Monoclonal Antibody to a Proximal Nephrogenic Renal ...cancerres.aacrjournals.org/content/canres/49/7/1802.full.pdf · Monoclonal Antibody to a Proximal Nephrogenic Renal Antigen:

[CANCER RESEARCH 49, 1802-1809, April 1, 1989]

Monoclonal Antibody to a Proximal Nephrogenic Renal Antigen:Immunohistochemical Analysis of Formalin-fixed, Paraffin-

embedded Human Renal Cell CarcinomasSamuel O. Yoshida1 and Ashraf Imam

Department of Pathology, CIGNA Hospital, and the University of Southern California, School of Medicine, Los Angeles, California 90033

ABSTRACT

An IgGl murine monoclonal antibody (PN-1S) to an epitope preservedin formalin-fixed, paraffin-embedded tissue sections of renal cell carcinomas was developed following immunization of mice with the microso-inal fraction of human renal cortical tissue homogenates. An immunohis-tochemical screening procedure was utilized against fixed, embeddedtissue sections of renal cell carcinomas. This antibody identified a M,200,000 determinant on immuno-Western blot analysis, and the recognized epitope was shown to reside in the carbohydrate domain of a M,200,000 glycoprotein (gp200). Analysis of its tissue distribution wasperformed by indirect immunoperoxidase staining of fixed, embeddedtissue sections of the following: 36 different types of normal tissues; 45primary and 38 metastatic renal cell carcinomas; and 124 cases of 39different types of neoplasms. In the normal kidney, gp200 was localizedalong the brush border of the proximal tubule and the luminal surface ofBowman's capsule adjoining the outgoing proximal tubule. Shared tissue

localization was also noted in breast lobules and ducts, parathyroid gland,and epididymis. The gp200 was expressed by 93% of primary and 84%of metastatic renal cell carcinomas, primarily on the cell surface membrane. Shared antigenic expression was noted in relatively few othertumors including: mammary carcinomas; teratocarcinomas; and parathyroid adenomas. The PN-15 monoclonal antibody holds promise as auseful reagent in retrospective investigations of fixed, embedded cases ofrenal cell carcinomas. The antibody may also be of practical utility indiagnostic immunohistochemistry by helping to retrospectively define theprimary site of metastatic clear cell tumors, especially when fresh frozentissue from such tumors is not available.

INTRODUCTION

There have been many recent studies dealing with the development of murine monoclonal antibodies immunoreactive withhuman renal cell carcinomas (1-12). Many of the antibodiesare directed against tissue antigens of relative restricted distribution, shared by renal cell carcinomas, segments of the normalrenal nephron, and a few other tissues. A few reports describemonoclonal antibodies that recognize "tumor-associated" an

tigenic determinants present on renal cell carcinomas but absentin the normal kidney (7-12).

With the exception of two reports (6, 11), these murinemonoclonal antibodies were shown to recognize epitopes thatare preserved only in cryostat sections. Many of these latterdeterminants were apparently destroyed by the process of fixation and/or paraffin embedment. Antibodies that react withantigens that are preserved in fixed-paraffin embedded tissuewould be of value in retrospective investigations and may be ofpractical use in diagnostic immunohistochemistry. With this inmind, a murine monoclonal antibody, designated PN-15, wasdeveloped against an epitope of a renal antigen preserved infixed-embedded tissue sections of renal cell carcinomas. Characterization of this antigen as well as extensive determination

Received 6/8/88; revised 12/6/88; accepted 1/3/89.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1To whom requests for reprints should be addressed, at CIGNA Hospital,1711 West Temple Street, Los Angeles, CA 90026.

of its distribution in neoplastic and normal human tissues hasbeen carried out.

MATERIALS AND METHODS

Affinity-purified immunoglobulin G fraction of horse antimouseimmunoglobulin and ABC2 were bought from Vector Laboratories,

Burlingame, CA. Reagents for electrophoresis and Western blottingwere purchased from Bio-Rad Laboratory, Richmond, CA. The rest ofthe reagents used were of the highest purity available from SigmaChemical Co., St. Louis, MO.

Preparation of Tissue Homogenate. Fresh normal renal cortical tissuewas obtained from a kidney immediately after nephrectomy for a renalcell carcinoma. The renal capsule was stripped, and normal renalcortical tissue, uninvolved by tumor, was dissected off. Subsequentpreparation of the microsomal fraction of the renal cortical tissue wascarried out at 0-4Â°Cas described by Nairn et al. (13). This preparation

was snap frozen in an acetone-dry ice freezing mixture and stored at-70Â°C.

Cell Fusion. Three-wk-old BALB/c mice were immunized with 3separate injections of the extract of renal cortical tissue at a 3-wkinterval. The spleen was removed on the fourth day after the lastinjection, and the spleen cells were fused with the mouse myeloma cellline (a nonsecretory variant of the mouse myeloma cell subline of SP2/OAg 14) as first described by Kohler and Milstein (14). Postfusionproduct was suspended in Dulbecco's modified Eagle's medium, containing high glucose (4 g/liter) and 1 x IO"4M hypoxanthine, 4 x 10~7M am inopter ine, and 6 x 10~5M thymidine, and plated into 2- x 96-well culture plates. The culture was incubated at 37Â°Cin a humidified

atmosphere of 5% CO2 for 12 days before screening of supernatant forthe presence of antibody as described below.

Screening, Selection, and Cloning of Hybrids. The initial screening ofsupernatants from wells exhibiting hybrid growth was performed im-munohistologically by using formalin-fixed and paraffin-embedded tissue sections that contained a case of a renal carcinoma and adjacentuninvolved kidney. Supernatant (100 pi) from wells showing hybridgrowth was applied to tissue sections and incubated for 30 min. Theslides were washed with PBS, and 100 n\ of biotinylated horse anti-mouse immunoglobulin, in an appropriate dilution, were added to each.After an incubation of 30 min, slides were again washed as above andthen incubated with 100 Â¿ilof appropriately diluted ABC for 30 min.Following a wash with PBS, the bound ABC was visualized by additionof a mixture of the substrate, H2Oz, and chromogen, aminoethylcarba-zole. Hybrids secreting antibodies in supernatants with strong reactivityto renal carcinoma cells and cells lining the proximal tubules werecloned by limiting dilution. Supernatants from wells with single clonalgrowth were again subjected to the above screening procedure thatincluded fixed and embedded tissue sections of primary and metastaticrenal cell carcinomas and also primary adenocarcinomas of lung andliver. The selected clones secreting antibody with strong reactivity toboth primary and metastatic renal carcinoma cells and no detectablereactivity to lung and liver carcinoma cells were subjected to one morecycle of recloning before making a final selection of clones.

Isotyping of the Selected Monoclonal Antibodies. The selected antibody, designated PN-1S, was characterized for its immunoglobulin

2The abbreviations used are: ABC, avidin-biotin-peroxidase complex; SDS,sodium dodecyl sulfate; PBS, 0.02 M sodium phosphate buffer, pH 7.2, containing0.15 M NaCl; TCA, trichloroacetic acid; PAGE, polyacrylamide gel electrophoresis; gp200, M, 200,000 glycoprotein; MFGM, milk fat globule membrane.

1802

on June 5, 2018. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

ANTIBODY TO MEMBRANE ANTIGEN OF HUMAN PROXIMAL NEPHRON

isotype using rabbit antibodies to isotypes of mouse immunoglobulinin an Ouchterlony double diffusion method.

Determination of Epitope Recognized by Monoclonal Antibody togp200. An investigation was conducted to determine whether the antibody was directed to the protein and/or the carbohydrate portion ofgp200 as described previously in detail (15). Briefly, 250 Â¿igof proteinof renal cortex extract were either treated with endo-/3-Ar-acetylglucos-aminidase H, pepsin, or heat. The supernatants and the TCA-precipi-table fractions resulting from treatments with endoglycosidase, pepsin,or heat were individually mixed with an equal volume of the antibody(1 mg of IgG/ml) solutions. Untreated PN-15 served as controls. Themixtures were incubated at 4Â°Cfor 16 h. Following incubation, themixtures were centrifuged at 100,000 x g and 4Â°Cfor 30 min, and the

absorbed antibody solutions were subsequently used for immunostain-ing sections of renal tissue.

Western Blot Analysis. Following the separation of detergent-solublematerial from the immunogen, breast carcinoma tissue, a mammarycarcinoma cell line, MCF7, or a renal cell carcinoma cell line, RC-7(16), by SDS-PAGE, the bands were electrophoretically transferred tonitrocellulose filter paper as described by Towbin et al. (17). Each lanewas cut from the filter paper and incubated with PBS containing 1%(w/v) gelatin for 1 h at room temperature, to block the nonspecificbinding of antibody to filter paper. Each strip of filter paper wasincubated for 30 min with 10 ml of PBS and 100 n\ of spent mediumcontaining the antibody or an equivalent amount of an irrelevantantibody of the same isotype. Following incubation the strips werewashed thoroughly with PBS containing 0.05% (v/v) Tween 20, thenincubated with horseradish peroxidase-conjugated goat anti-mouse immunoglobulin, in an appropriate dilution, for 30 min. The strips wereonce again washed as above. Finally, the color was developed byincubating the strips with PBS containing 1 m\i diaminobenzidine and0.01% hydrogen peroxide for 5 min.

Tissue Specimens and Immunohistochemistry. Formalin-fixed andparaffin-embedded tissue sections of 45 primary renal cell carcinomas,38 metastatic renal cell carcinomas, 124 other epithelial and mesenchy-mal neoplasms from multiple organ systems, and 36 types of normaltissue were obtained from the surgical pathology files of CIGNAHospital, Los Angeles, CA, and the Memorial Sloan-Kettering CancerCenter, New York, NY. For immunoperoxidase staining, 5-nm sectionsof paraffin-embedded tissue were stained using the avidin-biotin per-oxidase complex technique. Sections were incubated with 2 to 3 Mgofprotein of the IgG fraction of ascites fluid diluted with 5% bovineserum albumin/phosphate-buffered saline. Tissue sections were firstdigested with 0.05% protease XXIV (Sigma) for 10 min and thenincubated with the primary antibody for 45 min at 25Â°C.The rest of

the procedure was as described previously (18). For negative controls,nonimmune mouse serum was used in place of the primary antibody.Visual estimates of the percentage of cells staining were placed intofour rough categories: 0% staining (-); 1 to 9% staining (+); 10 to 49%

staining (++); 50 to 100% staining (+++).Immunohistochemical analysis was also performed on a limited

number of fresh frozen normal and neoplastic tissues obtained fromsurgically resected specimens. These included kidney, breast, parathyroid, thyroid, salivary gland, stomach, colon, lung, lymph node, spleen,muscle, skin, prostate, epidydimis, gastric adenocarcinoma, colonieadenocarcinoma, pulmonary adenocarcinoma, fibroadenoma, mammary duct carcinoma, non-Hodgkin's lymphoma (B-cell), Hodgkin's

disease (mixed cell), seminoma, and renal cell carcinoma.Absorption of Monoclonal Antibody to gp200 with Blood Group

Antigens. Absorption of the monoclonal antibody to gp200 againstblood group antigens was performed as follows. Equal volumes ofconcentrated human A,B substances (Neutr AB; DADE) and Lewisblood group substances (ORTHO) were each added separately to 5 ^gof protein of the IgG fraction of ascites fluid diluted to 250 n\ with 5%bovine serum albumin/phosphate-buffered saline. Following a 60-minincubation at room temperature and centrifugation, 400 ul of thesupernatant were analyzed immunohistochemically against tissue sections containing a renal cell carcinoma/adjacent normal kidney. Asimilar concentration of unabsorbed PN-15 antibody served as a positive control. Absorption against other minor blood group antigens was

performed using human Group O reagent red blood cells of knownminor blood group phenotypes (Spectrogen-Duo; ORGANON TEK-NIKA). The following blood group phenotypes are expressed by thereagent RBC: C; D; E; c; e; C"; f; V; K; k; Kp"; Kpb, Js"; Jsb; Fy"; Fy";Jka; Jkb; Le'; Le"; P,; M; N; S; s; Lu'; Lu"; and Xg". 250-^1 "packed cellvolume" of the latter red cells was added to 500 n\ (10 Â¿igof protein of

IgG) of cell culture supernatant from the PN-15 antibody-producingclone. Following 60-min incubation at room temperature and centrifugation, 400 id of the supernatant were used for immunohistochemicalanalysis as noted above. No hemolysis was encountered. UnabsorbedPN-15 antibody served as a positive control.

Absorption of Monoclonal Antibody to gp200 with Human Milk-Fat-Globule Membrane and Epithelial Membrane Antigens. The antibody(1 mg of immunoglobulin per ml) was separately mixed and incubatedwith 2 mg of protein of each of defatted human MFGM (19) orepithelial membrane antigen (20) immobilized to Sepharose 4B. Following the overnight incubation, the mixtures were centrifuged at12,000 x g and 4"C for 15 min. The supernatants containing absorbed

antibody were removed and subsequently applied to tissue sections fortheir immunostaining analysis.

RESULTS

Generation and Selection of Monoclonal Antibody

Immunization with the human renal cortical extract andsubsequent fusion using the spleen cells resulted in 116 wellswith growing hybridoma cultures. Immunohistochemical testing of the supernatants for reactivity with fixed and embeddedtissue sections of a renal cell carcinoma/adjacent normal kidneyshowed 4 that reacted with the proximal renal tubular epithelium and renal carcinoma cells. These latter 4 hybridoma cultures were subcloned by limiting dilution, resulting in 119 wellswith single-clonal growth. Immunohistochemical testing of thesupernatants from these subclones was performed with fixedand embedded tissue sections of normal kidney, primary andmetastatic renal cell carcinomas, pulmonary adenocarcinoma,and a hepatocellular carcinoma. The subclones from one ofthe original hybridoma cultures yielded supernatants that werebroadly reactive with all of the carcinomas tested. The supernatants of the subclones from another of the original hybridomacultures were only very weakly reactive with the proximal renaltubules and renal carcinoma cells. Subclones from the othertwo original cultures produced supernatants that reacted, withvariable intensity, with only proximal renal tubular epitheliumand renal carcinoma cells. Three subclones that showed themost intense reactivity with the proximal tubular epitheliumand renal carcinoma cells were expanded for further large scaleantibody production. Of these three, the antibody that demonstrated the most intense immunohistochemical staining, withthe least amount of background staining of tissue sections, wasselected for further studies.

The selected monoclonal antibody demonstrated strong immunohistochemical reactivity in formalin-fixed, paraffin-embedded tissue sections following predigestion with protease.Protease digestion enhanced immunoperoxidase staining forthe gp200 antigen, which was weak to moderate in intensityprior to protease digestion.

Pattern of Reactivity with Normal Human Tissue

The distribution of the gp200 antigen is noted in Table 1. Inthe normal kidney the gp200 antigen is expressed by the brushborder of the pars convoluta and pars recta segments of theproximal tubule, as well as Ideally along the luminal surface ofBowman's capsule adjoining the outgoing proximal tubule (Fig.l.-l). Of other normal tissues examined, the antigen is also

1803




localized along the luminal surfaces of the breast lobules andducts, the luminal surface of the epididymal tubular epithelium,within the cytoplasm of parathyroid parenchyma! cells, andfocally within the colloid of thyroid follicles. Thirty-one othernormal tissues failed to express similar or cross-reacting antigens.

Table 1 Immunohistochemical localization ofgp200 in normal human tissuesThree cases of each of the following normal organs were immunohistochemi-

cally tested for the expression of gp 200 and were found negative: adrenal gland;ampulla of Vater; appendix; bladder; bone; bone marrow; cartilage; colon; common bile duct; duodenum; endometrium; esophagus; fallopian tube; gallbladder;ileum; jejunum; liver; lung; lymph node; myometrium; ovary; pancreas; prostate;renal pelvis; skin; spleen; stomach; submandibular gland; testis; tonsil; thymus;and ureter.

OrganKidneyBreastEpididymis

ParathyroidThyroidComponent

stainedGlomerulus

Bowman's capsule

Proximal tubuleLoop of HenleDistal tubuleCollecting ductRenalpelvisLobule

Terminal ductSegmental ductNippleductEpithelial

cellsParenchyma!cellsColloidNo.

positive/no. ofcases3/33/33/3

4/43/3%

of cells"

stained+:++*++++

Â°Sections were recorded for the estimated percentage of cells staining on thefollowing scale: 0% (-); 1 to 9 (+); 10 to 49% (++); 50 to 100% (+++).

The pattern of immunoreactivity in the limited number offresh frozen normal tissues examined was similar to that offixed and embedded tissues. Positive staining was noted withinBowman's capsules and proximal tubules of the kidney, termi

nal ducts and lobules of the breast, parathyroid parenchymalcells, and tubules of the epidydimis.

Pattern of Reactivity with Neoplasms

The frequency and degree of antigen expression in primaryand metastatic renal cell carcinomas are shown in Tables 2 and3. Histological subtypes aside, 42 of 45 (93%) of primary and32 of 38 (84%) of metastatic renal cell carcinomas expressedthe gp200 antigen. There was considerable heterogeneity ofantigen expression in both primary and metastatic renal cellcarcinomas with a percentile range of 1% to 99% of tumor cellsstaining. While some tumors were uniformly positive and otherswere uniformly negative, many tumors demonstrated a "patchwork" heterogeneic pattern of antigen expression. Subpopula-

tions of tumor cells were seen to express the antigen, whileadjacent morphologically similar cells within the same tumorwere entirely negative (not illustrated). The gp200 antigen wasprimarily localized on the cell surface membrane of immuno-reactive tumor cells, while cytoplasmic localization was notedless consistently (Fig. 1, A to E). Areas of sarcomatoid transformation in otherwise typical renal cell carcinomas were non-reactive (not illustrated).

In 9 cases both the primary renal cell carcinoma and theircorresponding mÃ©tastaseswere available for analysis (Table 4).In general, no clear differences in gp200 antigen expression

Fig. 1. .!. immunohistological localizationof gp200 antigen with PN-15 monoclonal antibody using avidin-biotin-peroxidase complexstaining method. Strong immunoreactivity isnoted along the brush border membrane surfaces of proximal tubules. Focal immunoreactivity is also present on the luminal surface ofBowman's capsule adjoining the outgoing

proximal tubule (arrow). Glomerulus, distaltubules, and interstitium are unstained. Im-munoperoxidase technique, hematoxylincounterstain, X 100. B and C, primary renalcell carcinomas, clear cell (B), and papillarytypes (C). Intense expression of gp200 is notedon the membrane surfaces of tumor cells. Im-munoperoxidase technique, hematoxylincounterstain, x 100. D, metastatic renal cellcarcinoma in cervical lymph node. Lymphnode capsule is seen in right side of field.Stained tumor cells are seen surrounding unstained residual germinal centers. Immunoper-oxidase technique, hematoxylin counterstain,x 40. /. metastatic renal cell carcinoma inlung. Unstained bronchiole, with an intralu-minal "plug" of stained tumor cells, is seen in

right side of field. Immunoperoxidase technique, hematoxylin counterstain, x 100.

.â€¢¿�^,*^V2?&1*:r.

1804




Table 2 Immunohistochemical localization ofgp200 antigen in primary renalcell carcinomas

Table 4 Expression ofgp200 antigen in primary and corresponding metastaticrenal cell carcinomas from the same patient

HistolÃ³gica! subtypeNo. ofcases

No. positive/no. of cases

% of cells"

stainingClear cell-granular cell

Grades I-II* 17

Grades III-IV 20

PapillaryGrade II 4Grade III 3

Sarcomatoid 1

Total 45

12/173/172/17

17/1712/203/202/20T772D

4/43/31/1

42/45(93)'

" Sections were recorded for the estimated percentage of cells staining on thefollowing scale: 0% (-); 1 to 9% (+); 10 to 49% (++); 50 to 100% (+++).

* Nuclear grading of tumor assessed as outlined by Fuhrman et al. (35).c Number in parentheses, percentage.

Table 3 Immunohistochemical localization ofgp200 antigen in metastatic renalcell carcinomas

MetastaticsiteLungBoneDistant

lymphnodesRegional

lymphnodesSofttissuesBrainLiverAdrenal

glandThyroidglandDuraBreastTotalNo.

ofcases11942322211138No.

positive/ % of cells"

no. of casesstaining7/1

1+++2/11++97TT7/9

+++1/9++2/4

+++1/4++1/4+4732/2

+++1/3+++2/2

+1/2+++1/2'+++1/1

++1/1+++0/132/38

(84)c

" Sections were recorded for the estimated percentage of tumor cells stainingon the following scale: 0% (-); 1 to 9% (+); 10 to 49% (++); 50 to 100% (+++).

'' Negative case represents a purely sarcomatoid metastasis.c Number in parentheses, percentage.

were noted between the primary tumors and their mÃ©tastases,with 3 notable exceptions. In these latter 3 cases the mÃ©tastasesdemonstrated a considerable paucity of gp200 expression relative to their corresponding primaries. Also, these 3 cases differed clinically from the other 6 by the fact that the metastatictumor was excised 9 mo or longer after radical nephrectomyfor renal cell carcinoma. In the other 6 cases the mÃ©tastaseswere biopsied or excised at the time or, as in one case, veryshortly prior to nephrectomy.

One hundred twenty-four other epithelial and mesenchymalneoplasms of 39 histological subtypes were examined to determine the pattern of gp200 localization in tumors other thanrenal cell carcinomas (Table 5). As noted, the majority ofmammary duct carcinomas and all the mammary lobular carcinomas were immunoreactive. Similar to renal cell carcinomasthe antigen was localized on the cell surface membrane of tumorcells, with less consistent localization within the cytoplasm.Focal cell surface membrane immunoreactivity was also notedwithin the embryonal carcinomatous component of 2 of 4testicular teratocarcinomas. All parathyroid adenomas exam-

Case Site

% of cells"

staining

1 PrimaryMetastasis, humÃ©rus

2 PrimaryMetastasis, adrenal

3 PrimaryMetastasis, regional lymph node

4 PrimaryMetastasis, liver

5 PrimaryMetastasis, breast*

6 PrimaryMetastasis, paracaval lymph nodeMetastasis, soft tissue, arm''

7 PrimaryMetastasis, thyroid''Metastasis, lung'

8 PrimaryMetastasis, adrenal

9 PrimaryMetastasis, paracaval lymph nodeMetastasis, liver


* Nine mo postnephrectomy.' Ten mo postnephrectomy.* Eleven and one-half yr postnephrectomy.' Twelve and one-half yr postnephrectomy.

ined demonstrated both cell surface membrane and cytoplasmicimmunoreactivity. No evidence of gp200 localization was notedin multiple cases from 35 other histolÃ³gica! subtypes of neoplasms.

The pattern of immunoreactivity in the limited number offresh frozen neoplastic tissues examined was similar to thefixed and embedded neoplastic tissues. Positive staining wasnoted in a renal cell carcinoma and in the epithelial componentof a fibroadenoma. A case of a mammary duct carcinoma wasnonreactive, similar to one-third of the fixed and embeddedmammary duct carcinomas.

Immunohistological Analysis of the Absorbed Antibody

Absorption of the PN-15 monoclonal antibody with the concentrated A,B, substances, Lewis blood group substances, andGroup O human reagent red blood cells produced no diminutiveeffect upon the intensity or the percentage of renal carcinomacells staining by immunohistochemical methods. According tothe manufacturers' specifications the concentrated A and B

substances (DADE) should neutralize anti-A and anti-B antibodies, and the Lewis blood group substances (ORTHO) shouldneutralize anti-Le", anti-Leb, and anti-Le" antibodies.

Both unabsorbed and absorbed antibodies with defattedMFGM or epithelial membrane antigen also showed an indistinguishable intensity of immunostaining of cells in tissue sections.

Characterization of Epitope Recognized by Monoclonal Antibody. The immunological activity of the antibody was abolishedby absorption with TCA-soluble fractions of renal cortical tissuefollowing endo-0-jV-acetylglucosaminidase H treatment. Absorption with the TCA-precipitable fraction of the endoglyco-sidase-treated or pepsin-treated tissue extract had no effect onthe intensity of the immunostaining, suggesting that the recognized epitope is expressed in the carbohydrate domain of theglycoprotein.

Characterization of Antigen Recognized by the Antibody. Theantigen specifically recognized by the monoclonal antibody PN-15 was analyzed by immuno-Western blot analysis. Western

blot analysis of nonreduced extracts of the immunogen or a1805




Table 5 Immunohistochemical localization ofgp200 antigen in tumors otherthan renal cell carcinomas

HistologyNo. positive/no. of cases

% of cellsstaining"

Intraductal and infiltrating duct card- 8/12 + to +++noma, breast

In situ and infiltrating lobular carcinoma, 6/6 + to +++breast

Parathyroid adenoma 5/5 +++Teratocarcinoma (embryonal carcinoma 2/4 +

and teratoma),' testisAdenocarcinoma, lung 0/12 â€”¿�Bronchioloalveolar carcinoma, lung 0/2 -Adenocarcinoma, stomach 0/3 -Adenocarcinoma, ileum 0/1 â€”¿�Adenocarcinoma, colon 0/3 -Adenocarcinoma, pancreas 0/3 -Cholangiocarcinoma, extrahepatic bile 0/2 â€”¿�

ductHepatocellular carcinoma 0/2 â€”¿�Adenocarcinoma, gallbladder 0/1 -Squamous carcinomas, upper respiratory 0/3 -

tractAdenoid cystic carcinoma, salivary gland 0/4 -Mucoepidermoid carcinoma, salivary 0/1 â€”¿�

glandAdenocarcinoma, endometrium 0/4 â€”¿�Adenosquamous carcinoma, endocervix 0/2 â€”¿�Adenocarcinoma, endocervix 0/1 â€”¿�Clear cell adenocarcinoma, endocervix 0/1 -Serous cystadenocarcinoma, ovary 0/2 â€”¿�Mucinous cystadenocarcinoma, ovary 0/2 -Undifterentiated adenocarcinoma, ovary 0/2 -Endometrioid adenocarcinoma, ovary 0/2 â€”¿�Pure embryonal carcinoma, testis 0/1 -Seminoma, testis 0/4 -Adenocarcinoma, prostate 0/3 â€”¿�Transitional cell carcinoma, renal pelvis 0/4 â€”¿�Transitional cell carcinoma, bladder 0/3 â€”¿�Papillary adenocarcinoma, thyroid 0/6 â€”¿�Meningioma, dura 0/3 â€”¿�Glioblastoma multiforme, brain 0/3 â€”¿�Melanoma (4 primary and 1 metastatic to 0/5 -

lymph node)Non-Hodgkin's lymphoma, B-cell (2 B-im- 0/3 -

munoblastic, 1 large NO )Non-Hodgkin's lymphoma, T-cell, immu- 0/1 â€”¿�

noblasticLeiomyosarcoma, small intestine 0/1 â€”¿�Angiosarcoma, breast 0/1 â€”¿�Synovial sarcoma, triceps muscle 0/1 â€”¿�Granular cell tumor (skin, tongue, foot) 0/3 -


* Very few tumor cells demonstrated immunoreactivity in embryonal carcino-

matous component of tumor.' Noncleaved.

renal carcinoma cell line reacted with the monoclonal antibodyon nitrocellulose filter paper as shown in Fig. 2, Lanes B andC. A component with an apparent molecular weight of 200,000in the filter strip specifically reacted with the antibody. Western-ini nianobio t analysis of antibody PN-1S and the extracts oftissue containing mammary carcinoma cells or a mammarycarcinoma cell line (MCF7) gave a similar result (not illustrated). Reduction of the extract with mercaptoethanol beforeelectrophoresis had no effect on the molecular weight of theantigen. No band was visualized when the filter strip wasincubated with an irrelevant monoclonal antibody of the sameimmunoglobulin subtype (Fig. 2, Lane D).

DISCUSSION

The majority of renal cell carcinomas are thought to behistogenetically derived from, or differentiate along the line of,the proximal renal tubular epithelium as determined by theirexpression of proximal renal tubular antigens (21, 22). A few,however, lack the expression of these antigens, suggesting their

1mol wt.)

200*

116Â»

92Â»

66Â»

Â«9P200

45Â»

A B C D

Fig. 2. Western immunoblotting of gp200 with PN-15 monoclonal antibody.Following the separation of protein from the extract of human renal corticaltissue or renal carcinoma cell line (RC.7) by SDS-PAGE under nonreducingconditions, the protein bands were electrophoretically transferred to nitrocelluloseniter paper. Each lane containing the transferred bands was cut and incubatedwith either the specific antibody or an equivalent amount of an irrelevant monoclonal antibody of the same immunoglobulin class which served as a negativecontrol. Lane A, contains molecular weight marker proteins; Lane B, proteinextracted from the renal cortical tissue and immunoblotted with PN-15 antibody;Lane C, proteins extracted from the RC.7 renal carcinoma cell line and immunoblotted with PN-15 antibody; and Lane D, like Lane B, immunoblotted withan irrelevant monoclonal antibody.

derivation from, or differentiation along the lines of, othersegments of the nephron (22). Many of the monoclonal antibodies developed against cell lines or tissue homogenates ofrenal cell carcinomas demonstrate immunoreactivity to renalantigens distributed within the proximal tumor and, to a variable extent, within the glomerulus, Bowman's capsule, distal

tubule, and collecting duct (1-6). Some of these antigens arealso shared by other types of normal and neoplastic tissues, thedegree of apparent shared distribution dependent upon theantibody/antigen being analyzed. These antigens apparentlyrepresent normal "differentiation antigens" and are expressed

in the majority of renal cell carcinomas. A few studies, however,report on monoclonal antibodies to antigenic determinantsexpressed in renal cell carcinomas but not within the normalkidney (7-12). Three such described monoclonal antibodies(G250, K45, and S22) recognize various denned and undefinedantigens expressed by 50 to 97% of the renal cell carcinomasanalyzed (7-9). However, these antibodies also react with various other normal tissues, as well as a number of epithelial andmesenchymal neoplasms. One study on four monoclonal antibodies (E6, B7, C8, D8) demonstrated reactivity with an undefined putative tumor-specific antigen present only in renal cellcarcinomas, although only 4 other types of carcinomas wereexamined (10).

From another approach, Finnish investigators have demonstrated that murine monoclonal antibodies to a M, 130,000polypeptide of the endogenous feline retrovirus RD114p30

1806




(HPS-1) and to a M, 19,000 polypeptide of the human T-cellleukemia/lymphoma virus (anti-HTLVIpl9) react with 100%of all renal cell carcinomas tested but not with normal kidneytissue (11, 12). Both antibodies also react with choriocarcino-mas, invasive moles, hydatidiform moles, and syncytiotropho-blasts of human placentas. The anti-HTLVIpl9 also recognizesT-lymphocytes of adult T-cell leukemia/lymphomas, AIDS patients and healthy carriers of the HIV infection, normal thymicepithelium, and prostatic adenocarcinomas (23, 24). Examination of 5 other types of neoplasms showed no evidence ofexpression of either antigen. Of interest is that these antibodiesreact with paraffin-embedded tissue sections, a major objectiveof our study.

Our approach to the development of murine monoclonalantibodies to a renal antigen was primarily a practical one, asreflected by an immunohistochemical method of antibodyscreening using formalin-fixed, paraffin-embedded tissue sections of a renal cell carcinoma/adjacent normal kidney. Wewere able to develop a promising monoclonal antibody of IgGlisotype that recognizes an epitope on the carbohydrate component of a glycoprotein with an apparent molecular weight of200,000 (gp200). The epitope is neither destroyed by formalinfixation nor by the rigors of tissue processing for paraffinembedment. In the normal human nephron the gp200 antigenis localized within the brush border of the pars convoluta andpars recta segments of the proximal renal tubule, as well asfocally along the luminal surface of Bowman's capsule adjoining

the outgoing proximal tubule. By its shared localization alongthe luminal surface of breast lobules and ducts, the luminalsurface of the epididymal tubular epithelium, within parathyroid parenchymal cells, and focally within the colloid of thyroidfollicles, one might conjecture a transepithelial transport orsecretory function for the gp200. However, many additionalbiochemical studies and immunoelectron microscopy are required to elucidate the putative functional properties of thisglycoprotein, in conjunction with its topographical orientationwithin the cell membrane bilayer.

Absorption studies of the PN-15 monoclonal antibodyshowed no significant cross-reactivity with other carbohydratedeterminants, such as the Lewis blood group antigens, epithelialmembrane antigen, and AB blood group antigens, known to beexpressed in renal cell carcinomas and/or normal kidney tissues. The Le" and Le" determinants have been localized withinthe proximal tubule and some renal cell carcinomas (25-27).However, in contrast to the gp200 antigen, their distributionwithin other normal and neoplastic tissues is much more widespread. They are also expressed in the stomach, small intestine,colon, appendix, pancreas, salivary gland, biliary ducts, bronchial mucosa, gallbladder, squamous epithelium, transitionalepithelium, endocervical and endometrial glands, small cellcarcinomas of lung, gastric carcinomas, colonie carcinomas,pancreatic carcinomas, and gallbladder carcinomas (25-27).The epithelial membrane antigen, which is found in the majorityof renal cell carcinomas, is localized to the distal tubule andcollecting ducts of the normal kidney (28). By contrast, thegp200 antigen is localized to the proximal tubule and Bowman's

capsule. Furthermore, the epithelial membrane antigen also hasbeen identified in a wide variety of other normal and neoplastictissues (20, 29). The ABO (H) blood group antigens are expressed in the collecting ducts and urothelium and have notbeen shown to be expressed by any of 30 renal cell carcinomasexamined (30). Thus, by its molecular weight and pattern oflocalization within human tissues, the gp200 antigen appearsto represent a unique glycoprotein as determined here.

Although directed against a distinct glycoprotein, the antibody parallels other murine monoclonal antibodies directedagainst various defined and undefined kidney antigens locatedwithin the proximal segments of the nephron. Two independently reported monoclonal antibodies, MKi-1 and F23 (URO3),both recognize a M, 140,000 antigen localized to the proximalrenal tubule (4, 9). The MKi-1 antigen is also expressed inactivated lymphocytes, breast, and pancreatic epithelium, whilethe F23 defined antigen is otherwise expressed in fibroblasts,cartilage, bile ducts, and canaliculi. The S23 and S27 (URO4)monoclonal antibodies recognize the M, 120,000 glycoprotein,adenosine deaminase binding protein, which is expressed by theproximal tubule and loop of Henle (9, 31). Again there isvariable expression in other tissues including the prostate,placenta! trophoblasts, breast (S23), lung (S23), and colon(S23). Our antibody also demonstrates characteristics similarto the A6H and D5D monoclonal antibodies reported by Ves-

sella et al. (2) and the RC3 and RC69 antibodies described byOosterwijk et al. (3). These latter antibodies recognize as yetundefined renal antigens localized within the proximal tubuleand/or Bowman's capsule, but absent in all other normal tissues

examined.Immunohistochemical analysis demonstrated gp200 antigen

expression in 93% of primary and 84% of metastatic renal cellcarcinomas. The primary tumors negative for the gp200 antigenmay be histogenetically derived from, or differentiate along thelines of, more distal segments of the nephron, as has beenpostulated by others who have noted a similar percentage ofrenal cell carcinomas lacking differentiation antigens of proximal tubules (22). As noted with other renal antigens (3, 9, 32),not only were there considerable quantitative differences inantigen expression by different renal cell carcinomas but alsoprominent heterogeneity of antigen expression among subpop-

ulations of tumor cells in any particular renal cell carcinoma.Such heterogeneity in the expression of cell surface antigenshas been shown to be the result of cell cycle-dependent antigenexpression (33). One is tempted to postulate that such quantitative and qualitative phenotypic differences in gp200 antigenexpression may reflect upon the biological behavior of thetumor (i.e., independent of clinical stage and nuclear grade). Aretrospective clinicopathological study of a large number ofprimary renal cell carcinomas, of similar stage and grade, mayhelp determine whether gp200 antigen expression might represent an independent prognostic parameter.

In 9 cases where both the primary tumor and correspondingmÃ©tastaseswere available for analysis, we noted no significantdifference in the degree of gp200 antigen expression between 6primaries and their mÃ©tastaseswhen both presented synchronously. These findings are similar to those of others who havenoted no differences in phenotypic expression of renal antigensbetween primary and metastatic renal cell carcinomas from thesame patient (2, 9). However, in 3 of our cases of subsequentmetachronous mÃ©tastasesof 9 mo or more duration, thereappeared to be a semiquantitative loss (percentage of tumorcells positive) of antigenic expression by the metastatic tumor.The number of cases is small, but one might postulate protracted selective forces in the host producing an in vivo analogueof antigenic modulation/loss of antigenic expression, as seen invitro with sequential passages of tumor cells in cell culture (33).The effect of any interval treatment also cannot be excluded. Inany event, it would be of interest to retrospectively study alarger number of such cases of primary and metachronousmetastatic renal cell carcinomas to determine whether this

1807




temporal parameter may affect the phenotypic expression ofthis cell surface renal antigen.

Specificity analysis demonstrated that the gp200 antigen isrelatively restricted in its distribution in other neoplasms. It isalso expressed by mammary (ductal and lobular) carcinomasand parathyroid adenomas, corollary neoplasms of the normaltissues expressing the gp200 antigen. Evidence that we are notdealing with a carbohydrate determinant shared by distinctmolecules was provided by Western immunoblot analysis ofmammary carcinoma cells, which showed PN-15 reactivity witha similar Mr 200,000 component. That 2 of 4 teratocarcinomaswould focally react in the embryonal carcinomatous componentshould probably not be surprising. This might best be explainedby the fact that the tumor cell of an embryonal carcinoma is apluripotential cell that has the capacity to differentiate into"mature" components of all three germ lines (34). In compari

son, the S4 (URO2), F23 (URO3), S23, and S27 (URO4)monoclonal antibodies react with, depending on the particularantibody, some transitional cell carcinomas, colonie adenomas,epidermoid carcinomas of lung, prostatic adenocarcinomas,fibrosarcomas, chondrosarcomas, osteogenic sarcomas, me-sotheliomas, astrocytomas, melanomas, and leukemia/lympho-mas (9). The A6H antibody also stained colonie adenocarcinomas, a few transitional cell, testicular, and mammary carcinomas, while the D5D antibody demonstrated restricted reactivityto renal cell carcinomas only (2). The RC3, RC69, and RC38series of monoclonal antibodies also showed restricted reactivityto renal cell carcinomas (3).

While the sensitivity of reactivity with renal cell carcinomasand the degree of restricted tissue localization of previouslyreported antibodies vary, virtually all the studies were carriedout on frozen tissue sections and/or on tumor cell lines. Vessellaet al. (2) state simply that their A6H and D5D monoclonalantibodies react with paraffin-embedded tissues, although "thedegree of reactivity was usually weaker." Besides our study, two

others have analyzed antibodies directed against renal cell carcinomas using paraffin-embedded tissue sections (6, 11). Onestudy using an antibody to the endogenous feline retrovirusRDI 14p30 was mentioned above (11). However, as only 5 othertypes of neoplasms, besides trophoblastic tumors, were examined, further specificity analysis would seem indicated. Thesecond study by Kochevar described a monoclonal antibody(5F4) to an undefined antigen expressed in paraffin-embeddedtissue sections of 100% of renal cell carcinomas (6). Cyto-plasmic staining of some adjacent normal renal tubular epithelium (apparent distal tubules as illustrations appear to indicate),prostatic glandular acini, sweat gland epithelium, and thyroidfollicular epithelium was also noted. The 5F4 antibody demonstrates significantly broader reactivity with other neoplasmsthan the HPS-1 antibody or our antibody. It reacts with prostatic adenocarcinomas, adenocarcinoma of lung, cystadenocar-cinoma of ovary, duct carcinoma of breast, melanomas, meso-thelioma, nephroblastoma, and an osteosarcoma. Thus, fromour analysis, the PN-15 monoclonal antibody appears to holdpromise as a diagnostic immunohistochemical reagent thatwould help retrospectively define tumors of proximal nephro-genic differentiation. As the kidney often enters into the differential of possible occult primary sites in metastatic tumors,positive immunohistochemical reactivity for the gp200 antigenwould lend strong support to a kidney primary. A breast primary or a metastatic embryonal carcinoma can usually beexcluded morphologically, clinically, and/or immunohisto-chemically (e.g., a-fetoprotein and Â«i-antitrypsin-positive embryonal carcinomas). Negativity for the gp200 antigen in a

metastatic tumor would argue against a kidney primary, butsuch evidence must be taken in light of the 84% sensitivity ofgp200 antigen expression in metastatic renal cell carcinomasand the considerable degree of heterogeneic expression of thiscell surface antigen.

ACKNOWLEDGMENTS

We wish to express our gratitude to Myron Melamed, M.D., Department of Pathology, Memorial SIoan-Kettering Cancer Center, andWarren Heston, Ph.D., Department of Urology, Memorial Sloan-Kettering Cancer Center, for generously providing us with tissue sections of metastatic renal cell carcinomas and the renal carcinoma cellline, respectively.

REFERENCES

1. Ueda, R., Ogata, S., Morrissey, D. M., Finstad, C. L., Szkudlarek, J.,Whitmore, W. F., Oettgen, H. F., Lloyd, K. O., and Old, L. J. Cell surfaceantigens of human renal cancer defined by mouse monoclonal antibodies:identification of tissue-specific kidney glycoproteins. Proc. Nati. Acad. Sci.USA, 78: 5122-5126, 1981.

2. Vessella, R. L., Moon, T. D., Chiou, R. K., Nowak, J. A., Arfrnan. E. W.,Palme, D. F., Peterson, G. A., and Lange, P. H. Monoclonal antibodies tohuman renal cell carcinoma: recognition of shared and restricted tissueantigens. Cancer Res., 45:6131-6139, 1985.

3. Oosterwijk, E., Ruiter, D. J., Wakka, J. C., Huiskens, J. W., Meij, V. D.,Jonas, L'.. Fleuren, G. J., Zwartendijk, J., Hoedemaeker, P. J., and Warnaar,

S. O. Immunohistochemical analysis of monoclonal antibodies to renalantigens: application in the diagnosis of renal cell carcinoma. Am. J. Pathol.,123: 301-309, 1986.

4. Tagliabue, E., Canevari, S., Menard, S., Fossati, G., BalsarÃ,A., Della Torre,G., and Colnaghi, M. I. Human renal antigen defined by a murine monoclonalantibody. J. Nati. Cancer Inst., 73: 363-369, 1984.

5. Kinouchi, T., Nakayama, E., Ueda, R., Ishiguro, S., Uenaka, A., Oda, H.,and Kotake, T. Characterization of a kidney antigen defined by a mousemonoclonal antibody K 2.7. J. Urol., Â¡37:151-154, 1987.

6. Kochevar, J. A renal cell carcinoma neoplastic antigen detectable by immu-nohistochemistry is defined by a murine monoclonal antibody. Cancer(Phila.), 60: 2031-2036, 1987.

7. Oosterwijk, J., Ruiter, D. J., Hoedemaeker, P. J., Pauwels, E. K. J., Jonas,U.. Zwartendijk, J., and Warnaar, S. O. Monoclonal antibody G250 recognizes a determinant present in renal cell carcinoma and absent from normalkidney. Int. J. Cancer, 38:489-494, 1986.

8. Luner, S. J., Chose, T., Chatterjee, S., Cruz, H. N., and Belitsky, P.Monoclonal antibodies in kidney and tumor-associated surface antigens ofhuman renal cell carcinoma. Cancer Res., 46: 5816-5820, 1986.

9. Finstad, C. L., Cardon-Cardo, C., Bander, N. H., Whitmore, W. F., Melamed,M. R., and Old, L. J. Specificity analysis of mouse monoclonal antibodiesdefining cell surface antigens of human renal cancer. Proc. Nati. Acad. Sci.USA, 82: 2955-2959, 1985.

10. Scharfe, T., Bechi, E., Kaltwasser, R., Thuroff, J. W., Jacobi, G. H., andHohenfellner, R. Tumor-specific monoclonal antibodies for renal cell carcinoma. Eur. Urol., 11: 117-120, 1985.

11. Wahlstrom, T., Suni, J., Nieminen, P., Narvanen, A., Lehtonen. T., andVaheri, A. Renal cell adenocarcinoma and retrovirus p30-rclatcd antigenexcreted to urine. Lab. Invest., 53: 464-469, 1985.

12. Wahlstrom, T., Narvanen, A., Suni, J., Pakkanen, R., Lehtonen, T., Saksela,E., Vaheri, A., Copeland, T., Cohen, M., and Oroszlan, S. M, 75,000 protein,a tumor marker in renal adenocarcinoma, reacting with antibodies to asynthetic peptide based on a cloned human endogenous retroviral nucleotidesequence. Int. J. Cancer, 36: 379-382, 1985.

13. Nairn, R. C., Richmond, H. G., and McEntegart, M. G. Immunologicaldifferences between normal and malignant cells. Br. Med. J., 2: 1335-1340,1960.

14. Kohler, G., and Milstein, C. Continuous culture of fused cells secretingantibody of pre-defined specificity. Nature (Lond.), 256:495-497, 1975.

15. Imam, A., Drushella, M. M., Taylor, C. R., and Tokes, Z. A. Preferentialexpression of a M, 155,000 milk-fat-globule membrane glycoprotein onluminal epithelium of lobules in human breast. Cancer Res., 46:6374-6379,1986.

16. Ueda, R., Shiku, H., Pfreundshuh, M., Takahashi, T., Li, L. T. C., Whitmore,W. F., Oettgen, H. F., and Old, L. J. Cell surface antigens of human renalcancer defined by autologous typing. J. Exp. Med., 750:564-579, 1979.

17. Towbin, H., Slaehelin, T., and Gordon, J. Electrophoretic transfer of proteinsfrom polyacrylamide gel to nitrocellulose sheets: procedure and some applications. Proc. Nati. Acad. Sci. USA, 76:4350-4354, 1979.

18. Hsu, S.-M.. Raine, L., and Fanger, H. Use of avidin-biotin-peroxidasecomplex (ABC) in immunoperoxidase techniques: a comparison betweenABC and unlabeled antibody (PAP) procedures. J. Histochem. Cytochem.,29:577-580,1981.

19. Imam, A., Laurence, D. J. R., and Neville, A. M. Isolation and characteriza-

1808




tion of a major glycoprotein from milk-fat-globule membranes of humanbreast milk. Biochem. J., 193:47-54, 1981.

20. Heyderman, E., Steele, K., and Ormerod, M. G. A new antigen on theepithelial membrane: its immunoperoxidase localization in normal and neo-plastic tissues. J. Clin. Pathol., 32: 35-39, 1979.

21. Wallace, A. C, and Nairn, R. C. Renal tubular antigens in kidney tumors.Cancer (Phila.), 29: 977-981, 1972.

22. Bander, N. H., Cardon-Cardo, C., Finstad, C. L., Whitmore, W. F., Oettgen,H. F., Melamed, M. R., and Old, L. J. Immunopathology of renal cancer:identification of antigenically and clinically distinct subtypes with monoclonal antibodies. Proc. Am. Assoc. Cancer Res., 25: 253, 1984.

23. Suni, J., Narvanen, A., Wahlstrom, T., Lehtovirta, P., and Vaheri, A.Monoclonal antibody to human T-cell leukemia virus p 19 defines polypeptideantigen in human choriocarcinoma cells and syncytiotrophoblasts of tirsitrimester placentas. Int. J. Cancer, 33: 293-298, 1984.

24. Haynes, B. F., Robert-Guroff, M., Metzger, R. S., Franchimi, G., Kalyanar-aman, V. S., Palker, T. J., and Gallo, R. C. Monoclonal antibody againsthuman T-cell leukemia virus pi9 defines a human thymic epithelial antigenacquired during ontogeny. J. Exp. Med., 157:907-920, 1983.

25. Ernst, C., Atkinson, B., Wysocka, M., Blaszczyk, M., Herlyn, M., Sears, H.,Steplewski, /.. and Kaprowski, H. Monoclonal antibody localization of Lewisantigens in fixed tissue. Lab. Invest., 50: 394-400, 1984.

26. Combs, S. G., Marder, R. J., Minna, J. D., Mulshine, J. L., Polovina, M.R., and Rosen, S. T. Immunohistochemical localization of the immunodom-inant differentiation antigen lacto-A'-fucopentaose III in normal adult andfetal tissues. J. Histochem. Cytochem., 32:982-988, 1984.

27. Fukushi, Y., Hakomori, S., and Shepard, T. Localization and alteration of

mono-, di-, and trifucosyl alâ€”Â»3type 2 chain structures during humanembryogenesis and in human cancer. J. Exp. Med., 759: 506-520, 1984.

28. Fleming, S., and Matthews, T. J. Renal tubular antigens in regenerativeepithelium and renal carcinoma. Br. J. Urol., 60: 103-109, 1987.

29. Pinkus, G. S., and Kurtin, P. J. Epithelial membrane antibodyâ€”a diagnosticdiscriminant in surgical pathology: Â¡mmunohistochemical profile in epithelial, mesenchymal, and hematopoietic neoplasms using paraffin sections andmonoclonal antibodies. Hum. Pathol., 16:929-940, 1985.

30. Ghazizadeh, M., Kagawa, S., and Kurokawa, K. Immunohistochemical studies of human renal cell carcinomas for ABO(H) blood group antigens, Tantigen-like substance, and carcinoembryonic antigen. J. Urol., 133: 762-766, 1985.

31. Andy, R. J., Finstad, C. L., Old, L. J., Lloyd, K. O., and Kornfield, R. Theantigen identified by a mouse monoclonal antibody raised against humanrenal cancer cells is the adenosine deaminase binding protein. J. Biol. Chem.,259: 12844-12849, 1984.

32. Yoshida, S. O., Imam, A., Olson, C. A., and Taylor, C. R. Proximal renaltubular surface membrane antigens identified in primary and metastatic renalcell carcinomas. Arch. Pathol. Lab. Med., 110:825-832, 1986.

33. Greiner, J. W., Horan-Hand, P., Colcher, D., Weeks, M., Thor, A., Noguchi,P., Pestka, S., and Schlom, J. Modulation of human tumor antigen expression. J. Lab. Clin. Med., 109: 244-261, 1987.

34. Pierce, G. B., Dixon, F. J., and Verney, E. L. Teratocarcinogenic and tissue-forming potentials of the cell types comprising neoplastic embryoid bodies.Lab. Invest., 9: 583-602, 1960.

35. Fuhrman, S. A., Lasky, L. C., and Limas, C. Prognostic significance ofmorphologic parameters in renal cell carcinoma. Am. J. Surg. Pathol., 6:655-663, 1982.

1809



1989;49:1802-1809. Cancer Res Samuel O. Yoshida and Ashraf Imam Paraffin-embedded Human Renal Cell CarcinomasImmunohistochemical Analysis of Formalin-fixed, Monoclonal Antibody to a Proximal Nephrogenic Renal Antigen:

Updated version

http://cancerres.aacrjournals.org/content/49/7/1802

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/49/7/1802To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Monoclonal Antibody to a Proximal Nephrogenic Renal ...cancerres.aacrjournals.org/content/canres/49/7/1802.full.pdf · Monoclonal Antibody to a Proximal Nephrogenic Renal Antigen: