monoclonal antibodies in analysis of oncoplacental...

[CANCER RESEARCH 42. 2028-2033, May 1982]0008-5472/82/0042-OOOOS02.00

Monoclonal Antibodies in Analysis of Oncoplacental Protein SPi in Vivoand in Vitro1

Eva Engvall,2 Masato MiyashJta,3 and Erkki Ruoslahti4

Cancer Research Center, La Jo/la Cancer Research Foundation. La Jolla, California 9203 7

ABSTRACT

Three monoclonal antibodies were developed to the placenta-specific glycoprotein SPi. The antibodies were used in

the characterization of SPi in placental tissue, pregnancy serum and urine, cancer serum, and cell culture. The threeantibodies reacted similarly with purified SPi of placental originin radio- and enzyme Â¡mmunoassay, and all three decoratedthe syncytiotrophoblast layer of placental villi in immunoperox-

idase staining applied to fixed placental tissue. However, thethree antibodies were found to have different and unique specificities and affinities. One antibody of relatively low affinity wasused to isolate SPi from placenta and fibroblast culture medium. The other two antibodies were used to develop a newand simple immunoassay for SP,. A panel of samples includingcancer sera, pregnancy sera and urine, as well as cell culturefluids gave similar results in the monoclonal assay as in aconventional radioimmunoassay. Our work with monoclonalantibodies to SPi provides further evidence for the productionof authentic SPi by fibroblastic cells in vitro and offers improvedreagents for the isolation, characterization, and quantitation ofan important marker of cancer and fetal development.

INTRODUCTION

The placental cells synthesize a number of proteins thatunder normal circumstances are unique to this organ but maybecome reexpressed in various types of cancer (for review,see Refs. 4 to 6 and 14). The selectiveness of their expressionhas made several of the placental proteins important markersof placental function as well as of cancer. The placenta-specific

protein SPi is one of these proteins. Originally described byTatarinov and Masyukevich (25) and by Bohn (2), this proteinhas been characterized as a glycoprotein with a molecularweight of about 100,000 (3). The concentration of SP, innormal serum is below the detection limit of radioimmunoassay(<0.5 ng/ml) suggesting that there is very little if any synthesisof SPi by normal nonplacental cells in vivo (10). The SPi levelsfirst become elevated during the first 2 to 3 weeks of pregnancy, making SPi a possible alternate to human chorionicgonadotropin as a marker for the detection of early pregnancy(12). Later in pregnancy, the SPi level in maternal serumbecomes as high as 0.3 mg/ml (2, 3). The function of theprotein is unknown.

SPi is expressed by trophoblastic tumors, and its measurement in serum provides a marker found useful in monitoring of

' This work was supported in part by Grant P30 CA 30199 from the National

Cancer Institute. Department of Health and Human Services.2 Recipient of National Cancer Institute Grant CA 27464.3 Present address: Katata Research Center, Toyobo Company, Otsu, Shiga,

Japan.4 Recipient of National Cancer Institute Grant CA 27460.

Received October 19, 1981; accepted January 28, 1982.

the treatment of choriocarcinoma (22, 23). The same is true ofgerm cell tumors (18). Whether tumors derived from tissuesother than the trophoblast and testis express SP( is controversial. Elevated serum levels of SPi have been found by somelaboratories in a variety of cancers (26, 28), while others havenot observed such elevations (10, 18). These discrepanciesmay at least partly depend on differences in the immunologicalreagents and techniques used for the detection of SP,.

Production of significant amounts of immunoreactive SPi hasbeen demonstrated in cultures of a variety of normal andcancerous human cell types (1, 19, 20). Fibroblastic cell linesderived from normal adult skin appear to be consistent SPiproducers (19),5 while production of SPi by cell lines estab

lished from human tumors is less common (20).It seemed to us that improved immunological reagents and

techniques for the determination of SPi could be useful inevaluation of production of SPi by tumors in vivo and in thecharacterization of SPi-like material produced in vitro. To ob

tain such reagents, we set out to prepare monoclonal antibodies to SPi. We report here on the development of suchantibodies to SPi and of their reactivities with SPi from differentsources.

MATERIALS AND METHODS

Antigens and Antisera. SP, was purified from placental extract asdescribed (9). Antisera to SP, were raised in rabbits by monthlyinjections of 100 to 300 p.gpurified SP, emulsified in Freund's complete

adjuvant. The antisera were absorbed with male plasma proteins coupled to Sepharose. Two individual rabbit antisera were collected after5 (Antiserum 6364) and 11 (Antiserum 103) months of immunization,respectively.

Hybridomas. A female BALB/cDub mouse (Flow Laboratories, Dublin, Va.) was immunized with 100 fig of SPi emulsified in Freund's

complete adjuvant and boosted 3 weeks later with 100 /ig of SPi inPBS6 given i.p. for each of 3 consecutive days (24). On the fourth day,

the spleen was removed, and the lymphocytes were used for fusionwith the mouse myeloma cell line X63-Ag.8.563 (15). The methods

used for fusion, culture, cloning, and propagation as ascitic tumorshave been described (27). The cultures were screened for antibodyproduction using the enzyme immunoassay, ELISA (8). Antibodies werepurified from hybridoma cultures and ascites using chromatography onprotein A-Sepharose (27).

Immunoperoxidase Staining. Term placentas were obtained within24 hr of normal deliveries. Pieces of tissue were fixed in acetone,dehydrated in ethanol, embedded in paraffin, and sectioned. Theparaffin was removed, and rehydrated sections were incubated withhybridoma culture medium diluted 1:2 with PBS for 1 hr at roomtemperature. After a washing, peroxidase-labeled rabbit anti-mouse

IgG was incubated with the sections for 1 hr longer. The sections werewashed again and incubated with 0.05% 3,3'-diaminobenzidine tetra-

5 Our unpublished observations.6 The abbreviations used are: ELISA, enzyme-linked immunosorbent assay;

PBS, phosphate-buffered saline, 0.14 M NaCI, 0.015 M phosphate, pH 7.2.

2028 CANCER RESEARCH VOL. 42

on May 14, 2018. © 1982 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

,â€žT

hydrochloride (Litton Bionetics, Kensington, Md.) in 0.01 % H2C>2in 0.1M citrate-phosphate buffer, pH 5.0, for 10 min. The sections were

washed in H2O and mounted in glycerol.Labeling of Antigen and Antibodies. SP, was labeled with 125Iusing

chloramine-T to a specific activity of about 50 /jCi//ig. Antibodies were

labeled with either horseradish peroxidase (Toyobo Co., Osaka, Japan)or calf intestinal alkaline phosphatase (Sigma type VII-S; Sigma Chem

ical Co., St. Louis, Mo.). Procedures for enzyme labeling and measurement of enzyme activities have been described in detail (8).

Radioimmunoassay and Determination of Antibody Affinity. Quan-titation of SPi was done by competitive double antibody techniqueessentially as described (10) except that 10% bovine serum in PBSwas used as diluent. For measurement of antibody affinity, the assayswere carried out at an antibody concentration which bound 50% of theimmunoreactive tracer in the absence of inhibitor. The antibody, tracer,and various amounts of unlabeled SP, were incubated for 20 hr atroom temperature; an equal volume of 25% polyethylene glycol (M,4000; Sigma) was added; and 10 min later, the samples were centrifugea, and the precipitates were counted. Relative affinities wereestimated from the inhibition curves, taking the reciprocal of the concentration of SP, required for half-maximal inhibition as the relative

affinity constant (17).Enzyme Immunoassay of SPi. Nontreated polystyrene microtiter

plates were coated with purified monoclonal anti-SP, (1 /ig/ml) orpolyclonal anti-SP, (3 fig/ml). Serial dilutions of purified SPi or samples

containing SP, were prepared in PBS containing 5% normal bovineserum and 0.05% Tween 20. The samples were incubated with thesolid phase for 3 to 16 hr at room temperature. After washing, enzyme-labeled anti-SP,, diluted in PBS containing serum and Tween, was

added and incubated for 2 to 8 hr. After a final washing, the enzymebound to the solid phase was measured.

Culture of Human Fibroblasts. Human normal skin fibroblast cultures were obtained from skin biopsies of healthy laboratory personnel.The cells were routinely cultured in Dulbecco's modification of Eagle's

medium supplemented with 10% fetal bovine serum.

RESULTS

Production of Monoclonal Antibodies to SP,. Three cultures with anti-SP, activity were found among the 10 wells thatshowed growth in three 96-well plates. The positive hybrido-

mas, designated IC12, IID10, and IIID7, were cloned andpropagated in mice. All 3 produced antibodies of the lgG1subclass as tested by immunodiffusion of concentrated hybrid-

oma culture fluid against specific antisera.Binding of SPi by Monoclonal Antibodies and Antibody

Affinity. Radioimmunoassay binding curves obtained with themonoclonal mouse anti-SP, and polyclonal rabbit anti-SP,

showed that the monoclonal antibodies reacted with a smallerproportion of 125I-SP, than did the polyclonal rabbit antibodies

(Chart 1). The amount of each individual antibody preparationcapable of binding 50% of added immunoreactive labeled SP,was used to generate the inhibition curves shown in Chart 2.The relative affinity constants calculated from these assaysshowed that the IID10 antibody had the highest affinity of the3 monoclonal antibodies (Table 1). Its affinity was approachingthose of the hyperimmune rabbit antisera, while the 2 otherantibodies had lower affinities.

Monoclonal Antibodies in Immunohistology. All 3 monoclonal antibodies gave strong immunoperoxidase staining ofthe syncytiotrophoblast layer of placental villi. Fig. 1 shows thisfor 2 of the antibodies and shows that the staining patternswere identical to that seen with polyclonal rabbit antiserum. Nostaining was seen with a monoclonal mouse IgG, with a hybrid-

Monoclonal Antibodies in Analysis of Oncoplacental Protein SP,

100r

T3

OmÃ›L"

co50

io- 10" 10- 10'3

Dilution of AntibodyChart 1. Titration of antibody in radioimmunoassay. â€¢,rabbit 6364, serum;

O, rabbit 103, serum; A, IID10, ascites; O, IC12. ascites; A, IIID7, ascites.

100

O)

p05

50

Xco

10 100 1000 10000

ng/ml

Chart 2. Inhibition of binding of antibody to radiolabeled SP, by purifiedunlabeled SP,. O, rabbit 103; Â».rabbit 6364; A. IID10; D. IC12; A. IIID7.

orna antibody of irrelevant specificity, or with normal rabbitserum.

Specificities of Monoclonal Anti-SPiS. Sandwich ELISA was

used to evaluate whether the 3 monoclonal antibodies to SP,were reacting with the same or different antigenic determinantson the SP, molecule. All 3 mouse antibodies as well as standardrabbit antibody were used both for coating and as enzymeconjugates in a sandwich assay. The antibodies were tested inpairs in the presence or absence of a single dose of SP, (10ng/ml). In the sandwich technique, one can obtain a doseresponse in the presence of the antigen only if the antibody onthe solid phase and the labeled antibody are directed againstantigenic determinants that are different and/or properlyspaced. The results show that the IC12, IID10, and IIID7 allreact with different antigenic determinants since combinationof any 2 of these antibodies gave a dose response to thepresence of SP, (Chart 3). The reason for the relatively poor

MAY 1982 2029



E. Engvall et al.

Table 1Relative affinities of polyclonal rabbit antibodies and monoclonal mouse

antibodies to SP,Antibody fc(M~')

20-

Rabbit 6364Rabbit 103IC12IID10IIID7

3.0 x 10'Â°4.0 X 10'Â°3.9 X 10"5.6 X 10"2.1 X 10"

1.2

1.0

Â£ 0.8Tâ€”~i) 0.6

^ 0.4

0.2

IC12 noio mo?

Solid Phase AntibodyCharts. Two-site ELISA. Plates were coated with IC12, IID10. or IIID7.

Binding of SP to the solid phase was detected using enzyme conjugatesprepared from IC12 (s), IID10(â€¢).IIID7(a), or an IgG fraction of rabbit antiserumto SP, (D).

response to SP, seen in the combinations involving the IIID7antibody is unknown but cannot solely depend on its relativelylow affinity, since IC12 is of similar affinity.

One-Step, 2-Site Assay of SPi. Since the antibodies IC12and IID10 seemed to react with different antigenic determinantson SP, and to give a good response to a low dose of SPi, theywere used to develop a sandwich assay for SP,. IC12 wasused for coating of the solid phase, and IID10 was used forenzyme labeling. Alkaline phosphatase was used as enzymelabel for this assay. Since these antibodies reacted with different antigenic determinants, they did not interfere with eachother's binding to the antigen and could be incubated with the

antigen in a single step instead of the 2 consecutive steps of aconventional sandwich assay. The assay gave a linear doseresponse in the 3 to 100 ng/ml SP, dose range (Chart 4).Concentrations of SP, above 1 /ig/ml gave a negative doseresponse (not shown). This " high-dose hook effect' ' is obtained

when the amount of solid-phase antibody and/or enzyme-

conjugated antibody becomes a limiting factor in the assay.The one-step sandwich ELISA using monoclonal antibodies

was evaluated on a panel of samples representative of theclinical as well as basic research aspects of SP,. The 2-site

assay using the 2 monoclonal antibodies gave results similarto those obtained with a classical competitive, double antibodyradioimmunoassay. The samples tested included sera andurine from pregnant women, sera from cancer patients (invasivemole) and from healthy laboratory personnel, and samples ofmedia from cultured fibroblasts (Table 2). There was a tendency for the 2-site assay to give higher values than were

obtained in the radioimmunoassay. We do not know at presentif this discrepancy is due to some inherent characteristics ofthe monoclonal antibodies used or whether it might depend onthe fact that different assay procedures were used.

cEinCM

1.0

ÃŽ0.5

0.2

Â»256 1128 ,64 ,32dilution oi hbioDiast SP,

62 125 25ng/ml SP,

50 100

Chart 4. Standard dose-response curve for SP, ( ) in the 2-site one-stepELISA using monoclonal antibodies IC12 and 11010. , dilutions of fibroblastSP, concentrated by immunoadsorbance on monoclonal antibody IIID7. Bars,S.D.

Table 2Comparison of SP, levels /Â»various samples as measured by competitive

radioimmunoassay with conventional antiserum and by 2-site, one-step enzymeimmunoassay with monoclonal antibodies IC12 and IID10

SamplePregnancy

serumPregnancy

urineCancer

serumNormal

serumFibroblast

culture mediumSP,(ng/ml)

byRadioimmunoas

sayELISA200.000240,000

150,0002,900

8333270

15022012<3

<3<31620240,000

300,000200,0002,300

7340350

300310

<100<10<10

<1016

28

Reactivity of SP,-like Antigen from Fibroblast Cultures

with Monoclonal Antibodies. It is evident from Table 2 that the2 monoclonal antibodies IC12 and IID10 recognize SP,-like

material in fibroblast culture media in a way quantitativelysimilar to that of the polyclonal radioimmunoassay antibody.That the third monoclonal antibody IIID7 also reacted with thesame product of fibroblasts was shown by immunoadsorbance.The IIID7 antibody was coupled to Sepharose, and the resultingimmunoadsorbent was used to bind the SP,-like material fromfibroblast culture media. The SP,-immunoreactive materialcould be eluted from the IIID7 Â¡mmunoadsorbent by 8 M urea.A total of 4.2 fig of SP, was obtained from 2.3 liters of pooledmedia from cultured skin fibroblasts as determined by radioimmunoassay. This represents a 20% recovery since the mediumwhich initially contained SP, at 10 ng/ml had only 1 ng/mlafter absorption. The peak of eluted fractions was concentrated




Monoclonal Antibodies in Analysis of Oncoplacental Protein SP,

and further tested in the one-step sandwich ELISA (Chart 4)

and in immunodiffusion against precipitating rabbit antiserum(Fig. 2). The fibroblast SP, was Â¡mmunologically indistinguishable from placental SP, in these tests also. In conclusion, 3monoclonal antibodies of different specificities provide evidence that fibroblastic and placental SP, are closely similarmolecules.

DISCUSSION

Monoclonal antibodies have been found to be quite useful indefining new antigenic species in mixtures of macromolecules.They also provide specific probes for deciphering the antigenicstructure of macromolecules. Further applications for monoclonal antibodies include quantitative immunoassay and Â¡m-

munoadsorbent purification of antigens. The latter applicationsput more strict requirements on the isotype, specificity, andaffinity of the antibodies.

The average affinity of antibodies increases upon prolongedimmunization (21). A short immunization schedule such as iscommonly used for production of monoclonal antibodies givesantisera that are of too low an affinity to be useful in sensitiveimmunoassays. However, our results show that it is possible toobtain high-affinity antibodies, suitable for radio- and enzyme

immunoassays, even after a short immunization course. One ofour antibodies, IID10, had a relative affinity constant within 1order of magnitude of that of our conventional radioimmunoas-say antibody, obtained after 5 to 11 months of immunization.The other 2 antibodies, IC12 and IIID7, had affinities on theorder of 108 M~'. While this allowed quantitation of SP, only in

the fig range in competitive radioimmunoassay (Chart 2), theIC12 antibody could be used in combination with the high-

affinity antibody IID10 to construct a sandwich enzyme immunoassay capable of detecting SP, amounts as small as 10 ng/ml. That the same antibody gives a higher sensitivity in ELISAthan in radioimmunoassay probably depends on the higherlocal concentration that the antibody attains at the solid-phase

surface in the ELISA than as a solute in radioimmunoassay.All our monoclonal antibodies bound a lower percentage of

125l-labeled SP, than was bound by conventional anti-SP,. The

reason for this is not known, but it may be due to the fact thatmonoclonal antibodies react with a single determinant and thatany given determinant may be denatured in a certain proportionof molecules during the purification and iodination procedure.The fractional binding of the labeled SP, is probably not due tothe presence of antigenically distinct subfractions or variantsin SP, since more than 90% of SP, from pooled placentalextracts5 and from pooled fibroblast medium bound to these

antibodies.The monoclonal antibodies stained sections of acetone-fixed

and paraffin-embedded term placental tissue giving similar

patterns. Only the syncytiotrophoblast layer was specificallystained by the monoclonal as well as polyclonal antibodies.These results show that our monoclonal antibodies react withdeterminants in SP, that are not destroyed by standard fixationand embedding techniques and that these antibodies mightbecome useful in localization of SP, in tumors and tissues otherthan placenta.

Results obtained using ELISA showed that, in spite of theirsimilar reactivities in radioimmunoassay and immunohistology,all 3 monoclonal antibodies were directed against different

antigenic determinants. The sandwich assay requires bindingof both the solid-phase antibody and the labeled indicator

antibody to the same molecule of SP,. Such binding wasobtained in combinations of any 2 of the 3 antibodies but notif the same antibody was used for both functions. The 3monoclonal antibodies therefore appear to be directed againstdifferent and well-spaced structures on the SP, molecule.

Due to their different specificities, the monoclonal antibodiesdid not interfere with each other's binding to SP, when reacted

with SP, in the same step of a sandwich ELISA. This resultedin a greatly simplified assay. We have previously shown thathuman a-fetoprotein can be assayed based on the same principle (27). Others have independently developed a similarassay for a human melanoma antigen (7). A greater number ofsamples than studied here will have to be tested in order toevaluate a possible clinical usefulness of the assay for SP,determination. Some of the features of this assay make itattractive for routine use, especially the defined Â¡mmunologicalreagents used and the single binding step required.

With a lower limit of detection of 10 ng/ml, the assay basedon our monoclonal antibodies was not sensitive enough tostudy the important question of normal levels of SP,, if any, inhealthy nonpregnant individuals. We are currently developingadditional hybridomas with the hope that monoclonal antibodies with higher affinities would allow us to develop assaysof improved sensitivity.

We used the assay for measurement of the low but significantlevels of SP, immunoreactivity found in cultures of fibroblasticcells. We have previously tested by conventional radioimmunoassay fibroblast cell lines from over 20 different healthy,nonpregnant individuals of both sexes and found that all produce low but easily detectable and stable levels of SP,-likeactivity5. This finding is puzzling, because the lack of detectable

SP, in the circulation of normal nonpregnant individuals suggests that fibroblasts do not synthesize significant amounts ofSP, in vivo. Rosen et al. (19) showed in their initial studydescribing this phenomenon that the immunoreactive materialfrom fibroblasts had a molecular weight corresponding to thatof authentic placental SP,, but further characterization has notbeen done. In this paper, we present results that show that theSP,-like material from fibroblast culture media can be concen

trated and that it has the immunological properties of authenticSP,.

The fibroblast product reacted with our monoclonal antibodies in ELISA as well as in immunochromatography. Wewere able to isolate sufficient quantities of fibroblast SP, for itspartial characterization by immunochromatography on monoclonal antibodies. The Â¡mmunological identity of the isolatedfibroblast SP, with placental SP, in immunodiffusion againstrabbit anti-SP, and the reactivity with 3 individual monoclonal

antibodies each of which recognizes a different antigenic sitein the SP, polypeptide provide strong evidence for the identityof the fibroblast product as authentic SP,.

Since cultured fibroblasts are not necessarily representativeof the fibroblast population of intact adult tissue, it is possiblethat the synthesis of SP, is a characteristic of a mesenchymalstem cell. This hypothesis can be tested with the monoclonalantibodies. The placental staining shown here demonstratesthe suitability of our monoclonal antibodies for detection of SP,in tissues, and the specificity of these antibodies will allowmore definite conclusions on the presence of SP, in tissue than

MAY 1982 2031



E. Engvall et al.

has been possible previously. Similar studies will also be informative regarding the purported presence of SPi in granu-

locytes (5) and tumors such as mammary cancer (13, 28).Finally, it may be possible to use monoclonal antibodies to SP,in immunodetection of tumors in vivo. Promising immunodetec-

tion results have been obtained with conventional antibodies toother tumor-associated proteins such as carcinoembryonicantigen and a-fetoprotein (11,16). The specificity and unlimited

supply of monoclonal antibodies may be an advantage inimmunodetection as well.

ACKNOWLEDGMENTS

We thank Marsha L. Bell for excellent technical assistance and the ToyoboCompany for a generous gift of horseradish peroxidase.

REFERENCES

1. Azer, P. C., Braunstein. G. D., Van de Velde, R. L., Van de Velde, S., Kogan.R., and Engvall, E. Ectopie production of pregnancy-specific /8,-glycoproteinby a nontrophoblastic tumor in vitro. J. Clin. Endocrinol. Metab., 50: 234-239, 1980.

2. Bonn, H. Nachweis and Characterisierung von Schwangerschaftsproteinenin der menschlichen Plazenta, sowie ihre quantitative immunologische Bestimmung Â¡nserum schwangerer Frauen. Arch. Gynakol., 210: 440-457,

1971.3. Bohn, H. Isolierung and Characterisierung des Schwangerschaftsspezi

fischen /3,-glykoproteins. Blut, 24: 292-302, 1972.4. Bohn, H. Placental and pregnancy proteins. In: F.-G. Lehman (ed.), Carcino-

Embryonic Proteins, Vol. 1, pp. 289-299. Amsterdam; Eisevier Publishing

Co.. 1979.5. Bohn, H., Inaba, N., and LÃ¼ben,G. New placental proteins and their potential

diagnostic significance as tumor markers. Oncodev. Biol. Med., 2: 141-153, 1981.

6. Braunstein, G. D., Vaitukaitis, J. L., and Carbone, P. P. Ectopie productionof human chorionic gonadotropin. Ann. Intern. Med., 78: 39-45, 1973.

7. Brown, J. P., Woodbury, R. G., Hart. C., HellstrÃ¶m, I., and HellstrÃ¶m. K. E.Quantitative analysis of melanoma-associated antigen p97 in normal andneoplastic tissue. Proc. Nati. Acad. Sei. U. S. A.. 78: 539-543, 1981.

8. Engvall, E. Enzyme immunoassay: ELISA and EMIT. Methods Enzymol.,70A: 419-439, 1980.

9. Engvall, E. Pregnancy-specific /?,-glycoprotein (3P,). Purification and partialcharacterization. Oncodev. Biol. Med., 1: 113-122, 1980.

10. Engvall, E.. and Yonemoto. R. H. Is SP, (pregnancy-specific /ÃŽ,-glycoprotein)elevated in cancer patients? Int. J. Cancer, 23: 759-761, 1979.

11. Goldenberg, D. M.. Kim. E. E., DeLand. F. H.. Bennet, S., and Primus, F. J.Radioimmunodetection of cancer with radioactive antibodies to carcinoembryonic antigen. Cancer Res., 40: 2984-2992, 1980.

12. Grudzinskas, J. G.. Gordon, Y. B.. Jeffrey, D., and Chard, T. Specific andsensitive determination of pregnancy-specific /3i-glycoprotein by radioim-munoassay. A new pregnancy test. Lancet, /. 333-335, 1977.

13. Home, C. H. W., Reid, I. N., and Milne, G. D. Prognostic significance ofinappropriate production of pregnancy proteins by breast cancers. Lancet,2. 279-282, 1976.

14. Ibsen, K. H., and Fishman, W. H. Developmental gene expression in cancer.Biochim. Biophys. Acta, 560. 243-280, 1979.

15. Kearney, J. F., Radbruch, A., Liesegang, B., and Rajewski, K. A new mousemyeloma cell line that has lost immunoglobulin expression but permits theconstruction of antibody-secreting hybrid cell lines. J. Immunol., 123:1548-1550, 1979.

16. Kim, E. E., DeLand. F. H., Nelson, M. O., Bennet, S., Simmons, G., Alpert,E., and Goldenberg, D. M. Radioimmunodetection of cancer with radioactiveantibodies to a-fetoprotein. Cancer Res., 40: 3008-3012, 1980.

17. MÃ¼ller,R. Calculation of average antibody affinity in anti-hapten sera fromdata obtained by competitive radioimmunoassay. J. Immunol. Methods, 34:345-352, 1980.

18. Rosen, S. W., Javadpour, N., Calvert, I., and Kaminska, J. Increased"pregnancy-specific" betai-globulin in certain nonseminatous germ cell

tumors. J. Nati. Cancer Inst., 62. 1439-1441, 1979.19. Rosen, S. W., Kaminska. J., Calvert, I. S., and Aaronson, S. A. Human

fibroblasts produce "pregnancy-specific" beta-1 glycoprotein in vitro. Am.

J. Obstet. Gynecol., 134: 734-738, 1979.20. Rosen, S. W., Kaminska, J., Calvert, I. S., Ellmore, N., and Aaronson, S. A.

Ectopie production of "pregnancy-specific" beta-1-glycoprotein in vitro:

discordance with three other placental proteins. Am. J. Obstet. Gynecol.,137: 525-529, 1980.

21. Ruoslahti. E. Antigen-antibody interaction, antibody affinity, and dissociationof immune complexes. Scand. J. Immunol.. Suppl. 3, 3-7, 1976.

22. Searle, F., Leake. B. A., Bagshawe, K. D., and Dent. J. Serum-Sp.-pregnancy-specific /?i-glycoprotein in choriocarcinoma and other neoplasticdiseases. Lancet. Ã•.579-581. 1978.

23. SeppÃ¤lÃ¤,M., Rutanen, E.-M., Heikinheimo, M., Jalanko, H., and Engvall, E.Detection of trophoblastic tumor activity by pregnancy-specific beta-1-glycoprotein. Int. J. Cancer. 21: 265-267, 1978.

24. StÃ¤hli, C., Staehelin, T.. Miggiano, V.. Schmidt, J.. and HÃ¤ring, P. Highfrequencies of antigen-specific hybridomas: dependence on immunizationparameters and prediction by spleen cell analysis. J. Immunol. Methods, 32.297-304, 1980.

25. Tatarinov, Y. S., and Masyukevich, V. D. Immunological identification of anew beta-globulin in the blood serum of pregnant women. Byull. Eksp. Biol.Med., 69:66-68, 1970.

26. Tatarinov, Y. S., and Sokolov, A. V. Development of a radioimmunoassay forpregnancy-specific beta,-globulin and its measurement in serum of patientswith trophoblastic and non-trophoblastic tumors. Int. J. Cancer. 19: 161-166, 1977.

27. Dotila. M., Ruoslahti, E., and Engvall. E. Two-site sandwich enzyme immunoassay with monoclonal antibodies to human alpha-fetoprotein. J. Immunol.Methods, 42:11-15, 1981.

28. WÃ¼rz,H. Serum concentrations of SPi (pregnancy-specific /3,-glycoprotein)in healthy, non-pregnant individuals, and in patients with nontrophoblasticmalignant neoplasms. Arch. Gynecol., 227: 1-6, 1979.

Fig. 2. Immunodiffusion of SP,-like material isolated from fibroblasts (a) andpregnancy serum (100 fig SP, per ml; b) against rabbit anti-SP, (ci.




Monoclonal Antibodies in Analysis of Oncoplacental Protein SPi

V\'

v -K â€¢'.X % â€¢-â€¢ v

.

; / â€¢ *â€¢ â€¢ ' -I A. *"' *â€¢ - v \ r' t\, - L

r ;ves&laÃâ€¢â€¢--Ãâ€¢, ,, -.,-

, - ^,

9f1C

t - * r \ ., * ' , ^t â€¢r\ ^.^s.xiiv ^> s â€¢**Â».<A::: â€¢â€¢'(( ,i v" 'K.*/ ,

Fig. 1. Immunoperoxidase staining of placental sections using monoclonal antibodies IID10 (a) and IC12 (Â£>),normal rabbit serum (c). and rabbit anti-SPi Id).

MAY 1982 2033



1982;42:2028-2033. Cancer Res Eva Engvall, Masato Miyashita and Erkki Ruoslahti

in Vitro and in Vivo 1Monoclonal Antibodies in Analysis of Oncoplacental Protein SP

Updated version

http://cancerres.aacrjournals.org/content/42/5/2028

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/42/5/2028To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



monoclonal antibodies in analysis of oncoplacental...

Documents