometrium, uterine cervix, ovary, lung (non-small cell carcinomas), pancreas, urinary bladder, kidney and melanocytes. Lack of Vla-3 in the majority of the tumor cell population with a staining restricted to the periphery of tumor cell nests was detected in about 22% of the colon carcinomas, in 60% of the carcinomas arising from the uterine cervix and in parotid, in 5% of the ovarian carcinomas, in 25% of the thyroid carcinomas, in 10% of the NSCLC and renal adenocarcinomas. Among the eight liver malignancies tested, mAb M-Kid 2 stained homogeneously one cholangiocarcinoma and one trabecular hepatocarcinoma. The first lesion was strongly reactive with mAb M-Kid 2, showing a plasma membrane pattern of staining polarized at the periphery of the tumor cell nests. Table 3. Expression of VLA-3 integrin in solid primary human tumors Staining pattern PCN2 Cell

Tumors from

Fpl

Colon Esophagus Stomach Liver 3 CC HC Breast4

20123 6 4/4 2 12/12 10 1/1 0 1/8 0 8/32 DIC 4 6/10 LIC 1 3/3 MED 3 0/1 Muc 0 1/1 Und 0 10/10 6 5/5 2 19/20 10 30/32 7 0/9 0 3/3 0 3/3 2 4/4 0 10/10 2 10110 7 2/2 0 6/6 0 6/8 6 1/4 1

Endometrium Uterine cervix Ovary Lung (NSCLC) (SCLC) Parotid Pancreas Thyroid Urinary bladder Kidney Prostate Skin Melanocytes Brain (Gliomas)

5 0 0 0 0 0 0 0 0 0 0 3 1 3 0 2 0 1 0 1 0 0 0 0

Cell & PCN 9 2 2 1 1 4 5 0 0 1 4 0 8 20 0 1 1 3 8 2 2 6 0 0

1 FP = Fraction Positive. - 2 PCN = Periphery of neoplastic cell nests. - 3 CC = cholangiocarcinoma, HC = hepatocarcinoma. 4 DIC = ductal infiltrating carcinoma, LIC = Lobular infiltrating carcinoma, MED = medullary carcinoma, Muc = mucinous carcinomas, Und = undifferentiated carcinoma.

In all the other cases mAb M-Kid 2 stained very weakly only the neoformed biliary ducts and vessels. In breast carcinomas, various degrees of expression of Vla-3 were observed. 25% of the ductal infiltrating tumors showed a weak and heterogeneous staining with mAb M-Kid 2, whereas 60% of the lobular infiltrating carcinomas retained a homogeneous expression of the u3Bl heterodimer with an overall cell surface distribution. Three medullary breast carcinomas and one undifferentiated lesion also showed consistent expression of the Vla-3 integrin (Table 3). In all the tumor lesions analyzed, the expression of this adhesion molecule was apparently not correlated to the degree of tumor differentiation as evaluated histopathologically. SCLC and three gliomas out of four tested were unreactive with mAb M-Kid 2 over a wide range of antibody concentrations.

Discussion The current knowledge regarding the structural and functional relevance of the integrin class of molecules is mainly derived from in vitro studies l , 13. Scanty informat~on is available about their in vivo pattern of expresSIOn.

As experimental data indicate that these heterodimers may be involved in tumor invasion and metastasis, and that malignant transformation is associated with a change in integrin expression andlor structure9, 14,30, the study of their distribution in normal and transformed tissues is clearly of major relevance. The value of this information is underlined by the demonstration of I) the preferential expression of the lIb-IlIa like glycoproteins in metastatic melanoma 23 , II) the expression of the u6B4 complex in highly metastatic tumor variants 8, III) the modulation of expression of the u6B l 27 and B3 integrin subunits 2 in tumor progression and IV) the prognostic value of altered integrin expression in squamous carcinoma35 . Vla-3 identified only by a limited number of mAbs has been described as a promiscuous receptor for fibronectin, laminin and collagen 13 • Of interest from the functional point of view, expression of u3Bl has been correlated with the adherence of cells to culture dishes 3l . At present, a limited number of studies describing the in vivo distribution of Vla-3 in normal and transformed human tissues, which can clarify its relevance in the malignant phenotype, are available 29 • The recent development in our laboratory of the murine mAb M-KID 2 with high affinity for the U3~l complex3, has allowed us to address this issue. As reported by Rettig and Old3l using mAb J 143 to the U3 chain on a large panel of long-term tumor cell lines, we have confirmed both immunohistochemically and immunochemically that the expression of Vla-3 is mostly, although not exclusively, restricted to the plastic adherent cell fraction. In addition we have shown that this property is not the result of adaptation to in vitro growth, but it characterizes tumors of recent in vitro propagation as well as freshly collected tumor cells. Furthermore, despite their adherent growth pattern, some cells lines of neuroectodermal

392 . A. Bartolazzi et al.

lineage do not express detectable levels ofVla-3. At least in microcitoma this has been confirmed in vivo (see below). Testing mAb M -KID 2 on normal tissues, while confirming the expression of the a3~1 complex in epidermis 24, renal tubules, podocytes and Bowman's capsule 19 , breast myoepithelial and alveolar cells 18 and pancreatic ducts 12 has demonstrated that the heterodimer has a wide tissue distribution and is confined to the baso-Iateral or, less frequently, to the basal aspect of most mono- and multilayered epithelia. Testing the expression of the Vla-3 integrin in a large panel of solid primary tumors has shown that although changes in the pattern of distribution of the Vla-3 occur following malignant transformation, the expression of this heterodimer is rarely lost. Lack of detectable levels of this heterodimer has in fact been observed in 13% of the colon rectum cancers and in breast carcinomas 18 ,29, in which 75% of the invasive ductal carcinomas and 40% of the invasive lobular carcinomas were unreactive with mAb M-KID 2. Due to the unorganized multilayer arrangement of tumor cells, the ordered baso-Iateral stain is often completely lost. In neoplastic lesions we have identified three patterns of distribution of the a3~1 heterodimer with no apparent correlation with the degree of tumor differentiation, namely: A) homogeneous distribution on the neoplastic cell plasma membrane; B) distribution at the periphery of the neoplastic cell nests; C) homogeneous distribution on the cell plasma membrane with polarization at the periphery of the cell nests. As described by Hall et al. 12 we have observed a high degree of expression of Vla-3 in pancreatic adenocarcinomas. Furthermore, we have documented the expression of Vla-3 in cholangio carcinomas and as described for the a6~4 complex22 the lack of a3~1 in small cell lung tumors. In conclusion the findings of this study provide the basic information to critically evaluate the role of a3~1 complex in malignant transformation. Further areas of investigation should consider the study of this integrin molecule in metastatic disease and in a well-characterized model of tumor progression i.e. malignant melanoma.

Acknowledgement This study was supported by Associazione Italiana per la Ricerca sui Cancro; CNR PF ACRO; A.B. is a fellow of AIRC.

References 1 Albelda SM, Buck CA (1990a) Integrins and other cell adhesion molecules. FASEB J 4: 2868-2880 2 Albelda SM, Mette SA, Elder DE, Stewart R, Damjanovich L, Herlyn M, Buck CA (1990b) Integrin distribution in malignant melanoma: Association of the ~3 subunit with tumor progression. Cancer Res 50: 6757-6764 3 Bartolazzi A, Fraioli R, Tarone G, Natali PG (1991) Generation and characterization of the murine monoclonal

antibodies M-KID 2 to VLA-3 integrin. Hybridoma 10: 707-720 4 Bonner WM, Laskey RA (1974) A film detection method for tritium labelled proteins and nucleic acids in polyacrylamide gels. Eur J Bioch 46: 83-87 5 Chen FA, Repasky EA, Bankert RB (1991) Human lung tumor-associated antigen identified as an extracellular matrix adhesion molecule. J Exp Med 173: 1111-1119 6 Del Senno L, Anzanel D, Barbieri D, Barbieri R, Buzzoni D Rossi P, Piva R, Gambari R (1986) KJ 29 cell line isolated from a kidney adenocarcinoma: morphological aspect and effects of chemical inducers. Cell Bioi Int Rep 10: 195-196 7 De Strooper B, Van der Schueren B, Jaspers M, Saison M, Spaepen M, van Leuven F, van den Berghe H, CassimanJJ (1989) Distribution of the ~1 subgroup of the integrin in human cell and tissues. J Histochem Cytochem 37: 299-307 8 Falcioni R, Kennel SJ, Giacomini P, Zupi G, Sacchi A (1986) Expression of tumor antigen correlated with metastatic potential of Lewis lung carcinoma and B 16 melanoma clones in mice. Cancer Res 46: 5772-5778 9 Fearon ER, Cho KR, Nigro JM, Kern SE, Simons JW, Ruppert JM, Hamilton SM, Preisinger AC, Thomas G, Kinzler KW, Vogelstein B (1990) Identification of a chromosome 18 q gene that is altered in colorectal cancers. Science 247: 49-56 10 Giacomini P, Aguzzi A, Pestka S, Fisher PB, Ferrone S (1989) Modulation by recombinant DNA leukocyte (a) and fibroblast (~) interferons of the expression and shedding of HLA and tumor-associated antigens by melanoma cells. Eur J Immunol15: 946-951 11 Giancotti FG, Ruoslahti E (1988) Elevated levels of the aSP1 fibronectin receptor suppress the transformed phenotype of Chinese Hamster ovary cells. Cell 60: 849-859 12 Hall PA, Coates P, Lemoine NR, Horton MA (1991) Characterization of integrin chains in normal and neoplastic human pancreas. J Pathol165: 33-41 13 Hemler ME (1990) Via proteins in the integrin family: Structures, functions, and their role on leukocytes. Ann Rev Immunol 8: 365-400 14 Hirst R, Horwitz A, Buck C, Rohrschneider L (1986) Phosphorylation of the fibronectin receptor complex in cells transformed by oncogenes that encode tyrosine kinases. Proc Nat! Acad Sci USA 83: 6470-6474 15 Humphries MJ, Yamada KM, Olden K (1988) Investigation of the biological effects of anti-cell adhesion synthetic peptides that inhibit experimental metastasis of B 16-F 10 murine melanoma cells. J Clin Invest 81: 782-790 16 Hynes RO (1987) Integrins: a family of cell surface receptors. Cell 48: 549-554 17 Juliano R (1987) Membrane receptors for extracellular matrix macromolecules: relationships to cell adhesion and tumor metastasis. Biochim Biophys Acta 907: 261-278 18 Koukoulis GK, Virtanen I, Korhonen M, Laitinen L, Quaranta V, Gould VE (1991) Immunohistochemicallocalization of integrins in the normal, hyperplastic and neoplastic breast. Am J Path 139: 787-799 19 Korhonen M, Ylanne J, Laitinen L, Virtanen I (1990) The ar-U6 subunits of integrins are characteristically expressed in distinct segments of developing and adult human nephron. J Cell Bioi 111: 1245-1254 20 Laemli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T 4. Nature 227: 680-685 21 Laudonio N, Zupi G, Storniello G, Salvati F, Greco C (1989) Establishment of cell lines from serial samples from two patients with lung tumours before and after chemotherapy. Cytotechnology 2: 39-47 22 Mariani Costantini R, Falcioni R, Battista P, Zupi G, Kenel

Vla-3 in Human Tissues . 393 SJ, Colasante A, Venturo I, Gallo Curcio G, Sacchi A (1990) Integrin a6~4 expression in human lung cancer as monitored by specific monoclonal antibody. Cancer Res 50: 6107-6112 23 McGregor BC, McGregor JL, Weiss LM, Wood GS, ChungHong H, Boukerche H, Warnke RA (1989) Presence of cytoadhesines (lIb-IlIa like glycoproteins) on human metastatic melanomas but not on benign melanocytes. Am J Clin Pathol 92: 495-499 24 Morhenn VB, Schreiber AB, Soriero 0, McMillian W, Allison AC (1985) A monoclonal antibody against basal cells of human epidermis. J Clin Invest 76: 1978-1983 25 Mottolese M, Venturo I, Digiesi G, Perrone Donnorso R, Bigotti A, Muraro R, Aluffi A, Natali PG (1990) Use of MoAb D 612 in combination with a panel of MoAbs for the immunocytochemical identification of metastases from colon-rectum. Br J Cancer 61: 626-630 26 Natali PG, Imai K, Wilson BS, Bigotti A, Cavaliere R, Pellegrino MA, Ferrone S (1981) Structural properties and tissue distribution of the antigen recognized by the monoclonal antibody 653-40S to human melanoma cells. J Nat! Cancer Inst 67: 591-601 27 Natali PG, Nicotra MR, Cavaliere R, Giannarelli D, Bigotti A (1991) Tumor progression in human melanoma is associated with changes in a6/~1 laminin receptor. Int J Cancer 49: 168-172

28 Pignatelli M, Bodmer WF (1990) Integrin cell adhesion molecules and colorectal cancer. J Pathol 162: 95-97 29 Pignatelli M, Hanby AM and Stamp GWH (1991) Low expression of ~h a2 and a3 subunits of Via integrins in malignant mammary tumours. J Pathol165: 25-32 30 Plantefaber LC, Hynes RO (1989) Changes in integrin receptors on oncogenically transformed cells. Cell 56: 281-290 31 Rettig WJ, Old LJ (1989) Immunogenetics of human cell surface differentiation. Ann Rev Immunol 7: 481-511 32 Ruoslahti E, Pierschbacher MD (1987) New perspectives in cell adhesion: RGD and integrins. Science 238: 491-497 33 Ruoslahti E, Giancotti FG (1989) Integrins and tumor cell dissemination. Cancer Cells 1: 119-126 34 Stamp GWH, Pignatelli M (1991) Distribution of ~h ah a2, and a3 integrin receptors in basal cell carcinomas. J Pathol163: 307-313 35 Wolf GT, Carey TE, Schmaltz SP, McClatchey KD, Poore J, Glaser L, Hayashida DJS, Hsu S (1990) Altered antigen expression predicts outcome in squamous carcinoma of the head and neck. J Nat! Cancer Inst 82: 1566-1572 36 Zutter MM, Mazoujian G, Santoro SA (1990) Decreased expression of adhesive protein receptors in adenocarcinoma of the breast. Am J Pathol137: 863-870

Received July 6, 1992 . Accepted October 5, 1992

Key words: Adhesion molecules - Integrins - Monoclonal Antibodies Dr. Pier Giorgio Natali, Laboratory of Immunology, Regina Elena Cancer Institute, Viale Regina Elena 291, I-00161-Rome, Italy, Tel: 39-6-4985527, Fax: 39-6-4180473