Cleavage of M#{252}llerian Inhibiting Substance Activates … · Vol. 1. 343-349. March 1995 Clinical Cancer Research 343 Cleavage of M#{252}llerian Inhibiting Substance Activates

Vol. 1. 343-349. March 1995 Clinical Cancer Research 343

Cleavage of M#{252}llerian Inhibiting Substance Activates

Antiproliferative Effects in Vivo1

Marina S. Kurian, Ricardo Sainz de Ia Cuesta,

Gerald L. Waneck, David T. MacLaughlin,2

Thomas F. Manganaro, and Patricia K. Donahoe

The Pediatric Surgical Research Laboratories [M. S. K., R. S. d. I. C.,D. T. M., T. F. M., P. K. D.J, Surgical Immunology Laboratory[G. L. W.J, and the Division of Gynecologic Oncology [R. S. d. 1. C.],Massachusetts General Hospital, and the Departments of Surgery

[M. S. K.. G. L. W., D. T. M., 1. F. M., P. K. D.] and Obstetrics and

Gynecology [R. S. d. I. C.. D. T. M.], Harvard Medical School.

Boston, Massachusetts 02114

ABSTRACT

M#{252}llerian inhibiting substance (MIS), an inhibitor of

growth and development of the female reproductive ducts in

male fetuses, requires precise proteolytic cleavage to yield its

biologically active species. Human plasmin is now used to

cleave and, thereby, activate immunoaffinity-purified re-

combinant human MIS at its monobasic arginine-serine site

at residues 427-428. To avoid the need for exogenous enzy-

matic cleavage and to simplify purification, we created an

arginine-arginine dibasic cleavage site (MIS RR) using site-

directed mutagenesis to change the serine at position 428

(AGC) to an arginine (cGC). The mutant eDNA was thenstably transfected into a MIS-responsive ocular melanoma

cell line, 0M431, followed by cloning for amplified expres-

sion to test its biological activity in vitro and in vivo.

Media from each clone were assayed for production of

MIS RR by a sensitive ELISA for holo-MIS, and high- and

low-producing clones were selected for further study. Media

from the highest MIS RR producer caused M#{252}llerian duct

regression in an organ culture bioassay. Other transfections

were done with an empty vector (pcDNAI Neo) or a con-

struct lacking the leader sequence and thus failing to secrete

MIS, to serve as controls. The 0M431 clones containing the

MIS RR mutant were growth inhibited in monolayer cul-

ture. The high- and low-producing MIS RR 0M431 clones,

along with transfected 0M431 controls, were injected into

the tail veins of immunosuppressed severe combined immu-

nodeficiency mice for in vivo analyses. Four to 6 weeks later,

pulmonary metastases were counted in uniformly inflated

lungs. 0M431 clones containing the more easily cleaved MIS

RR displayed a significant dose-dependent reduction in pul-

monary metastases when compared to the lungs of animals

given injections of 0M431 clones containing empty vector,

leaderless MIS, or wild-type MIS that requires activation by

plasmin cleavage. Since the purification protocol of MIS RR

is less complicated than that for wild-type MIS, which re-quires subsequent enzymatic activation, MIS RR can be

used for scale-up production with increased yields for fur-

ther therapeutic trials against MIS-sensitive tumors.

INTRODUCTION

MIS,3 initially described by Jost (1-3) as the ‘ ‘ M#{252}llerian

inhibitor,’ ‘ is secreted by the fetal testis and, early in male

development, causes regression of the M#{252}llerian ducts, the

precursors for the Fallopian tubes, uterus, and the upper third of

the vagina. MIS is secreted by Sertohi cells of embryonic,

perinatal, and adult testis (4-6) and by the granulosa cells of the

preadolescent and adult ovary (7-9). MIS is reported to play a

role in inhibition of maturation of oocytes (10, 11) and sper-

matogonia4 (12). It may also be influential in pulmonary devel-

opment (13) and has been implicated in testicular descent (14).

In addition to inhibiting normal growth and development, MIS

has antiprohiferative effects on tumors of MUllerian duct origin

(15-18) as well as on 0M431, a human ocular melanoma of

neural crest derivation (19-21). Candidate type I (22) and

type ii5 (23, 24) MIS receptors recently have been cloned, but

convincing saturable binding of MIS to these cloned products

has yet to be demonstrated. Although the candidate type I

receptor is more widespread and may be shared among other

members of the TGF-�3 family (25, 26), the type II receptor is

highly expressed in the testis and ovary5 (23, 24).

MIS is a Mr 140,000 glycosylated disulfide-linked ho-

modimer (27, 28) which has certain similarities to other mem-

bers of the TGF43 superfamily (29). Proteolytic cleavage is

required to generate the biologically active carboxy-terminal

fragment (30) which is required for MOllerian duct regression

and which shares homology with processed TGF-�3. In vito,

cleavage occurs at a monobasic Arg(427)-Ser(428) site (30, 31)

by an unidentified endogenous enzyme. Unlike MIS, TGF-�3

and other members of the superfamily contain a tetrabasic

cleavage site that is more efficiently processed (32). MIS is

produced in our laboratory predominantly in a noncleaved form

from transfected amplified DHFR-deficient, CHO cells (29) in

which monobasic cleavage is inefficient and variable. After

immunoaffinity purification of the secreted recombinant human

Received 8/24/94; accepted I 1/8/94.

I This work was partially funded by National Cancer Institute Grant

CA17393 (P. K. D., D. T. M., T. F. M.), Grant HD 30812 (P. K. D.),

American Cancer Society Postdoctoral Fellowship ACS PRTF 158

(M. S. K.), NIH Grant GM 46467 (G. L. W.), and MassachusettsGeneral Hospital Gynecologic Oncology Fellowship (R. S. C.), and

tissue culture facilities were provided by the Cell Culture Core of the

Reproductive Sciences Center, NICHD P30HD28138 (P. K. D.).2 To whom requests for reprints should be addressed.

3 The abbreviations used are: MIS, MUllerian inhibiting substance;

TGF-�3, transforming growth factor �3; CHO, Chinese hamster ovary;a-MEM+, ca-MEM with ribonucleosides; SCID, severe combined im-munodeficiency; DHFR, dihydrofolate reductase.

4 A. Bellve and P. K. Donahoe, unpublished data.5 P. K. Donahoe et al. , unpublished data.

on April 5, 2020. © 1995 American Association for Cancer Research.clincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

344 Cleavage of MIS Activates Antiproliferative Effects

MIS, plasmin cleavage followed by molecular sieve chromatog-

raphy in acid has been used to isolate the purified active carboxy

terminus (30, 31). Alternatively, plasmin cleavage of holo-MIS

without subsequent chromatography is used to prepare samples

in which amino- and carboxy-terminal fragments remain in

noncovalent association (33).

MIS, as a mediator of regression of a fetal organ system

and as an inducer of programmed cell death (34, 35), is a paradigm

for mechanistic studies and for development as a tumor sup-

pressor chemotherapeutic agent. In the selection of an effective

therapeutic agent, specificity for the target tumor, the appropri-

ate stage of the cell cycle, molecular stability, efficacy, toxicity,

route of delivery, the state of the patient, and drug/drug com-

patibility are all factors that must be considered. Potentially, one

may use as the therapeutic agent, the activated ligand, its recep-

tor, a downstream factor in its pathway of activation, a product

activated by the inhibitor, or an activating or inhibiting tran-

scription factor. In the case of MIS, the higand has been cloned

for some time. However, its therapeutic efficacy was variable,

depending on the serendipitous degree of activation of holo-MIS

at its monobasic cleavage site in the engineered Chinese hamster

cell of production. Therefore, to explore alternatives of produc-

ing a more efficacious and practical therapeutic agent, we ex-

plored methods of more reliably and reproducibly cleaving MIS,

created the appropriate cleavage constructs, and tested them for

their ability to produce MIS capable of inhibiting tumor growth.

Site-directed mutagenesis using the Kunkel method of M13

mutagenesis (36) was used to convert the monobasic cleavage

site of MIS to a dibasic (RR) cleavage site and to create, as a

control, a leaderless MIS protein that should be translated, but

not secreted by the cell. The MIS RR and leaderless MIS mutant

cDNAs and empty vector were transfected into MIS-responsive

ocular melanoma cells, OM431. Biological activity of the mod-

ified MIS protein secreted by transfected OM431 cells was

assessed by MUllerian duct regression in organ culture (37),

using concentrated culture media and immunoaffinity-purified

MIS RR. Once biological activity was confirmed, the antipro-

liferative efficacy was studied in cells transfected with MIS RR

in constructs in in vitro growth and in vivo metastases assays.

MATERIALS AND METHODSModification of MIS. Using the Kunkel method of site-

directed mutagenesis (36) human MIS eDNA was excised from

the pDl plasmid (38) and subcloned into pcDNAI Amp (In-

vitrogen) using HindIII and Nod restriction endonuclease sites

in the polylinker. The MIS eDNA was then subcloned into the

M13mp19 replicative form using HindIlI and XbaI sites in the

polylinker. A deoxyuracil-containing single-stranded template

was generated by transformation of CJ236 (dut-, ung-) with

M13mp19/hMIS and purification of the phage DNA. Mutagenic

ohigonucleotides were designed with unique restriction enzyme

sites to generate (a) a leaderless MIS that would be translated

but not secreted and (b) a dibasic RR cleavage site (see below);

the mutagenic oligonucleotides were then annealed to the de-

oxyuracil-containing template. Second-strand synthesis was

performed and the double-stranded product transformed into

XL1 Blue Escherichia coli that results in degradation of the

wild-type deoxyuracil-containing strand and replication of the

mutagenic strand. Colonies were then selected and mini-prep

replicative form DNA tested by restriction enzyme digest for

presence of the mutation. The mutant hMIS cDNAs were then

subcloned into PUC18 using the HindIII/XbaI sites, and the

presence of the mutations was further confirmed by sequencing.

The mutagenic oligonucleotides (lower case) in the antisense

orientation are as follows:

Leaderless MIS:

Kozak Box HindIII

5’-TGGCFCCTCFGCI’CTcAtGGtggCAGTCCCaAGCttAGCCC

CCA000CAG-3’

MIS RR:

Mg ApaLI

5’-TGGCCCCCGCGCgGCGCTGTGCaCGACCCGGCCCG

C-3’

Cell Lines. Several different human ocular melanoma

cell lines had been established (39); when tested for growth

inhibitory response, OM431 was best inhibited by MIS (21).

OM431 cells were grown in the a modification of MEM with

ribonucleosides (a-MEM+) and 5% MIS-free female FCS. The

wild-type human MIS cDNA, the MIS RR mutant, and the

leaderless MIS mutant were subcloned into pcDNAI Neo (In-

vitrogen) using the HindIII and Not! sites in the polylinker. The

completed constructs as well as the empty vector control were

stably transfected into 0M431 using the calcium phosphate

transfection method. Clonal populations were isolated under

0418 (Geneticin) selection.

MIS Analysis. A MIS ELISA (40) was used to detect the

comparative secretion of MIS by wild-type MIS, MIS RR, or

leaderless MIS clones. All clones were grown in a-MEM+ with

5% female FCS and 0.5 mg/ml Geneticin (Sigma). Samples of

culture media were collected for ELISA analysis either after the

clones had been in culture from 4 to 7 days and reached

confluency, or every day in experiments that assessed the mono-

layer growth of the clones. There was no detectable MIS in the

culture media obtained from either the leaderless MIS/0M431

clones, except when excess cell lysis occurred and low levels of

MIS could be detected (5 ng/ml), or the empty vector-trans-

fected 0M431 clones, which were also assayed by Southern blot

for presence of the neomycin-resistance gene. High- and low-

MIS RR-producing clones were identified by the MIS ELISA

based on protein levels per million cells at confluency. Protein

gel electrophoresis of immunoaffinity-purified MIS RR and

wild-type MIS was performed on a 15% acrylamide gel at 30

mA for 1 h. The gel was then stained with Coomassie blue

overnight and destained in 20% methanol/10% acetic acid.

Regression of the MUllerian duct in response to different doses

of MIS RR and wild-type holo-MIS was demonstrated in an

organ culture bioassay (37) in which urogenital ridges of 14.5-

day timed pregnant Holtzman rat fetuses were harvested and

placed for 3 days at 37#{176}Cin organ culture dishes on agar over

Cambridge Medical Research Laboratory 1066 media contain-

ing 10% female FCS. The urogenital ridges were then oriented,

fixed, stained with hematoxylin and eosin, and serially sec-

tioned. Regression of the MUllerian duct was graded using a

morphological scale (37).



Clinical Cancer Research 345

Monolayer Growth. Monolayer growth of cells trans-

fected with the different constructs was determined by compar-

ing cell numbers obtained on different days after seeding P100

Petri dishes (Falcon) with the same initial cell number. Prior to

plating the clones for the experiments, the OM431-transfected

clones were washed in HBSS, trypsinized, and seeded at a high

density, and allowed to grow for 2 days. This was repeated to

ensure that the cells were in logarithmic growth. The cells were

then seeded in a similar manner and allowed to grow to density

arrest over 4 days to assure exposure to secreted MIS; the cells

were then trypsinized and further resuspended in fresh media

after centrifugation to remove the trypsin. The cells were

counted using a Coulter counter, resuspended to a concentration

of 14,000 cells/ml, and recounted to verify the accuracy of the

dilution. Transfected cells (140,000) were plated in P100 dishes

and the cells in 2-5 plates counted for each clone on days 1, 2,

4, and 6 to assess growth. All clones were grown in a-MEM,

a-MEM+, 10% female FCS (with no exogenous MIS as doe-

umented by immunoassay for bovine MIS; Ref. 30), and

0.5 mg/ml Geneticin.

Lung Metastasis Assay. Cells were grown and cycled as

described above, trypsinized, counted, and resuspended to a

concentration of400,000 cells/mi. Female SCID (41) mice were

obtained from the Massachusetts General Hospital Radiation

Biology Department where all work was done in accordance

with a protocol approved by the Massachusetts General Hospital

Animal Care Institutional Review Board. Using sterile tech-

nique, the tail veins of these mice were washed with alcohol

followed by sterile mineral oil, and then injected with 0.25 ml of

each of the transfected OM43 1 clones using tuberculin syringes

and a 27-gauge needle, and the animals were returned to their

cages. Four to 5 weeks later the lungs were harvested, inflated

uniformly with 4% paraformaldehyde at 20 cm H,O pressure,

and embedded in paraffin. Central coronal 8-mm sections were

cut and stained with hematoxyhin and eosin. Using the AppleS-

can (Macintosh) program, three central coronal lung sections

from each mouse were scanned and analyzed for pulmonary

metastases after screening by light microscopy to eliminate

areas of hemorrhage. Comparisons are of area occupied by

tumor to that occupied by the total inflated lung expressed as a

ratio, using the ImageGraphics software (version 1.37). Values

for pulmonary tumor area ratio are expressed as the mean ± SE.

Reduction in percentage of tumor area occupied by MIS RR

cells was analyzed statistically with post hoc testing by ANOVA

using the ‘ ‘SuperANOVA’ ‘ software package from ABACUS

concepts. A P of < 0.05 was considered to he statistically

significant.

RESULTS

Prior to determining the biological activity of the mutant

MIS RR protein, production was measured by ELISA in eight

MIS RR OM431 clones. Samples of culture media were ob-

tamed when cells reached near confluency. Although all clones

secreted MIS RR, two were selected for further evaluation. A

high MIS RR OM431 clone that produced an average of6.#{244}l p.g

MIS RR/1,000,000 cells (ii = 10) after 6-7 days in culture was

compared to a low-producing clone that made 1.59 p.g MIS

RR/1,000,000 cells (ii = 9) over the same time period. The

kD

1O6� -

Qrt_,_l_u�J ----� � -

49.5- y::-�..� ‘-1

32.5- � ----- � -.�:.,..

275- �

1 8.5- ______� -

1 2Fig. I Polyacrylamide gel electrophoretic analysis of MIS RR is rep-

resentative of the degree of endogenous cleavage of MIS RR resulting

in the production of 55- (large arrow) and l2.5-kDa (s,nall arrow)

fragments separated on these reducing gels and stained with Coomassie

brilliant blue (Lane 2). Intact 70-kDa holo-MIS RR monomer appears

above the 55-kDa band (Lane 2). The position of the prestained molec-

ular weight markers is given on the left in Lane 1.

ELISA of media samples for the empty vector OM431 clone

could detect no MIS (n = 10).

When purified to homogeneity by immunoaffinity chroma-

tography, MIS RR protein exhibited 70-, 55-, and 12.5-kDa

protein bands, visualized after Coomassie blue staining of poly-

acrylamide clectrophorcsis gels, indicating that reproducibly a

considerable degree of cleavage of MIS had occurred (Fig. 1).

The biological activity of MIS RR in causing regression of the

MOllerian ducts was determined in the standard rat urogenital

ridge assay. The dose response of a representative immunoaf-

finity-purified MIS RR preparation showed partial (0.925 p.gJ

ml) to complete (2-4 p.g/ml) regression of M#{252}llerian ducts,

similar to that caused by wild-type MIS.

Both the OM431/MIS RR high-producing (n = 8) and

low-producing (n = 6) clones were significantly growth inhib-

ited (P < 0.05) in vitro compared to the empty vector OM431

clone (ii = 10) (Fig. 14). In these experiments the MIS RR

OM43 1 high-producing clone produced a mean of 8.7 p.g MIS



0 VECTOR

t2 MIS RR LOW

0 MIS RR HIGH

100

75

50

25

0

B

Lu�

z-j-jLuC.) 5.+5

-a-- EMPTY VECTOR

-.-- LEADERLESS-‘--- MIS RR HIGH

-�0 1 2 3 4

DAY OF INCUBATION

6 M. A. Maggard, E. A. Cathin, P. L. Hudson, P. K. Donahoe, and D. T.

MacLaughlin. Activated MIS blocks EGF receptor phosphorylation viaa tyrosine phosphatase. Metabolism, submitted for publication.

346 Cleavage of MIS Activates Antiprohiferative Effects

Fig. 2 A, MIS RR inhibits 0M431 monolayer cell growth. High (n =

8, 0) and low (n = 6, �) MIS RR-producing OM431 clones were bothsignificantly growth inhibited (P < 0.01) relative to OM431 clonal cellstransfected with empty vector alone (n = 10, 0). Furthermore, the MISRR high-producing clone grew significantly less than the low-producingclone (P < 0.02). In each case, cells were counted on day 6 of culture.Cell counts were normalized to the empty vector as a control, andpresented as mean ± SE. B, inhibition of OM431 cell growth dependenton MIS secretion. Clonal cells transfected with empty vector (n = 5, 0)or with the leaderless MIS mutant (n = 5, S), which does not secreteMIS, grew equally well, whereas growth of the MIS RR high-producingclonal cells (n = 5, A) was significantly inhibited (�, P < 0.0002) afterday 4 of incubation.

RR/106 cells and the low-producing clone generated a mean of

2.2 p.g MIS RR/106 cells. Monolayer growth of the MIS RR

0M431 high-producing clone was also significantly inhibited

compared to the growth of the MIS RR clone (P < 0.05)

producing lower levels of MIS. Secretion of MIS appears to be

required for its biological activity in these cells since the pro-

liferation of an 0M431 cell line transfected with the leaderless

MIS mutant, which lacks a signal peptide, was not significantly

different from the 0M431 clone transfected with empty vector

(n 5) but was significantly greater (P < 0.05) than the MIS

RR high-producing clone (n = 5, Fig. 2B).

MIS RR-transfected 0M431 cells were significantly

growth inhibited in vivo. Representative mid-sagittal lung sec-

tions from mice given injections of the 0M431 clone producing

high levels of MIS RR show a significant reduction in total

percentage area of lung occupied by metastases compared to

either 0M431 cells containing the MIS RR low producer or the

leaderless MIS clones (Fig. 3). Growth of tumors in all animals

is compared in Table 1. The 0M4311M!S RR high-producer

clonal cells resulted in fewer metastases as compared to the

clones transfected with the low MIS producer (P < 0.002),

empty vector (P < 0.01), or leaderless MIS (P < 0.02) con-

structs.

DISCUSSION

In the male fetus, MIS induces regression of the M#{252}llerian

ducts, preventing the formation of the Fallopian tubes, uterus,

and upper third of the vagina (1-4). This inhibitory effect of

MIS on normal development provoked an early interest in the

effect of MIS on tumors arising from MUllerian duct precursor

cells. Previously, we showed that MIS has an antiproliferative

effect on established and primary endometrial and ovarian car-

cinoma cells in vitro (15-18). Subsequently, an inhibitory effectof MIS was observed on the human ocular melanoma cell line,

0M431, of neural crest derivation (19-21). Based on observed

autocrine inhibitory effects on granulosa cell growth and pro-

gesterone production (42, 43) and paracrine inhibitory effects on

oocytes (10, 11) and spermatogonia4 (12), as well as the pattern

of mRNA expression of candidate MIS receptors5 (22-24), we

suspect that germ cell and granulosa cells may be additional

therapeutic targets. In order to consider in vivo tests of efficacy

against such a wide range of human tumors, we anticipated the

need to identify a reliable and stable MIS ligand and to stream-

line its production.

MIS requires proteolytic processing to generate the biolog-

ically active carboxy terminus which has TGF-�3 homology (31).

Immunoaffinity-purified recombinant human MIS, as currently

generated from transfected CHO cells (29), is, however, only

partially cleaved. We previously demonstrated that the antipro-

liferative effect of wild-type MIS on ocular melanoma (21)

varies with the amount of cleavage present during production

and purification (30). Similar variability was seen when wild-

type MIS preparations were assessed for ability to inhibit auto-

phosphorylation of the epidermal growth factor receptor, i.e.,

increased cleavage resulted in enhanced inhibition of autophos-

phorylation.6 The specific endogenous monobasic cleavage en-

zyme responsible for this activity in the Sertoli cells of the testis

or in target tissues has not yet been identified; however, it is

clear that such cleavage is essential for bioactivity (30). A more

efficient TGF-�3-like dibasic R427-R428 cleavage site was crc-

ated to replace MIS R427-S428 as a strategy to facilitate cleav-

age of MIS, and the present experiments were designed to assess

the effect of this modification on the antiproliferative activity of

MIS.

Culture media from stably transfected MIS RR 0M431

cells were concentrated and tested in the organ culture Mftllerian

duct regression bioassay (37) to assess biological activity of the

mutant protein. Once biological activity was confirmed, culture

media from the MIS RR 0M431 high-producing clone were

immunoaffinity purified and, on gel electrophoresis, found to be

the cleaved MIS species. Small quantities of this immunoaffin-



Fig. 3 Lung metastases were inhibited in ani�als given injections of transfected OM43 1 cells. Lungs from SCID mice given injections 4 weeksearlier with either 100,000 cells of leaderless MIS (A), MIS RR low-producing (B), or high-producing (C) OM431 clones, uniformly inflated with20 cm H2O pressure at the time of harvest, were embedded in paraffin, sectioned, and H & E stained. Compared to leaderless MIS and MIS RRlow-producing OM431 clones, the lungs from the SCID mice given injections of the MIS RR OM431 high-producing clone showed the least numberof metastases.


Table I MIS RR results in significant reduction in metastases of0M431 cells

OM431 Clone

Mean % tumorarea:

experiment 1

Mean % tumorarea:

experiment 2

Empty vector 20.473 ± 2.365a

(n 7)

Leaderless MIS 17.420 ± 4.378

(n3)

18.590 ± 2.746

(n10)

MIS RR high producing 1.696 ± 0.732 1.039 ± 0.256

MIS RR low producing(n5)

6.816 ± 4.378

(n7)

(n10)

14.854 ± 2.729

(n8)

a The mean percentage of tumor area ± SE of lungs from animalsgiven injections of the OM431 MIS RR high-producing clone indicatedthat this clonal line produced the least number of metastases whencompared to MIS RR low-producing, empty vector (n = 7, P < 0.0028),and leaderless MIS (n = 3, P < 0.0096 and n = 10, P < 0.0001)OM431 clones.

ity-purified MIS RR variant could also induce M#{252}llerian duct

regression in the urogenital ridge bioassay similar to the regres-

sion caused by wild-type MIS. When the efficacy of this more

easily cleaved product was tested by comparing growth of

MIS-sensitive cells transfected with MIS RR, vector alone, or

with the transport-defective leaderless MIS, OM431 clones

transfected with MIS RR were found to be the most growth

inhibited in vitro in monolayer culture. Furthermore, high-pro-

ducing clones were more inhibited than low-producing clones.

The increased antiproliferative effect of MIS RR on MIS-

responsive cells observed in the present experiments correlated

with increased cleavage facilitated by the mutated dibasic cleav-

age site, which is observed under the acidified conditions of

purification (44), as with TGF-�3 (45). This inhibitory effect was

also seen in vivo when transfected OM431 cells were injected

into the tail vein of female SCID mice and pulmonary metas-

tases counted. The area and, by extrapolation, the volume of

lung covered by tumor was significantly less in the mice given

injections of the MIS RR 0M431 high-producing clone when

compared to those given injections of similar numbers of clonal

cells secreting no MIS. The variety of clones used in this study

also ruled against positional effects as the cause of the inhibi-

tion.

One of the theoretical appeals of developing biological

modifiers for cancer chemotherapy is their anticipated relative

specificity and lack of toxicity. An explosion of research into

genetic defects correlated with cancers has led to a host of

discoveries of abnormalities in cell cycle control, transcription

factors, and DNA repair (46-53). Even if all of the cancer-

related mutations in the downstream machinery controlling

growth progression can be identified, it is probable that effective

treatment strategies may entail both replacement of the defective

gene in that pathway, then treatment of the tumor with the

natural ligand that initiates the inhibitory cascade, since rescue

therapies that bypass a defect in a universal growth arrest

pathway may lack the specificity required to avoid toxicity to

nontumor cells.

Given the promising inhibitory effects seen with the more

easily cleaved MIS ligand compared to either the noncleaved or

nonsecreted MIS, we are now using methotrexate-driven ampli-

fication to increase production of MIS RR in DHFR-deficient



34�8 Cleavage of MIS Activates Antiproliferative Effects

31. Pepinsky, R. B., Sinclair, L. K., Chow, E. P., Mattaliano, R. J.,Manganaro, T. F., Donahoe, P. K., and Cate, R. L. Proteolytic process-

CHO cells as well as a novel yeast. Purified MIS RR protein can

then be compared with wild-type MIS for specific activity and

stability, and production of the most effective ligand scaled up

for use in in vivo animal trials and eventually in human clinical

trials against MIS-sensitive tumors preselected by the presence

of MIS receptor. Meanwhile a search is underway to uncover the

endogenous enzyme that cleaves MIS, with the expectation that

cotransfecting its gene with the MIS gene may yield the most

efficient production of cleaved and hence activated MIS.

ACKNOWLEDGMENTS

We are grateful to Gretchen Crist for assuming the responsibility of

performing all of the MIS ELISAS necessary for the study and Steven

Kim for his technical expertise in assisting with the animal dissections.

REFERENCES

1. Jost, A. Sur la differenciation sexuelle de l’embryon de lapin re-marques au sujet des certaines operations chirurgical sur l’embryon.C. R. Seances Acad. Sci. Ser. III Sci. Vie., 140: 461, 1946.

2. Jost, A. Rescherches sur la differenciation sexuelle de l’embryon de

lapin. Arch. Anat. Microsc. Morphol. Exp., 36: 271-315, 1947.

3. Jost, A. Sur les derives mulleriens d’embryons de lapin des deuxsexes castres a21 jours. C. R. Seances Acad. Sci. 5cr. III Sci. Vie., 41:135-136, 1947.

4. Blanchard, M., and Josso, N. Source of anti-Mullerian hormonesynthesized by the fetal testis: Mullenan inhibiting activity of fetalbovine Sertoli cells in tissue culture. Pediatr. Res., 8: 968-971, 1974.

5. Donahoe, P. K., Ito, Y., Price, J. M., and Hendren, W. H. M#{252}llerianinhibiting substance activity in bovine fetal, newborn, and prepubertaltestes. Biol. Reprod., 16: 238-243, 1977.

6. Hayashi, M., Shima, H., Hayashi, K., Trelstad, R. L., and Donahoe,P. K. Immunocytochemical localization of M#{252}llerianinhibiting sub-stance in the rough endoplasmic recticulum and Golgi apparatus inSertoli cells of the neonatal calf testis using a monoclonal antibody. J.Histochem. Cytochem., 32: 649-654, 1984.

7. Takahashi, M., Hayashi, M., Manganaro, T. F., and Donahoe, P. K.

The ontogeny of M#{252}llerianinhibiting substance in granulosa cells of thebovine ovarian follicle. Biol. Reprod., 35: 447-453, 1986.

8. Ueno, S., Kuroda, T., MacLaughlin, D. T., Ragin, R. C., Manganaro,T. F., and Donahoe, P. K. Mullerian inhibiting substance in the adult ratovary during various stages of the estrous cycle. Endocrinology, 125:

1060-1066, 1989.

9. Hirobe, S., He, W. W., Gustafson, M. L., MacLaughlin, D. T., andDonahoe, P. K. M#{252}llerianInhibiting Substance gene expression in thecycling rat ovary correlates with dominant follicle selection. Biol. Re-prod., 50: 1238-1243, 1994.

10. Takahashi, M., Koide, S. S., and Donahoe, P. K. Mullerian inhib-iting substance as oocyte meiosis inhibitor. Mol. Cell. Endrocinol., 47:

225-234, 1986.

1 1 . Ueno, S., Manganaro, T. F., and Donahoe, P. K. Human recombi-nant Mullerian inhibiting substance inhibition of rat oocyte meiosis isreversed by epidermal growth factor in vitro. Endocrinology, 123:

1652-1659, 1988.

12. Taketo, T., Saeed, J., Nishioka, Y., and Donahoe, P. K. Delay of

testicular differentiation in the B6.YDOM ovotestis demonstrated by

immunocytochemical staining for Mullerian inhibiting substance. Dcv.Biol., 146: 386-395, 1991.

13. Catlin, E. A., Manganaro, T. F., and Donahoe, P. K. MullerianInhibiting Substance depresses accumulation in vitro of disaturatedphosphatidylcholine in fetal rat lung. Am. J. Obstet. Gynecol., 159:

1299-1303, 1988.

14. Hutson, J. M., and Donahoe, P. K. The hormonal control of testic-ular descent. Endocr. Rev., 7: 270-283, 1986.

15. Donahoe, P. K., Swann, D. A., Hayashi, A., and Sullivan, M. D.Mullerian duct regression in the embryo is correlated with cytotoxicactivity against a human ovarian cancer. Science (Washington DC),205: 913-915, 1979.

16. Donahoe, P. K., Fuller, A. F., Jr., Scully, R. E., Guy, S. R., andBudzik, G. P. Mullerian inhibiting substance inhibits growth of a humanovarian cancer in nude mice. Ann. Surg., 194: 472-480, 1981.

17. Fuller, A. F., Jr., Guy, S. R., Budzik, G. P., and Donahoe, P. K.Mullerian inhibiting substance inhibits colony growth of a human ovar-ian cancer cell line. J. Clin. Endocrinol. Metab. 54: 1051-1055, 1982.

18. Fuller, A. F., Jr., Krane, I. M., Budzik, G. P., and Donahoe, P. K.Mullerian inhibiting substance reduction of colony growth of humangynecologic cancers in a stem cell assay. Gynecol. Oncol., 22: 135-148,1985.

19. Chin, T., Parry, R. L., and Donahoe, P. K. Human Mullerianinhibiting substance inhibits tumor growth in vitro and in vivo. CancerRes.,51: 2101-2106, 1991.

20. Parry, R. L., Chin, T. W., Epstein, J., Powell, D. M., Hudson, P. L.,and Donahoe, P. K. Recombinant human Mullerian inhibiting substanceinhibits ocular melanoma cell lines in vitro and in vivo. Cancer Res., 52:

1182-1186, 1992.

21. Boveri, J. F., Parry, R. L., Ruffin, W. K., Gustafson, M. L., Lee,K. W., He, W. W., and Donahoe, P. K. Transfection of the Mullerianinhibiting substance gene inhibits local and metastatic growth. Int. J.Oncol., 2: 135-143, 1993.

22. He, W. W., Gustafson, M. L., Hirobe, S., and Donahoe, P. K.

Developmental expression of four novel serine/threonine kinase recep-

tors homologous to the activin/transforming growth factor �3 Type IIreceptor family. Dcv. Dyn., 196: 133-142, 1993.

23. Baarends, W. M., van Helmond, M. J. L., Post, M., van der Schoot,

P. J. C. M., Hoogerbrugge, J. W., de Winter, J. P., Uilenbroek, J. T. J.,Karels, B., Wilming, L. G., Meijers, J. H. C., Themmen, A. P. N., andGrootegoed, J. A. A novel member of the serine/threonine kinasereceptor family is expressed in the gonads and in mesenchymal cells

adjacent to the mullerian duct. Development, 120: 189-197, 1994.

24. diClemente, N., Wilson, C., Faure, E., Boussin, L., Carmillo, P.,

Tizard, R., Picard, J-Y., Vigier, B., Josso, N., and Cate, R. Cloning,expression, and alternative splicing of the receptor for anti-Mullerianhormone. Mol. Endocrinol., 8: 1006-1020, 1994.

25. Attisano, L., Carcamo, J., Ventura, F., Weis, F. M. B., Massague, J.,and Wrana, J. L. Identification of human activin and TGF�3 type Ireceptors that form heteromeric kinase complexes with type II receptors.Cell, 75: 671-680, 1993.

26. Chen, R., Ebner, R., and Derynck, R. Inactivation of the type IIreceptor reveals two receptor pathways for the diverse TGF�3 activities.Science (Washington DC), 260: 1335-1338, 1993.

27. Budzik, G. P., Powell, S. M., Kamagata, S., and Donahoe, P. K.Mullerian Inhibiting Substance fractionation by dye affinity chromatog-raphy. Cell, 34: 307-314, 1983.

28. Picard, J. Y., and Josso, N. Purification of testicular anti-Mullerian

hormone allowing direct visualization of the pure glycoprotein anddetermination of yield and purification factor. Mol. Cell. Endocrinol.,34: 23-29, 1984.

29. Cate, R. L., Mattaliano, R. J., Hession, C., Tizard, R., Farber, N. M.,Cheung, A., Ninfa, E. G., Frey, A. Z., Gash, D. J., Chow, E. P., Fisher,R. A., Bertonis, J. M., Tories, G., Wailner, B. P., Ramachandran, K. L.,Ragin, R. C., Manganaro, T. F., MacLaughlin, D. T., and Donahoe,P. K. Isolation of the bovine and human genes for Mullerian inhibitingsubstance and expression of the human gene in animal cells. Cell, 45:

685-698, 1986.

30. MacLaughlin, D. T., Hudson, P. L., Graciano, A. L., Kenneally,M. K., Ragin, R. C., Manganaro, T. F., and Donahoe, P. K. Mullerianduct regression and antiproliferative bioactivities of Mullerian inhibiting

substance resides in its carboxy-terminal domain. Endocrinology, 131:

291-296, 1992.




ing of Mullerian inhibiting substance produces a transforming growthfactor-�3-like fragment. J. Biol. Chem., 263: 18961-18964, 1988.

32. Smeekers, S. P., and Steiner, D. F. Processing of peptide precursors:

identification of a new family of converting proteases. Cell Biophys.,

19: 45-55, 1991.

33. Catlin, E. A., Ezzell, R. M., Donahoe, P. K., Gustafson, M. K., Son,E. V., and MacLaughlin, D. T. Identification of a receptor for humanMullerian inhibiting substance. Endocrinology, 133: 3007-3013, 1993.

34. Price, J. M., Donahoe, P. K., Ito, Y., and Hendren, W. H. Pro-

grammed cell death in the M#{252}llerianduct induced by MUllerian inhib-iting substance. Am. J. Anat., 149: 353-376, 1977.

35. Trelstad, R. L., Hayashi, A., Hayashi, K., and Donahoe, P. K. Theepithelial-mesenchymal interface of the male rat MUllerian duct: loss ofbasement membrane integrity and ductal regression. Dcv. Biol., 92:

27-40, 1982.

36. Kunkel, T. A., Bebenek, K., and McClary, T. Efficient site directedmutagenesis using uracil-containing DNA. Methods Enzymol., 204:

125-139, 1991.

37. Donahoe, P. K., Ito, Y., Marfatia, S., and Hendren, W. H. A graded

organ culture assay for the detection of Mullerian Inhibiting Substance.J. Surg. Res., 23: 141-148, 1977.

38. Cate, R. L., Donahoe, P. K., and MacLaughlin, D. T. MOllerianinhibiting substance. In: M. B. Sporn and A. B. Roberts (eds.), Hand-book Experimental Pharmacology, Vol. 95/Il, pp. 179-210. New York:Springer-Verlag, 1990.

39. Albert, D. M., Ruzzo, M. A., McLaughlin, M. A., Robinson, N. L.,Craft, J. L., and Epstein, J. Establishment of cell lines of uveal mela-noma. Invest. Ophthalmol. Visual Sci., 25: 1284-1299, 1984.

40. Hudson, P. L., Dougas, I., Donahoe, P. K., Cate, R. L., Epstein, J.,Pepinsky, R. B., and MacLaughlin, D. T. An immunoassay to detecthuman Mullerian inhibiting substance in males and females during

normal development. J. Clin. Endocrinol. Metab., 70: 16-22, 1990.

41. Custer, R. P., Bosma, G. C., and Bosma, M. J. Severe combinedimmunodeficiency (SCID) in the mouse. Am. J. Pathol., 120: 464-477,

1985.

42. Kim, J. H., Seibel, M., MacLaughlin, D. T., Donahoe, P. K., Ransil,B. J., Hametz, P. A., and Richards, C. J. The inhibitory effects of

Mullerian inhibiting substance on epidermal growth factor inducedproliferation and progesterone production of human granulosa-lutealcells. J. Clin. Endocrinol. Metab., 75: 91 1-917, 1992.

43. Seifer, D. B., MacLaughlin, D. T., Penzias, A. S., Behrman, H. R.,

Asmundson, L., Donahoe, P. K., Haning, Jr., R. V., and Flynn, S. D.

Gonadotropin-releasing hormone agonist-induced differences in granu-

losa cell cycle kinetics are associated with alterations in follicular fluid

Mullerian inhibiting substance and androgen content. J. Clin. Endocri-nol. Metab., 76: 711-714, 1993.

44. Ragin, R. C., Donahoe, P. K., Kenneally, M. K., Ahmad, M., and

MacLaughlin, D. T. Human Mullerian inhibiting substance: enhanced

purificatione imparts biochemical stability and restores antiproliferative

effects. Protein Expr. Purif., 3: 236-245, 1992.

45. Assoian, R. K. Purification of type-B transforming growth factor

from human platelets. Methods Enzymol., 146: 153-167, 1987.

46. Levine, A. Cancer in AIDS. Curr. Opin. Oncol., 4: 863-866, 1992.

47. Nevins, J. R. E2F: a link between the Rb tumor suppressor protein

and viral oncogenesis. Science (Washington DC), 258: 424-429, 1992.

48. Scherr, C. J. Mammalian Gi cyclins. Cell, 73: 1059-1065, 1993.

49. Xiong, Y., Hannon, G. J., Zhang, H., Casso, D., Kobayashi, R., andBeach, D. p21 is a universal inhibitor of cyclin kinases. Nature (Lond.),366: 701-704, 1993.

50. Polyak, K., Lee, M-H., Erdjument-Bromage, H., Koff, A., Roberts,J. M., Tempst, P., and Massague, J. Cloning of p27�”, a cyclin-dependent kinase inhibitor and a potential mediator of extracellular

antimitogenic signals. Cell, 78: 59-66, 1994.

51. Toyoshima, H., and Hunter, T. p27, a novel inhibitor of Gi cyclin-

Cdk protein kinase activity, is related to p21. Cell, 78: 67-74, 1994.

52. Kamb, A., Gruis, N. A., Weaver-Feldhaus, J., Liu, Q., Harshman,

K., Tavtigian, S. V., Stockert, E., Day, R. S., Johnson, B. E., andSkolnick, M. H. A cell cycle regulator potentially involved in genesis ofmany tumor types. Science (Washington DC), 264: 436-440, 1994.

53. Nobori, T., Miura, K., Wu, J. D., Lois, A., Takabayashi, K., and

Carlson, D. A. Deletions of the cyclin-dependent kinase-4 inhibitor genein multiple human cancers. Nature (Lond.), 368: 753-756, 1994.



1995;1:343-349. Clin Cancer Res M S Kurian, R S de la Cuesta, G L Waneck, et al. antiproliferative effects in vivo.Cleavage of Müllerian inhibiting substance activates

Updated version

http://clincancerres.aacrjournals.org/content/1/3/343

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://clincancerres.aacrjournals.org/content/1/3/343To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



Documents

Cleavage of M#{252}llerian Inhibiting Substance Activates … · Vol. 1. 343-349. March 1995 Clinical Cancer Research 343 Cleavage of M#{252}llerian Inhibiting Substance Activates