Note: Descriptions are shown in the official language in which they were submitted.
W092/10l99 -207530~
PCT/VS9l/08986
--1--
Des~ription
Therapeutic And Diagnostic
Applications Of Fetal Fibronection
:~.
Field of the Invention
The present invention relates to the field of
mammalian reproduction and in particular, the invention
relates to therapeutic and diagnostic applications
through the manipulation of the protein, fetal
5 fibronectin (FFN). More particularly, therapeutic and `~
diagnostic applications relating to fertility
enhancement, contraception and contragestion are
provided.
,
Back~rou~d of the Invention
10In the field of mammalian reproduction, many
diagnostic procedures exist to aid the
obstetrician-gynecologist in making a diagnosis and
choosing an appropriate course of action.
Infertility in humans is defined as one year of --
unprotected coitus without contraception. Although
approximately 10-15% of couples are affected by
infertility, the risk of infertility is doubled for
women between the ages of 35 to 44 as compared to women
between the ages of 30 and 34. Changes in fertility
patterns such as re~ult from increased maternal age,
will have a significant impact on the make-up of
populations. It has been calculated that, without
immigxation, the population in the United States will
decline by t:he middle of the next century.
Furthermore, the percent of people over the age of 65
will increaE,e to over 23% in the next 100 years, resul-
ting in an older and smaller work force.
In the United States, 40% of infertility can be
accounted for by problems in the male. There~ore,
WO 92/1019g 7$ ~ U 9
;; PCT/US 91/0X986.-
semen analysis should be an early diagnostic step forinvestigating fertility. In evaluating male factor
infertility, tests have been developed to assess the
quality of a mammalian spe~ sample. Examples include
monitoring sperm morphology and motility and assessment
of the ability of sperm to penetrate zona free hamster
; ova and undergo an acrosomF- reaction. In addition,
antisperm antibodies can be detected in semen, and
; maternal sera. Post-coital examination of the female
partner has been used to evaluate the ability of sperm
to reach and survive in the cervical mucus.
Evaluating a female for infertility can be co-
mplex. Examination of the fallopian tubes is an
important step in mammalian fextility evaluation due to
the evidence of increase of pelvic inflammatory
disease. Currently, a hysterosalpolingogram (HSG) is
the procedure of choice to examine the patency of the
fallopian tubes. In addition to HSG, hysteroscopy,
the direct examination of the uterus by a fiber optic
device, can identity endometrial polyps, submucous
leiomyomas Mullerian anomalies, and other abnormalities
within the uterus itself.
Another category of diagnostis~ procedures includes
examination of ovarian function including ovuslation and
the secretion of progesterone during the luteal phase
of the menstrual cycle. Ovarian function can be
crudely assessed by measuring basal body temperatures
during the menstrual cycle and cervical mucous testing
around the time of ovulation. More accurate tasting
can be performed by measuring luteinizing hormone, a
pituitary hor~one which induces ovulation after a
mid-cycle surge. Finally, serum progesterone levels
can be measured to assess the adequacy of the luteal
pAase of the menstrual cycle.
In orcler to rule out inadequate ovarian
progesterone effects in secretory endometrium,
W0~2/10199 2 0 7 5 3 0 0 PCT/US91/08986
-3-
so-called luteal phase defects, the endometrium itself
has been directly assessed by performing an endometrial
biopsy three days before the ~uspected onset of menses.
Endometrial biopsies are examined by hematoxylin and
eosin staining of paraffin embedded specimens. For
infertility patients, the reading of these biopsies may
provide clinically useful information about the
adequacy of the luteal phase, and other potential data,
such as the presence of infection, inflammation, or
; 10 neoplasia of the endometrium.
Finally, the infertility patient could undergo
endoscopic examination through an incision in the
; abdomen to directly visualize the external surfaces of
the ovary, fallopian tubes and uterus, therefore
identifying any gross pathology which was not detected
by previous examinations.
Endometriosis is associated with severe
dysmenorrhea, and can result in infertility and
debilitating pelvic pain. There are no current methods
for establishing the definite diagnosis of
endometriosis without resorti~g to direct visualization
of the pelvis and abdominal cavity by operative
laparoscopy. Therefore there i5 a pressing need for a '
relatively non-invasive method of imaging endometrial
implants.
Mullerian anomalies such as unicornate and
bicornate uteri can be associated with infertility and
recurrent pregnancy 105s. Currently, the diaynosis of
these abnormalities frequently requires operative
procedures such as hysteroscopy and laparoscopy. A
non-invasive method of delineating the anatomic
distribution of functional Mullerian Ppithelium and
thus anatomy of these structures would bP clinically ;
useful.
In vitro fertilization (IVF) requires the removal
of ova from a mammalian ovary, and exposure of these
~',
.
. .
~O92/10l99 2 0 7 5 3 0 ~ PCT/VS91/08986t--
~
; -4~
ova to sperm outside the body. Fertilization of each
ovum requires that at least one living sperm penetrates
the zona pellucida touter covering) of the ovum and
fuses with the pronucleus. Once this has occurred and
the ova are fertilized, they can be transferred to a
uterus where they then can become implanted on the
uterine wall. If implantation occurs, the pregnancy
proceeds as if fertilization had occurred within the
body. IVF has gained widespread professional and
public acceptance. However, despite the ever
increasing frequency and refinement of this procedure,
IVF attempks most often do not result in pregnancy.
IVF pregnancy rat~s are currently only about 15 to 20
percent. For a variety of reasons, exposing the ova to
sperm does not nacessarily result in fertilization.
More often, fertilized ova fail to implant in the
uterus. The low success rate of IVF often leads to an
excessive financial and psychological burden for the
infertile couple.
other assisted reproductive technologies include
two modifications of the IVF technique. The first is
gamete intra-fallopian transfer (GIFT), the second is
zygote intrafallopian transfer (ZIFT). In the GIFT
procedure, the retrieved oocyte and sperm are mixed
together and placed ~ack into the fallopian tube where
fertilization takes place. The fertilized zygote then
travels down through the fallopian tube into the
endometrial cavity, where implantation may or may not
take place. The ZIFT procedure allows for fertiliz- -
ation to take place in vitro as in standard IVF, the
fertilized zygote then is placed back in the fallopian
tube where it travels down into the uterus to implant.
Finally, hyper-stimulation protocols necessary to
retrieve many oocytes from the donor women may have
35 deleterious effects on the endometrium itself and ~ ;
decrease the rates of implantation. Two basic
2B7~30~
WO92/10l99
; ~ PCT/US91/08986
.,,; , . .~ , . .
--5--
procedures have been utilized to help overcome this
pro~lem. The first is non-stimulated oocyte retrieval.
A single egg is retrieved, allowed to be fertilized and
placed back into the fallopian tube or uterus for
implantation. The other technique involves the
hyperstimulation portion of the IVF procedure to
retrieve the eggs and allow for fertilization in vitro.
- The æygote is then frozen to be placed back into the
patient after several normal cycles, with the hope
that the endometrium will then be more receptive to
- implantation. All of these techniques attempt to
; maximize the quality of the eggs, zygotes produced
after ~ertilization and the receptivity of the end-
ometrium. Any procedurP which would enhance the
implantation rate above the standard 15 to 20% would
have a markedly positive e~fect on any of these
technologies.
Even if implantation occurs, pregnancy loss may
occur. Pregnancy loss during the first six weeks has
bean shown to occur in 35 to 50% of pregnancies
confirmed by hCG analysis. Wilcox et al., "Incid~nce
of Early Loss of Pregnancy" New Engl. J~ Med. (1988)
319:189-194. Furthermore, the chance of a successful
live birth a~ter three consecutive abortions without a
live birth may be as low as 40-50~. There is clear
precedence for autoimmune pregnancy loss and
infertility. It has been shown that patients who have
general autoimmune diseases have a high incidence of
reproductive failure. Gleicher et al., '~The
Reproductive Autoimmune Failure Syndrome~" Am. J.
Ostet. Gynlecol. (1988) 159:223-227. There is also
evidence oE autoimmune mediated recurrent pregnanoy
loss in patients without clinically apparent autoimmune
disease. Lockwood et al., Lancet 11:742-45, (1986);
and Lockwood et al., "The Prevalence and Biological
Significance of Lupus Anticoagulent and Anticardiolipin
WO92/10199 2 0 7:S 3 0 0 PCT/US91/0~986 ~-
--6--
Antibodies in a General Obstetric Population," Am. J.
Obstet. Gynecol., 161:369~373 (1989).
The incidence of Down Syndrome (DS) is
; approximately 1/1000 live births. Adams et al.,
"Down's Syndrome: Recent Trends in the United States",
JAMA (1981) 246:758-760. This incidence is expected to
increase modestly over the next 10 years as the number
of women becoming parents between 35 and 49 years of
- age increases. Goodwin and Huether, "Revised Estimates
- 10 and Projections of Down Syndrome Births in the United
States, and the Effects of Prenatal Diagnosis
Utilization, 1970-2002", Prenatal Diagnosis (1987)
7:261-271. However, by the year 2000 women over the
age of 34 will still account for only 8.1% of all live
births and 39% of DS cases. Thus even a fully
subscribed to maternal age-based DS screening program
would fail to identify 61% of DS cases. Goodwin and
Huether (1987). Moreover, only a minority (20% - 30%)
of women 35 years of age or older avail themselves of
fetal karyotype analysis. Goodwin and Huether id.; and
Hook et al., "Use of Cytogenetic Diagnosis in New York
State", New Engl. J. Med. (1981) 305:1410-1413.
This inherent inefficiency and practical
underutilization of DS scre~ning has prompted a
concerted effort to improve the ability to detect the
DS fetus. Lockwood et al., "A Sonographic Screening
Method for Down's Syndrome", Am J. Obstet. Gynecol.,
157:803-808 (1987). These efforts have included the
development and expansion of maternal serum biochemical ~;~
markers and the evaluation of potential sonographic
indicators of the DS fetus. The association of
decreased maternal serum ~-fetoprotein (MSAFP) levels
with fetal DS has now been well established. Ris~
deter~ination based on a combination of MSAFP value and
maternal age allows for the detection of a third of DS
fetuses whlle necessitating karyotype analysis in 5% of
:
WO92/l~l99 ? 0 7~ 3 ~ 0 PC~/US~l/08986
pregnant patients under 35. Wald et al., "Maternal
Serum Screening for Down's Syndrome in Early
Pregnancy", Brit. Med. J. (1988) 297:883-887.
Recently, Bogart and associates observed higher
; 5 than expected concentrations of the HCG ~ subunit in
women bearing DS fetuses b~etween 8 and 25 weeks
gestation. Bogart et al., "Abnormal Maternal Serum
Chorionic Gonadotropin Levels in Pregnancies with Fetal
Chromosome Abnormalities", Prenatal Diagnosis (1987)
7:623-630. Since HCG values generally plateau at
between 18 and 25 weeks of gestation, an elevated HCG
value appears to be a reliable index af DS risk.
Maternal serum unconjugated estriol levels may also be
marginally lower in DS pregnancies. Canick et al.,
"Low Second Trimester Maternal Serum Uncon~ugated
Oestriol in Pregnancies with Down's Syndrome", Br. J.
Obstet. Gynaecol. (1988) 95:330-33; and Macri et al.,
"Maternal Serum Down Syndrome Screening: Unconjugated
Estriol is not Useful", Am. J. Obstet. Gynecol. (1990)
162:672-673. However, DS risk assignment paradigms
which employ maternal age with MSAFP, HCG and estriol
values may identify 60% of DS fetuses while
necessitating karyotype analysis in only 5% of pregnant
women. Wald et al., Id. Clearly additional maternal
biochemical markers for ~etal DS and other chromosomal
abnormalities would have great utility.
Localization of ectopic pregnancies and metastatic
gestational trophoblastic disease can be challenging.
The manifestations of tubal pregnancy include
amenorrhea, vaginal spotting or bleeding, abdominal or
pelvic pain and the presence of a pelvic mass.
Laboratory testing for suspected ectopic pregnancy
includes hPmoglobin and hematocrit, white blood cell
count, urine and serum human chorionic gonadotropin
(HCG~ pregnancy test, ultrasound including vaginal
probe ultrasound, culdocentesis (a diagnostic procedure
W0 92/l~l~9 2q7~0 PC~/U591/~8~861
to detect blood in the perit:oneum), curettage of the
endometrium to rule out the presence of products of
conception within the uterus, laparoscopy and finally
in emergency cases, laparotomy.
Given the wide variability of patient
presentations and the course of ectopic pregnancy, the
accurate diagnosis of this disorder is sometimes
difficult. In one study of three hundred women,
approximately a third were seen more than once, and 11%
were seen more than twice before the correct diagnosis
was made. In addition, in a recent study of deaths
from ~ctopic pregnancy, more than half of the cases
were misdiagnosed leading to fatal maternal outcome.
Clearly, accurate and rapid diagnosis and/or treatment
of an ectopic pregnancy would be an important advancP
for the field of obstetrics and gynecology.
The current modalities for detecting metastatic
gestational trophoblastic disease include serum HCG
level determination, chest X ray, pelvic ultrasound,
magnetic resonance imaging (MRI~ computerized
tomography (CT) scan of the abdomen, pelvis and head.
Like in other solid tumors, small mètastases can be
missed by these procedures. The knowledge of the
presence of metastases is critical for the successful
treatment of these and other tumors. Therefore, a
method which can localize and detect small quantities
of trophoblast tissue would be very helpful for the
management of this disease.
In the United States, commonly employed
contraceptive techniques include oral steroidal
contraceptives, injected or implanted steroidal
contraceptives, intrauterine devices, physical,
chemical or physiocochemical barrier techniques,
withdrawal, sexual abstinence around the time of
ovulation, breast feeding and permanent sterilization.
In addition to the high failure rates of some of these
; WO92/10t99 2~753~o
PCT/US9l/0898
_g_
methods, a number of these methods have serious pote-
ntial complications for the users. For example, in
addition to metabolic changles induced by some oral
contraceptives, there is a ]potentially increased risk
of untoward cardiovascular ~effects and thromboembolisms
in susceptible patients. Moreover many o~ these
techniques are expensive and require meticulous medical
surveillance, and as such they are not applicable to
poorer patients in the third world and underprivileged
patients in the developed world.
Notwithstanding the work reported in this field, a
need still exists for improved diagnostic and
therapeutic applications in the field of mammalian
reproduction.
The present invention provides methods of
regulating the reproductive potential of a mammal by
regulating the amount of FFN to which the mammal's
cells and/or tissues are exposed. Also provided are
methods of monitorinq the reproductive potential of a
mammal by detecting the presence of, quantitating the
amount FFN, or determining the ratio of total
fibronectin to FFN to which the mammal's tissues are
exposed. Further provided by the invention are methods
of imaging and/or treating tissues expressing or
exposed to FFN.
Summary of the Invention
The present invention provides methods of
diagnosing fertili~y problems, methods of enhancing
fertility and methods of decreasing fertility
potential.
Before discussing the various embodiments of the
invention, it should first be noted that the inventor
has determi.ned that F~N plays a significant role or
roles in the mammalian reproductive process. While the
inventor has developed certain theories xegardin~ the
~, , , ,., - , , ; , ' ~ '
WO92/l0199 ~ 7 5 ~ O o PCT/~91/08986 ~
--10--
role(s) thus played, the usefulness of the diagnostic
and therapeutic methods disclosed herein do not
necessarily depend upon the accuracy of any given
theory. Indeed, certain theories may not be entirely
consistent with one another.
In this regard, and as is more fully explain~d
below, FFN has been determined to be localized in
various reproductive tissue, including the implantation
zone of the uterus. This would seem to indicate that
FFN plays a role, and perhaps a substantial or critical
one, of promoting or permitting the appropriate binding
of cells, and in particular, the fertilization of the
ova and the implantation of the blastocyst. Consistent
with this role, FFN may enhance or control sperm-ova
transport and/or contact, the adhesion of the
blastocyst to the endometrial epithelia and the
adherence of the placenta and membrane to the
endometrial decidua, and may thus be a critical factor
in determining whether fertilization and implantation
occur and the manner and timing of placental and
membrane separation, whether by way of premature
abortion or full term delivery.
For example, the unique glycosylation properties -
of FF~ might serve to enhance cell and ECM binding
affinity with or without altering proteolytic
susceptibility when compared to adult forms of
fibronectin. Therefore enhancement of seminal,
follicular, blastocyst or endometrial FFN synthesis -~
could facilitate fertilization, implantation and
placentation, while inhibition of FFN synthesis could
serve as a contraceptive or contragestational agent.
on the other hand, the inventor has theorized that
FFN might serve to inhibit inappropriate cell binding
(e.g. tubal implantation) and thus promote
fertilization or the pre-implantation transport of the
blastocyst.
.
WO92/101~9 2075~0
~ PCT/US91/08~6
- Should FFN have a relatively reduced cell and ECM
binding affinity compared with adult forms, it could
serve as a "lubricant" preventing injudicious binding
of gametes, zygote or blastocyst. Specifically,
seminal and Mullerian tract epithelial FFN secretion
could prevent sperm binding to the endocervix,
endometrium or tubal epithelium prior to fertiliæation,
- thus enhancing the latter's probability of occurrence.
Similarly, tubal epithelial and ~ollicular FFN could
act to prevent oocyte adherence to the tubal epithelium
with consequent failure of fertilization or, in the
; event of fertilization, ectopic implantation. Finally
the presence of endometrial glandular epithelial FFN
synthesis and secretion in the very early luteal phase
(prior to 19 days of the menstrual cycle) could prevent
i premature blastocyst adherence or implantation prior to
the period of optimal endometrial morphological and
biochemical receptivity (menstrual days 19-21). Analo-
gously, localization of FFN during pregnancy to the
sites of future placental and membrane separation could
facilitate separation of the placenta and membranes
following delivery at term. In addition, reduced FFN
cell and ECM binding affinity together with increased
proteolytic susceptibility could facilitate preterm
delivery secondary to either contractions or
inflammation. This latter view is consistent with the
association of preterm delivery with placental
separation (abruption) and preterm membrane rupture. `
Therefore, manipulation of FFN concentrations or
fibronectin moiety synthesi~ at these various sites
could serve to enhance or inhibit fertilization,
implantation and the maintenance of placental and
membrane structural integrity throughout gestation.
Also, depending upon whether the FFN acts as a
binding agent or a "lubricant", its presence at
different :Locations and at different times during the
WO92/10199 2 0 7 5 3 ~ O PCT/US91/08986
-12-
reproductive process may have opposite effects. For
example, if, as is presumab:Ly believed, FFN does indeed
act as a "lubricant", its presence may facilitate the
movement of sperm through the vagina, the cervix, the
uterus, and ultimately to the fallopian tubes where
fertiliæation occurs. Similarly, the presence of a FFN
"lubricant" would serve to lessen the probability that
the fertilized or unfertilized ova would implant
prematurely, e.g. in the fallopian tube. It would
further appear that an FFN "lubricant" would retard the
premature implantation of the blastocyst in the uterus.
However, FFN might also serve as a binding agent to
promote trophoblast-decidual contact and the formation
of placental and membrane connections to the uterus
sufficiently strong to maintain pregnancy but
sufficiently tenuous to facilitate expulsion of the
placenta and membranes following expulsion of the
fetus.
Hence, it is within the scope of the present
invention to enhance reproductive potential both by
enhancing and retarding FFN levels depending upon the
type of tissue involved and the time of treatment
relative to the reproductive process. As a general
proposition, it is expected that increased FFN levels
would enhance reproductive potential in in vlvo
fertilization (including artificial semination) at all
stages except at the time and location of implantation.
Methods of raising and lowering FFN levels are ~ -
discussed below.
It is within the scope of the present invention to
regulate FFM levels as a means of affecting
reproductive potential regardless of whether the FFN
level is increased or decreased, and regardless of
whether the FFN acts in all instances as a "lubricant"
or a binding agent.
- WO92/10199 2075~
PCT/VS91/08986
-13-
., ~
Thus, FFN may serve one of two dissimilar
functions i.e., either the enhancement or inhibition of
cell-cell or cell-ECM bonding -- or may serve both
functions at dlfferent locations and/or times -- e.g.
retardation of cell binding in the Mullerian tract, and
` promotion of implantation in the uterus.
In the circumstances, the discussion below and
disclosure of the various embodiments of the invention
may imply or presuppose a theory of FFN function that
may not be entirely consistent with a theory which is
implied or presupposed with respect to another
embodiment. It is, however, believed that one skilled
in the art will appreciate that the different or even
; inconsistent theories which are here proposed are not
intended to detract from the scope or significance of
the present invention, but rather to promote a better
understanding of the invention and the potential uses
' herein described.
The invention provides a novel method of
monitoring the fertilization potential of a mammal by
analyzing samples such as a semen sample or the
post-coital cervico-vaginal mucus or follicle aspirates
from a mammal to determine the relative likelihood that
pre-coital or post-coital semen from that mammal is `
capable of fertilizing an ovum from a female mammal for
the same species. Other types of tissue which may be
sampled include, but are not limited to, serum, sperm,
ova, endometrium, placenta, uterine fluid and
follicular fluid.
It has, however, been demonstrated that FFN is
present in siynificant quantities in normal semen and
post-coital cervico-vaginal mucus samples as well as
from follicle aspirates. Thus, the method of this
invention provides information as to whether the mammal
is producing FFN and in appropriate amounts for optimal
fertilization potential. For example, the method
.
~ WO92/10199
2 ~ 7 5 3 (3 ~ PCT/US91/~8986
- -14-
includes providing at least one sample of semen or
-~ post-coital or cervico-vaginal mucus and assaying the
sample for FFN, whereby a relative change in the co-
ncentration of FFN in said sample as compared to normal
- 5 levels of FFN or the ratio of FFN to adult fibronectin
in fertile mammals of the same species is indicative of
the fertilization potential.
, Concentrations of FFN in follicular fluid are also
substantial and may be correlated with fertility. The
lO invention also provides a diagnostic procedure for
ultrasound guided aspiration of follicles with measur-
ement of FFN concentrations of follicle aspirates
related to fertile mammals of the same species to
determine the likelihood of fertilization. Similar
15 comparison of FFN concentrations to adult fibronectin
7 concentrations can also be utilized for assessing
fertility potential.
; The invention further provides a novel infertility
and recurrent abortion screening test comprising
20 assaying at least one bodily fluid or cell type from a
' mammal suspected of being infertile or of having
unexplained recurrent abortions for the presence of FFN
autoantibodies. The association of the protein FFN
with normal mammalian reproduction has now been reco- ;
25 gnized. Thus, the method of this invention provides a
tool for cliagnosing mammals with an autoimmune
infertility.
This invention further provides a novel method for
altering the fertility potential of a mammal by
30 regulating the level of fibronectin in mammalian
reproductive tract tissues or secretions. Increases in
FFN can result from the infusion of exogenous FFN in
semen, follicular fluids or onto tissues or by
increasing the level of endogenous FFN by the use of
35 pharmaceutically acceptable substances. In one
embodiment of the present invention, the probability
:
. '. . . . . . ..
WO92/10~99 2 0 7 s 3 0 0 PCT/US91/0898~
-15-
that a conceptus will become implanted in a mammalian
uterus is altered by regulating the amount of FFN in
the uterine cavity at about the time the uterus will be
contacted with the conceptus. It has now been
recognized that during normal mammalian pregnancy, FFN
has been localized in at least several locations
including the i~plantation zone where tropAoblastic
cells make contact and attach to the uterus and within
the chorionic membrane where trophoblastic cells make
direct contact with the maternal decidua. In addition,
it has now been established that decidualized
endometrial stromal cells and endometrial glands
synthesize and secrete FFN in vitro. Thus the method
of this invention changes the local environment of the
surface of the endometrium at about the time the uterus
will be contacted with the conceptus thereby altering
the chance that a conceptus will become adherent and/or
subsequently enhances FFN synthesis following
implantation in the uterus.
Further provided by this invention is a novel
method of regulating FFN synthesis so as to modify the
probability of fertilization, blastocyst adhesion to
the endometrium and/or implantation by regulating FFN
synthesis in the mammalian uterus, ovarian follicle or
seminal vesicle. The method comprises administering to
said mammal a compound which regulates FFN synthesis by
mammalian endometrial glands in culture. It hais ~ow
been found that estrogen plus progesterone inhibit FFN -
synthesis in cultured endometrial glandular epithelia
whereas the same hormonal treatment maintains FFN
synthesis by cultured endometrial stromal cells.
Either estrogen or progesterone alone inhibit stromal
cell FFN synthesis. Therefore the inventian includes a
method of regulating FFN synthesis by administering
systemically or topically a therapeuticalLy effective
amount of putative paracrine effectors which are known
: ,:: -- , . . .
. .
~ -- wo s2/lnlss ' ' ' ';
~ o 7 ~ 3 o Q PCT/~'S91/08986
-16-
regulators of fibronectin synthesis such as
transforming growth factor ~ tTFG-~). Ignotz et al.,
"Regulation of Fibronectin and Type I Collagen mRNA
Levels by Transforming Growth Factor ~", J. Biol.
Chem., 262:6443-6446 (1987).
Additionally, hormones including but not limited
to estrogen, estrogen analogues, antagonists or
agonists, progesterone analogues, antagonists or
agonists can also be used, depending on the desired
clinical effects. Analogous treatment can be
administered to preovulatory women to regulate
follicular FFN production. Similarly, paracrine and/or
hormonal manipulation can regulate seminal vesicle FFN
production in the male. The importance of the protein
FFN in reproduction has now been recognized; thus
regulating the production of FFN by the method of this
invention, provides inter alia methods of contraception
' and fertility enhancement.
Further provided by this invention is a novel
method of decreasing free FFN in a mammal comprising
administering anti-FFN antibodies to said mammal.
Thus, by the method of this invention, anti-FFN
antibodies bind to FFN expressed by the mammal
effectively decreasing the amount of available or free
FFN regulating the abi~lity of the sperm to *ertilize
the ova or the conception to implant.
Fuxther provided by this invention is a novel
method of reducing the reproductive potential a mammal.
By the method of this invention, reproductive potential
is regulated by administering to the mammal a FFN
antigen in an amount sufficient to raise antibodies to
FFN, whereby, for example, the probability that sperm
will reach the fallopian tube, or fertilization will
occur or that an adherent blastocyst will success~ully
migrate to the uterus so as to implant.
WO92/1~199 2 0 7 5 3 0 0 PCT/US91/0~986
-17-
:'
Further provided by t:his invention is a novel
method of detecting the presence of aneuploid fetuses
(including trisomy 21) in the first and second
trimesters of pregnancy by assaying concentrations of
FFN in maternal plasma.
Further provided ~y this invention is a novel
method of imaging trophoblastic tissue, M~llerian tract
epithelia and endometreotic implants using magnetic
resonance imaging (MRI). The method comprises
administering to a mammal a labeled antibody specific
for FFN wherein the label is not radioactive so as to
avoid irradiating reproductive tissues. In a preferred
embodiment the label has paramagnetic properties such
that the label can be detected with MRI. The labeled
antibody is used in an amount sufficient to bind to
said trophoblastic tissue in an amount sufficient to
detect the label bound to said FFN antibody, whereby
the location of said trophoblastic tissue is dete-
rmined. The amount of labeled antibody used and the
port of entry vary according to the tissue to be
imaged. For instance, in the case of suspected ectopic
pregnancies or malignancies the port of entry may be by
direct infusion into the bloodstream.
Further provided by this invention is a method of
imaging ectopic or normally localized Mullerian
epithelial cells. The method comprises administering
to a mammal a labeled antibody, wherein the label is
not radioactive. In a preferred embodiment the label
has paramagnetic properties and is administered
systemica:Lly or infused into the cervix and thence
uterus, fallopian tubes and abdomen. For example,
endometriosis could be identified in patients without
resorting to operative diagno~tic procedures. In the
case of imaging of the uterus or other organs, direct
topical application may be appropriate. As the
concentratlon of the labeled antibody increases, the
- WO92/10199 2 ~ 7 5 3`0-~ PCT/US91/08986
-18-
greater the nonspecific binding becomes allowing
visualization of larger areas of tissue to determine
for instance organ abnormalities along the Mullerian
tract.
.
Brief Description of the Drawin~
Figure l is a set of graphs depicting
immunoreactive FFN (ir-onfFN), immunoreactive total FFN
and immunoreactive prolactin in the media of cultured
endometrial stromal cells treated as indicated in the
~; lO figure.
.~
Detailed Description of the Invention
"Fetal Fibronectin" (FFN) as used herein refers to
a class of proteins which bind to FDC-6 and are
produced by mammalian trophoblasts and Mullerian
epithelial cells, and are found in other mammalian
reproductive fluids, e.g., seminal fluid, vaginal
secretions and ovarian follicular fluids. The
immunological binding partner (FDC-6), although
previously thought not to bind to normal human adult
tissues, binds to FFN produced by certain normal
reproductive tissues.
"FDC-6" as used herein refers to the monoclonal
antibody defining the oncofetal structure of
fibronectin as disclosed in U.S. Patent No. 4,894,326.
25 U.S. 4,894,326 is incorporated by reference as if fully
set forth herein.
U.S. 4,894,326 disclose~ an IgGl monoclonal
antibody (FDC-6) which defines a fibronectin structure
or structures located between the "Hep-2" and the
"Fib-Z" domains in the COOH-tPrminal region of
fibronectins isolated from hepatoma, sarcoma, and fetal
~ibroblasts. It is disclssed that this antibody
indicates the presance of two classes of human
fibronectin. First, fibronectin from fetal connective
~
WO92/10199 2 ~ 7 5 ~ O O PCT/US91/08986
-19-
tissue, placenta, amniotic fluid, hepatoma and colon
; carcinoma as well as cell lines from fetal tissue
hepatomas and sarcoma was characterized by the presence
; of the structure recognized by FDC-6 and by a slightly
higher molecular weight. In contrast, fibronectin from
normal adult tissues and plasma was characterized by a
slightly lower molecular weight and lack of reactivity
with FDC-6, and is therefore devoid of the FDC-6
defined structure. The FDC-6 defined structure is
referred to as the "oncofetal structure", or epitope
and fibronectin containing this structure has been
called "oncofetal fibronectin." Fibronectin from most
normal adult tissues and plasma lacking the oncofetal
- structure, is characterized as "normal fibronectin."
Development of fibronectin from fetal to adult
forms is disclosed to be associated with loss of the
oncofetal structure defined by the FDC-6 antibody, and
oncogenic transformation is disclosed to be associated
with activation in synthesis of the oncofetal structure
defined by the FDC-6 antibody. Since the structure
defined by FDC-6 antibody expressed in oncofetal
fibronectin is a useful marker of cancer, it is
disclosed that the FDC-6 antibody and other antibodies
raised against the oncofetal fibronectin structure will
Z5 be useful for diagnosing human cancer and for
monitoring and implementing various cancer treatments.
The mono~lonal antibody FDC-6, a product of
hybridoma cell li~e ATCC No. HB9018, is suitable for
use in this invention, as a specific probe for a
variety of immunoassays such as, immunohistochemical
localization, ELISA assays and immunoblot analyses.
FDC-6 can be prepared substantially as described in
U.S. 4,894,326. Briefly, the murine monoclonal
antibody FOC-6 was established following immunization
of mice with fibronectin isolat2d from human hepatoma
cells. The hybridoma ATCC No. HB9018 was selected by
WO92/10199 0 ~ ~
2 ~ 7 ~ 3 PCT/US91/08986
-20-
positive reactivity of its antibody (FDC~6) with
- fibronectins from hepatoma, fibrosarcoma, and fetal
fibroblasts and by negative reactivity with fibronectin
from plasma. Other antibod:ies recognizing
trophoblastic and/or pregnancy-specific decidual ECM
components are expected to b~ useful in this invention.
It has now been found t:hat FFN is present at sites
of fertilization and implantation. Specifically, FFN
is present in both seminal and follicular fluid and
along the Mullerian tract in the apical cytoplasm and
surface of epithelial cells indicating a role in
fertilization. FFN has also been found in Mullerian,
epithelial and trophoblastic tissue indicating a role
in gamete, zygote and blastocyst transport as well as
implantation and maintenance of pregnancy.
FFN is produced by the method of this invention by
recombinant DNA techniques and subsequent protein
purification. A preferred method of protein
purification is affinity column chromatography using
FDC-6 as the affinity ligand. Recombinant DNA
technology allows for production of large quantities of
FFN as compared to endogenous production which in the
case of cultured trophoblast tissue FFN synthesis
decreases to minimal levels after 72 hours. The
protein thus obtained can be subject to modifications
such as altered glycosylation patterns or alterations
in the amino acid sequence that can be affected by a
variety of recombinant DNA techniques. Glycosylation
can be achieved by selection of a particular host cell
or by chemical modification after protein synthesis has
taken place. The recombinant DNA technique of
oligonucleotide-directed site-specific mutagenesis can
be used to alter various aspects of the protein. For
instance specific activity can be changed by amino acid
- 35 substitutions, deletions and insertions distant from
the FFN epitope which result in either enhanced or
WO92/10199 20~5300
PCT/US91/08986
-21-
reduced cell binding affinity respectively and thus
either promote or inhibit respectively fertilization,
implankation and placentation.
Generally, in order to create a site specific
mutation, the DNA sequence ~which is to be mutated must
first be cloned into a vector which is capable of being
replicated in both single-stranded and double-stranded
forms. Appropriate vectors include the M13 cloning
vectors which can be purified as either single-stranded
bacteriophage or as double-stranded plasmids. A
single-stranded oligodeoxyribonucleotide is then
synthesized to contain both the desired mutation and a
number of flanking nucleotide bases complementary to
and sufficient to ~nneal to the single-~tranded form of
the DNA sequence to be mutated. The synthetic
oligodeoxyribonucleotide is then annealed to the
single-stranded vector and a new strand of DNA is
- synthesized by DNA polymerase using the synthetic
oligodeoxyribonucleotide as the primer and the
single-stranded vector as the template. The new
double-stranded heteroduplex vector now contains one
non-mutated strand and one mutated strand. The
heteroduplex vector is then transfected into suitable
host cells which replicate and amplify the
double-stranded form of the vector. Due to
semiconservative replication, some of the transformed
progeny will contain vectors with only the mutant DNA,
so~e will contain vectors with only the non-mutant DNA
and others will contain both mutant and non-mutant
vectors. The cells containing only the mutant vectors
must then be selected from all others. The selection
procedure most often usd consists of radiolabeling the
synthetic oligodeoxyribonucleotide used in the
mutagenesic; and hybridizing it to the transformed cells
under-stringent conditions. Cells which contain the
mutant vector will retain the radiolabeled probe and
WO92/10199
PCT/US91/08986 -
207~300 22
,~;
become radioactive under such conditions. The cells
carrying the mutant can therefore be distinguished from
the non-mutant, non-radioactive cells. Cells
containing both mutant and non-mutant vectors must be
separated from those containing only mutant vectors
necessitating another round of selection. The mutation
must then be confirmed by sequencing. Further, since
the DNA polymerase may have incorporated other
mutations into the vector, the specifically mutated DNA
is usually recloned into a new vector which has not
been subject to the mutagenesis procedure.
In addition, the unique glycosylation pattern of
FFN may alter its susceptibility to proteases. Given
the anatomic distri~ution of FFN in pregnancy (see Ex.
` 15 l), such an enhanced susceptibility to proteolysis
could contribute to preterm rupture of the membranes or
premature placental separation (abruption). The
identification of specific proteases which selectively ~`
cleave FFN, allows appropriate pharmacological `
interventions in order to prevent these untoward
pregnancy events.
Recombinant DNA techniques also allow for
production of protein fragments and fusion proteins.
In a preferred embodiment, recombinant DNA encoding FFN
would be expressed in mammalian cell lines so as to
conserve the native FFN glycosylation patterns. It is
thought that the glycosylation pattern is responsible
for FFN acti~ity. Specifically, FFN identified by the
FDC-6 monoclonal antibody differs from adult
fibronectin moieties only by the presence of a
~-N-acetylgalactosamine at the threonine residue within
the hexapeptide Val-Thr-His-Pro-Gly-Tyr sequence
located within the IIICS region. Matsuura et al., 3'An
~-N-acetylgalactosaminylation at the Threonine Residue
of a Defined Peptide Sequence Creates the Oncofetal
! WO 92/10199 2 0 7 5 3 ~ O PCT/US91/08986
--23--
Peptide ~pitope in Human Fibronectin", J. Biol. Chem.,
264:10472-10476 (1989).
- Trophoblast derived EC~ components can be
extracted from the ECM deposited by trophoblasts in
culture. Similarly, endometrial glandular or
decidualized stromal - specific ECM components can be
deposited by glands and decidualized stromal cells in
culture. This trophoblastic-specific or glandular- or
decidual-specific ECM can be injected into mice or
other animals for the generation of an antibody
response and the preparation of many different
polyclonal or monoclonal antibodies to these ECM
components by known techniques. See e.g., "Antibodies,
A Laboratory Manual," Cold Spring Harbor (1988), which
manual is incorporated by reference as if fully set
forth herein.
Once anti-trophoblast- or decidualized stromal- or
glandular-ECM-antibodies are generated, as described
above, those antibodies which specifically bind to
trophoblast- specific ECM or decidualized stromal- or
glandular-specific ECM, but not other adult ECM
components, can be identified.
Measurement of secreted FFN in reproductive fluids
is expected to be an important diagnostic test for a
variety of conditions. Males and females with abnormal
levels of FFN diagnosed by these methods can then be
candidates for FFN therapy or pharmacologic
~anipulations to regulate FFN synthesis, as described
in detail below.
In another aspect of this invention, a method of
diagnosing the fertilization potential of semen,
post-coital or follicular fluid samples from a mammal
to determine the relative likelihood that semen from
that mammal is capable of fertilizing an ovum from a
~5 female mam~al of the same species, comprising the steps
of pro~iding at least one sample of semen or
-- WO92/lO199 2 a 7 5 3 b o Pcr,us~l,089~6,
-24-
cervico-vaginal secretion or ovarian pre-ovulation
~ follicular fluid; and assaying the sample for FFN, as
- described above, whereby a relative change in the
concentration of FFN in said semen, post-coital or
follicular fluid sample as compared to normal levels of
FFN in fertile mammals of the same species is
indicative of an altered fertilization potential.
The amount of FFN present in the sample as
compared with the quantity of FFN in a fertile control
semen, post-coital or follicular sample can thus, e.g.,
be used to evaluate the fertilization potential of the
sample in cases of unexplained male or female factor ~-~
infertility. Additionally, males with abnormal levels
of seminal FFN or couples with abnormal cervico-vaginal
FFN levels as diagnosed by this method can then be
considered as candidates for FFN therapy or
pharmacologic manipulations to modulate FFN synthesis,
as described in detail below. Further, the
information provided by the method of this invention is
useful in selecting semen and follicular samples more
suitable for assisted reproduction technologies, e.g.,
selecting a semen sample with altered FFN levels for
use in an in vitro fertilization procedure.
The methods of this invention, wherein FFN is
assayed or imaged in reproductive fluids of
non-pregnant females and semen and cervico-vaginal
mucus, are then generally expected to be useful as
diagnostic screens for mammals suspected of being
infertile. Further, these assays or imaging techniques
have important applicability for mammals undergoing
a~sisted reproductive technologies in more precisely
optimizing attempted fertilization. These ~ethods of
detecting the presence of FFN in endometrial,
fallopian, cervical, vaginal, or other fluids of
non-pregnant females permit optimizing transfer of
sperm or fertilized or unfertilized ova to the uterine
.
'. ,.
W~92/~0199 2 0 7 5 3 0 ~ PCT/US9l/08~86
-25-
cavity or fallopian tube as judged by FFN levels in
normal controls. Such assays or imaging techniques are
expected to be easy, non-invasive ways to assess levels
of FFN in mammals.
Further provided by this invention is a method to
image potentially pathological trophoblastic tissue as
well as physiological or pathological M~llerian tract
epithelium in a mammal. By the method of this
invention, a mammal is administered an effective amount
of an antibody specific for FFN, whereby the antibody
is conjugated to a non radioactive label. The labelled
antibody is introduced into the mammal either by direct
- injection into the blood stream or by topical
application such as douches or rinses. The labelled
~- 15 antibody couples to FFN and is imaged by an appropriate
medical imaging device such as MRI. The use of radi-
olabeled antibody subjects the patient's ov~ries,
fallopian, endocervical and perhaps normal intrauterine `
pregnancies to potentially damaging levels of
20 radioactivity. Thus, nonradiolabeled antibodies are
t the method of choice for imaging and treating tissue
expressing FFN. In a preferred embodiment, MRI is used
as the imaging method.
Conveniently, FDC-6 is a suitable antibody for use
25 in this invention. Other FFN antibodies prepared as
described above are also expected to be suitable for
use in the invention.
Examples of trophoblastic tissue to be Iocalized
include but are not limited to ectopic pregnancies,
30 placenta previa, other placental abnormalities and
metastatic gestational trophoblastic disease and
primary ovarian choriocarcinoma and other germ line
tumors producing FFN as confirmed by evaluation of
these malignant tissues. In addition, endometriosis or
35 M~llerian tract anomalies can be similarly imaged.
WO92/10199 2 0 7 ~ 3 Q ~ PCT/US91/0~86 --
-26-
Suitable labels include but are not limited to
paramagnetic labels such as microspheres with
paramagnetic impurities or ]proteins with paramagnetic
ligands.
- 5 In another aspect of this invention, infertility
and recurrent pregnancy loss screening tests are
provided. At least one bodily fluid or cell type from
a mammal suspected of being infertile or of having
recurrent abortions is assayed for the presence of FFN
autoantibodies.
Conveniently, the presence of autoantibodies that
bind to FFN can be assayed by ~asic enzyme-linked
immunoassay (ELISA) (where purified FFN is immobilized
on a plastic surface, and the subject's bodily fluid,
e.g., serum, plasma, cervicouterine secretion, or
seminal fluid, is applied to these wells in various
dilutions), or immunoblot techniques. A positive test
would occur by using as a marker a secondary anti-human
(IgM, IgA and IgG antigen) antibody, which would only
bind in the assay if the patient's bodily fluids
contained an anti-FFN autoantibody.
Bodily fluids expected to be useful include but
are not limited to plasma, serum, semen, follicular
fluids, (peritoneal fluids) cervico-vaginal secretions
and cervicouterine aspirates. Generally, any bodily
fluid or cell type associated with FFN in a fertile
control are believed to be useful.
Certain causes of previously unexplained
in~ertility are believed due to an autoimmune process ~`
leading to endogenous production of anti-FFN
antibodies. In one aspect, this autoimmune process is
believed to prevent the proper biological function o~
FFN, causiny sperm aggregation, impeding fertilization
and/or implantation or gamete transport and playing a
role in repetitive miscarriage later in pregnancy.
- W092/10l99
-27- PCT/US91/~8986
The first step in treating mammals with autoimmune
infertility is a screening method to identify this
group. Once identified, these patients might benefit
from specific treatments used for patients with
autoimmune diseases, such as immunoglobulin or
immunosuppressive therapy.
In another aspect of this invention, a method of
`~ altering the probability that gametes will fertilize or
that a conceptus will become implanted in a mammalian
uterus is provided. By the method of this invention,
FFN is infused into the uterine cavlty prior to coitus
or about the time the uterus will be contacted with a
~` conceptus, said infusing introducing a sufficient
~, amount of FFN onto the surface of the uterine cavity to
alter the probability that fertilization or
implantation will occur.
To decrease the probability of implantation, the
infusion will be given at the time a uterus will be
contacted with a conceptus e.g., at about the time of
presumed implantation during a natural cycle of
conception. Further, agents found to increase
endometrial glandular FFN synthesis can be employed ,
(vide infra).
FFN could be infused in a variety of ways as are
other agents for introduction into the uterus. For
example, FFN could be in a dissolved form, either in
solution, within a gel, or in a slow release capsule to
allow for an appropriate time-dependent concentration
in the uterine cavity.
An effective amount of FFN is that amou~t of FFN
which when introduced onto the surfa~e of the uterine
cavity alters the probability that gamete fertilization
or implantation of a conceptus will occur.
Concentrations of FFN in the range from about 0.1 ~g/ml
to about 1 mg/ml are expected to be useful, since FFN
W~92/10199 2~7~ ~ ~
PCr/US91/08~6f.-
-28-
concentrations in human re]productive fluids fall in ~-
this range.
Since FFN is also present in sperm, semen and
follicular fluid, it is be:Lieved to play a critical
role in fertilization. Therefore therapeutic regime~s
utilizing FFN, e.g., the addition of FFN to a sample of
semen intended for use in artificial insemination or
into the vagina at the time of coitus, are believed
applicable in helping to overcome certain male or
female infertility problems due to insufficient FFN
levels. Prior to/or following coitus FFN could be
administered directly into the endometrial cavity to
promote sperm migration or subsequent fertilization.
Literature is known describing intrauterine
infusions, gels or sponges for the t-eatment of a
variety of conditions. It has been shown that the
endocrine function of an ovary could be markedly
changed by an intrauterine infusion. Helmer et al.
"Intrauterine Infusion of Highly Enriched Bovine
Trophoblast Protein-1 Complex Exerts an Antiluteolytic
Effect to Extend Corpus Luteum Life Span in Cyclic
Cattle," J. Reprod. Fertil., 87:89-101 (1989). It has
besn shown that rat uteri which received an ;;
intrauterine injection of luteinizing releasing hormone
had a significantly increased rate of impl~ntation
compared to uteri which had no injection. Jones,
"Blastocyst Attachment in the Ovariectomized Rat
Treated With an Intrauterine Injection of Luteinizing
Hormone-releasing Hormone (LRH) r ~I Acta Endocrinol.
(Copenh), 103:266-8 (1983). In addition to ~he use of
solutions, there are references citing use of gels
which are instilled intracervically to facilitate labor
and delivery. See e.g., Ekman et al., "Intracervical
Instillation of PGE2-gel in Patients with Missed
Abortion or Intrauterine Fetal Death," Arch Gynecol.,
233241-245 (1983). Finally, an intrauterine vehicle
WO92/10199 2 ~ 7 5 3 U O ~ P~T/US91/08~86
-29-
such as a slow release capsule either similar to those
currently existing on the m;rket or modified to
~ facilitate slower release O:e a pharmacologic agent
; which might either enhance or decrease the synthesis of
FFN could be utilized.
Other compounds includ:ing putative
stromal-epithelial paracrine effectors such as TGF-~,
or ECM components are expected to be useful for
enhancing FFN synthesis in mammalian Mullerian
epithelium including endometrial glands. Endometrial
glands in culture, within six days constitutively
produce FFN de novo, while administration of estradiol
and progesterone inhibits synthesis. Therefore,
putative paracrine effectors and those compounds in
plasma, ECM extracts, or maternal decidua which induce
or inhibit FFN synthesis can be readily identified in
the in vitro endometrium culture assay described
herein. These may be employed in the method of this
invention in an amount sufficient to augment or alter
FFN synthesis in a mammal. Such compounds are expected
to include steroids, peptides, and glycopeptide
hormones, gonadotropins, growth factors, cytokines,
antibodies, as well as portions of ECM proteins
including other fibronectins, laminin, collagen types I
and IV, vitrone~tin, or proteoglycans.
FFN synthesis can be modified in a mammal about
the time the uterus of the mammal will be contacted
with a conceptus whereby altering the probability that
the concept:us will implant in the uterus.
In another aspect of this invention, a method of
regulating FFN synthesis in a mammal is provided
comprising aclministering ~o said mammal a
pharmaceutically acceptable compound which regulates
FFN synthes;is by mam~alian M~llerian epithelial cells
and other reproductive tract tissue in appropriate
amounts.
: . ,
WO92/101~9 2 0 7 5 3 Q ~ PCT/~'S91/0898~i -
-30-
Compounds which regulal:e FFN synthesis by
mammalian endometrial gland~; in culture can be selected
~ from the group consisting of estrogen, estrogen
- analogues and progesterone i.nhibitors and antagonists.
Modification of FFN synthesis in a mammal has a
variety of utilities. For example, an infertile mammal
determined to have a level of FFN different ~rom that
of a normal fertile control, may be a candidate for FFN
synthesis modification. In contrast, FFN can be
employed to achieve a method of contraception.
Pharmacologic manipulation to yield a m~thod of
contragestion can also be achieved, as previously
described, by introducing estrogen, estrogen analogues
and progesterone inhibitors and antagonists to the
15 local environment of the uterus through a suppository, `
gel or sponge or by direct systemic treatment. These
manipulations are expected to be effective both for
intrauterine and ectopic implantations, the majority o~
which are intratubal implantations. Pharmacologic
manipulation of FFN is expected to be a useful
non-surgical, non-invasive alternative to treatment of
chromosomally abnormal pregnancies incompatible with
life, missed abortions, incomplete abortions or for
other indications. Termination of tubal ectopic
pregnancies could occur by either systemic or direct
installation of the pharmacologic agent to the tubal
pregnancy via laparoscopic, ultrasonic, or retrograde
cervicouterine irrigation.
In another aspect of this invention, a method of
decreasing t:he reproductive potential of a mammal is
provided. E~y the method of this invention, the
reprodu~tive potential of the mammal is decreased by
administering to the mammal a F~N antigen with or
without adjuvant in an amount sufficient to raise
antibodies to FFN, whereby the probability that
:,
WO92/10199 2 0 7 ~ 3 ~ 0 ~ PCT/US91/08986
-31-
gametes will meet and fertilize or a conceptus will
migrate to the uterus is decreased.
Since FFN appears in the Mullerian tract
epithelia, follicular fluid and trophoblasts during
pregnancy, a novel method of permanent female
sterilization based on FFN immunization is provided.
In addition, since FFN is present in the seminal fluid,
it is believed that immunizing men or women against FFN
will induce antibodies specific for unique regions of
FFN, and that such antibodies should not have general
systemic effects. These antibodies would bind to FFN
secreted by the mammal e.g., Mullerian epithelia,
follicular or trophoblastic cells, as well as the
seminal fluid FFN. The presence of these antibodies
may cause sperm agglutination or adherence to Mullerian
epithelial cells, alter sperm motility, inhibit
ovulation, impair ovum transport, prevent
fertilization, and alter the potential for i~plantation
by the blastocyst or otherwise induce sterilization.
In addition, if a blastocyst were to initiate
implantation, the presence of these antibodies is
expected to prevent or impair attachment of the
developing trophoblast to the endometrial decidua, thus
ending the gestation.
Male or female mammals could be immunized against
FFN by using the whole molecule or just the IIICS
domain of FFN with the proper glycosylation moieties.
Generally the protein can be dissolved at between about
l to 50 ~g/ml in sterile saline or saline with 0.4 mg
aluminum hydroxide per ml as a vehicle. Generally 0.5
to l.O ml of the protein solution is injected
intramuscularly and then followed by booster injections
at one and 6-12 months after the initial immunization.
Such immunizations thus prevents antikodies from
developing against portions of FFN in common with
normal adult plasma or cellular fibronectin.
WO92/10199 2 0 ~ ~ 3 Q ~ PCT/US91/089~6 -
-32-
~7
There iC7i precedence for i~munizing patients and
animals against various products of pregnancy to induce
contraception. For example, 88 subjects which were
immunized with an hCG base vaccine have been
investigated. Kharat et al , "Analysis of Menstrual
Records of Women Immunized with Anti-hCG Vaccines
Inducing Antibodies Partially Cross-Reactive with hLH,"
Contraception 41:293-9 (1990). In animals, it has been
shown that antibodies made against pig zona pellucida
could induce contraception in mares. Liu et al.,
"Contraception in Mares Heteroimmunized with Pig Zonae
Pellucidae", J. Reprod. Fertil., 85:19-29 (1989).
Finally, in a study in dogs, it has been shown that
contraception could be induced by immunizing dogs
against gonadotropin releasing hormone. Gonzalez et
al., "Immunological Approaches to Contraception in
Dogs," J. Reprod. Fertil. Suppl. 39:189-198 (1989). -`
In another aspect of this invention, a method of
detecting the presence of chromosomally abnormal
conceptuses during mammalian gestation is providPd,
comprising measuring maternal plasma concentrations of
FFN from 6 to 20 weeks gestation. Approximately 5-10
ml of maternal blood is collected in a tube containing
EDTA to produce a final concentration of EDTA of at
least 5 mM. The plasma from this specimen is then
collected and assayed for FFN concentrations by
immunoassay or radioimmunoassay.
Given the elevated concentrations of FFN that the
inventor has measured in the media of primary
trophoblast cultures and detected in th~
placental-endometrial interface of normal pregnancies,
neonatal cord plasma and in the amniotic fluid and
plasma of healthy pregnant women, measurement of
maternal plasma FFN would be expected to reflect
synthesis ~y the trophoblast or fetal tissue. Given
that aneuploid trophoblasts have been shown to produce
WO 92/101~9 2 ~ 7 5 3 0 0 PCT/US91/08~86
--3 1_ .
abnormal quantities of HCG and other trophoblast
specific products, the production of FFN by aneuploid
conceptuses is anticipated to be markedly abnormal and
this abnormality is expected to be reflected in
maternal plasma FFN concentrations.
Example ~
Immunohistochemical Studies of Placentae,
Membranes and Beniqn and Maliqnant Reproductive
10 Tract Tissues obtained from Non-Preqnant Women
The inventor has identified FFN in the cervical
and vaginal secretions of patients with uncomplicated
- pregnancies prior to 21 weeks gestation.
Cervico-vaginal FFN was rarely identified in patients
with uncomplicated pregnancies between 21 and 37 weeks
gestation. However, when present at this gestational
age interval cervico-vaginal FFN was frequently
associated with preterm delivery. This latter finding
promoted the inventor to characterize the tissue source
of FFN during pregnancy. Placentae, decidua and
external membranes for immunohistochemical analysis
were obtained from term pregnancies, a hysterectomy
specimen and from elective first trimester abortions.
Tubal ectop~c preqnancy specimens with associate
uterine decidual specimens were also obtained. Fetal
tissue was obtained following midtrimester termin-
ations. In addition, to account the inventor's
observation of the occasional but consistent presence
of cervico--vaginal FFN in healthy non-pregnant
patients, the presence and the localization of proteins
recognized by FFN specific antibodies in
non-qestational reproductive tract tissues (benign and
neoplastic) was evaluated. Immunoperoxidase staining
was performed accordi~g to the manufacturer's
instructionis, using an avidin-biotin peroxidase
detection system (Vector Laboratories, Burlingame, CA~.
:
.
.,
. , ., . : i,, .. . . -.. .. . .: . . , , . :
WO92~10199
i P~/US91/0~9~6 , --
~ 2~300
.
Antisera were utilized at 4 ~g/ml for FDC-6, a product
of hybridoma cell line ATCC HB9018 (American Type
Culture Collection, Rockville, MD provided by Adeza
Biomedical, Sunnyvale, CA), and undiluted for the
negative rabbit and goat serum or control P3X63Ag8
; mouse myeloma cell line supernatant. (American Type
Culture Collection, Rockville, MD).
5 ~m sections from formalin on Bouin's-fixed and
paraffin embedded tissue were placed on glass slides
previously coated with a film of 1% poly-d-lysine,
30-70,000 molecular weight (Sigma), dried at
temperatures no greater than 60C and stored at room
temperature until u ed. Routine immunoperoxidase
staining was carried out.
The results showed that in pregnancy tissues,
there was diffuse immunohistochemical staining for
total fibronectin (tFN) throughout the decidua and
villous stroma as well as the syncytiotrophoblast
basement membrane and surface. There was highly
specific intense staining for FFN on the ECM of the
decidua basalis in proximity to extravillous anchoring
trophoblasts and trophoblast cell columns of the
pla~ental-decidual junction. This pattern of FFN
staining was seen in both term and first trimester
specimens and in the tubal-trophoblast junction of
ectopic pregnancies. The villi, decidua distant from
the utero- placental junction, uterine decidua from
ectopic gestations, and maternal blood vessels were
consistently negative for FFN. Although the source of
FFN in the tissues in which FFN is found is not clear,
likely possibilitles are that FFN is secreted by the
extravillous trophoblasts derived from the
cytotrophoblastic shelf located in the decidua basalis
and/or the decidual cells. In view of the acknowledged
importance of cell-cell interactions in regulating the
differentiat:ed phenotype of either or both cell types,
WO92/10199 7 5 3 ~ ~ PCT/US91/08986
the specific localization of FFN only to ECM in which
both extravillous cytotrophoblasts and decidual cells
are present suggests that paracrine interactions
between these two cell types regulates placental FFN
production.
Analyses of membran~ preparations revealed an
analogous distribution of FFN in the chorionic ECM.
This site would appear to be ontogenically analogous to
the localization of FFN in the placental-decidual
junction since the chorion is a remnant of the
trophoblastic shelf. Patients with severe chori-
oamnionitis displayed an apparent reduction in FFN
staining, perhaps as a consequence of proteolysis.
Again, there was diffuse staining for tFN throughout
the amnion and chorion ECM.
In fetal tissues, FFN was identified in the
cytoplasm of hepatocytes and renal collecting tubules
and transitional epithelia. The presence of FFN in
fetal kidneys may reflect excretion of a fibronectin
fragment bearing the oncofetal epitope or synthesis of
FFN in the genito-urinary tract. However, it is
unlikely that fetal urine is a source of the high
levels of FFN present in the amniotic fluid since
immunoassays of the urines of newborns failed to
display evidence of FFN, despite high (12 ~g/ml) cord
plasma concentrations.
The results obtained with Mullerian tract tissue
are shown in Table I which compares the
immunohistochemical distribution of FFN in
non-gestational reproductive tract tissue and
reproductive tract malignancies. In benign tis~ues FFN
was localized to the juxta-luminal cytoplasm and
secretions in some cervical, endometrial, and tubal
epithelial specimens. It was similàrly localized to
the epithelium of a mucinous cystadenoma of the ovary
(Table I). Thus juxta-luminal cytoplasmic localization
~092/10199 2 0 7~3 0V PCT/US91/OB98i6
-~6-
in Mullerian epithelia stands in contrast to the
findings in placentae and membranes where FFN was
restricted to the ECM.
TABLE I
DISTRIBUTION OF FFN IN REPRODUCTIVE TRACT TISSUES
A) Benign Tissues:
1) Normal vagina: no staining for FFN epitope,
however, diffuse stromal staining for total
fibronectin (tFN).
2) Normal cervix: positive discrete
juxta-luminal endocervical gland cytoplasmic
staining for FFN in some patients' specimens,
however diffuse stromal ECM staining for tFN.
3) Tubo-ovarian complex: no staining for FFN,
but diffuse staining of stromal ECM for tFN.
4) Normal fallopian tubes: positive staining for
FFN in the juxta-luminal cytoplasm of the
tubal epithelium in most patients' specimens
and diffuse staining of stromal ECM for tFN.
5) Normal premenopausal endometrium:
a) Proliferative Phase: No FFN staining,
but di*fuse staining of stromal ECM for
tFN.
b? Secretory Phase positive discrete
juxta-luminal endometrial gland
cytoplasmic staining, but only in
tissues obtained from some patients in
the early luteal pha~ie.
6) Menopausal endometrium: no staining for FFN,
diffuse staining of stromal ECM for tFN.
7) Endometriosis o~ the colon: intense speci~ic
staining ~or FFN in the juxta luminal
cytoplasm of the endometrial glands with
dif~use staining of stromal ECM and
histiocytes for tFN;
.
WO92/10199 2 0 7 5 3 ~ O pcr/us9l/o8986
-37-
B) Maliqnancies:
8) Endometrial carcinomas and endometrial
hyperplasia: no ~taining for FFN, diffuse
staining of stromal ECM for tFN.
9) Endometriod tumor of the ovary: no staining
of the tumor for FFN, however, the fallopian
tube in this specimen displayed an FFN
staining pattern similar to #4 with luminal
surface cytoplasmic staining of the tubal
epithelium, throughout the specimen there was
diffuse staining of stromal ECM for tFN.
lO) Squamous cervical carcinomas (moderate and
poorly differentiated) and an endocervical
adenocarcinoma: no staining for FFN, diffuse
staining of stromal ECM for tFN.
ll) Mixed M~llerian tumor of the uterus: no
staining for FFN, diffuse staining of stromal
ECM for tFN.
12) Mucinous cystadenoma of the ovary: intense
juxta-luminal staining of the epithelial cell
cytoplasm for FFN with diffuse staining of
both ovarian stromal ECM and a fibrin clot ~-~
for tFN.
I3) Ovarian lymphoma: no staining for FFN,
whereas a normal lymph node demonstrated
trace intracytoplasmic staining for FFN an~
tFN in macrophages the (the latter may
represent a peroxidase artifact). ~ -
14) Dysgerminoma: no staining for FFN.
l~) A Papillary serous cystadenocarcinoma: no
staining for FFN but diffuse staining of
stromal ECM for tFN.
~ '.
:
' ~
~:
:
W092/10l99 2 07 ~ 3 ~ o
PCT/US91/08986
-38-
Example ?
Human Trophoblasts in Cult re
Synthesize Fetal Fibronectin De Novo
5ELISA assays were performed to measure the
concentration of FFN in trophoblast media and cell
extract. ELISA assays were performed using ROMC~E~K~
according to the manufacturer's instructions (Adeza
~iomedicals, Sunnyvale, CA).
10Human cytotrophoblasts were purified as previously
described~ Kliman et al., "Purification,
Characterization and in vitro Differentiation of
Cytotrophoblasts from Human Term Placentae",
Endocrinol. 118: 1567-1582 (1986). The
cytotrophoblasts were cultured in Dulbecco's Modified
Eagles' Medium (DMEM) containing 25 mM ~lucose and
25 mM HEPES (DMEMHG) supplemented with gentamicin
(50 ~g/ml), glutamine (4 mM), and 20% (v/v)
heat-inactivated fetal calf serum. For preparation of
cell extracts, cells were washed with phosphate
buffered saline, scraped from the culture dish, and
total cellular protein was extracted with an
S~S-didechoate buffer.
ELISA assay results showed that cell protein
extract from freshly purified villous cytotrophoblasts
contained barely detectable FFN on ELISA (<50 ng/mg
cell protein), in agreement with the negati~e villous
staining for FFN in the placenta. A small cruantity of
FFN was present intracellularly a~ter 24 h, (125 ng/mg
cell protein), suggesting that FFN synthesis had been
initiated by the cultured cells. After 96 h,
trophoblasts contained 18 fold more FFN (2200 ng/mg
cell protein) than the 24 hour cells, representing 0.2%
of total trophoblast cell protein. Thus
cytotrophoblasts, while not synthesizing FFN in vivc?,
are induced in culture to produce significant FFN. It
WO92/10199 2 0 7 ~ 3 O ~ PCT/US91/08986
was also determined that cultured trophoblasts secrete
FFN. Although very little FFN could be measured in
conditioned media from the first 24 h of culture,
during the time interval from 24 to 48 h the media
concentration of FFN averaged 4.5 ~g/ml.
This result indicated trophoblast secretion of
newly synthesized FFN into the culture media. Based on
ELISA, lO0 percent of trophoblast fibronectin FFN
contains the oncofetal domain, and is therefore
reactive with FDC-6.
Example 3
Immunoassay Quantitation of FFN
Concentrations in Sperm. Follicular Fluid
Samples of semen obtained from specimens sent to
the Mount Sinai Perinatal Laboratory for routine semen
analysis were analyzed for FFN concentrations by
immunoassay according to the manufacturer's
instructions (Romcheck~ Adeza Biomedical). Very high
concentrations were found in semen, 77 - 648 ~g/ml,
including the semen of an aspermic vasectomized male.
Subsequent studies suggested tight binding of FFN to
sperm since washed sperm retained relatively high
concentrations of FFN (95 - 1040 ng/ml). Residual
follirular fluid obtained from mature follicles
aspirated under ultrasound guidance during (IVF) oocyte
retrieval was assayed for FFN and total fibronectin.
Concentrations of 2.9 ~g/ml and 24.6 ~g/ml for FYN and
total fibronectin, respectively, were observed. Thus
follicular fluid from mature follicles appear to
contain bot:h adult and oncofetal forms.
:'.
::
' ': ,' ~ : ' .' . ' : :.
WO92/10199 2~7~3~ - P~ S91/~89~6,~-
-40-
Example 4
Analyses of FFN Secretion by Primary Cultures
of_Endometrial Epithelial and Stromal Cells
Glands were isolated f;rom biopsies of
proliferative endometrium and cultured on collagen I
under polarizing conditions according to a modification
of the method described by Schatz et al, "Culture of
Endometrial Cells Under Polarizing Conditions",
i 10 Differentiation, 42:184-190 (1990) while stroma was
isolated from proliferative and secretory phase
biopsies and cultured on plastic. The original tissue
specimen was formalin fixed, paraffin embedded,
sectioned and stained with hematoxylin-eosin and
assigned a menstrual date according to the criteria of
Noyes et al., "Dating the Endometrial Biopsy", Fertil.
Steril. (1950) 1:3 2.
Stromal cells were derived from a specimen of
secretory endometrium and grown to confluence on
plastic culture dishes in Basal Media (BM: Serum free
DMEM ~ Ham's F12, 1/1) + 10% stripped calf serum (SCS).
At confluence within 15 days of plating, the
experimental period was initiated by adding in BM + 10
SCS, either lOnM estradiol (E2), 1 ~M
Medroxyprogesterone acetate (MPA), E2 + MPA, or BM +
10% SCS with 0.1% ethanol (the vehicle control). The
cultures were maintained in a 37C 95% air 5% C02
incubator. The medium was replenished every 4 - 5 days
with the coxresponding experimental or control medium. -
The conditioned medium was concentrated and assayed for
levels of Prolactin (PRL~, (Amersham RIA) total
fibronectin (Total Fibronectin Assay0, Ade~a
8iomedical) and (Romcheck~, Adeza Biomedical) FFNo At
17 days, media were collected and the cells harvested
by scraping with rubber "policeman" into ice cold Hanks
Balanced Sa-lt Solution (HBSS) containing a cocktail of
v ~J
WO92/10199 -41- PCT/US91/08986
protease inhibitors (PI: soybean trypsin inhibitor,
pepstatin, leupeptin, aprotinin, PMSF, all at a final
concentration of lO ~g/ml). The cell pellets were
resuspended in HBSS + PI, and divided into aliquots to
measure levels of protein by Modified Bradford Assay
(Bio-Rad), total fibronectin and FFN.
Figure 1 shows the effects of E2, MPA and E2 + MPA
on levels of PRL, fibronectin and FFN in the
conditioned medium of primary stromal cell cultures,
and normalized to protein content in the harvested
cells. Virtually no PRL (Fig. lA) was detected in
control or E2-supplemented medium, whereas MPA elicited
only a small increase in PRL levels. However, the
combination of E2 + MPA produced a marked synergism in
PRL output by the 5-9 days collection period with
further increases evident by 13-17 days. This
elevation in PRL production appears to be
progesterone-regulated, while the action of E2 likely
reflects maintenance or enhancement of progesterone
receptors. Both total fibronectin (Fig. lB) and FFN
tFig. lC) were readily detected in the stromal cell
conditioned medium. Estradiol produced the most
striking effect of any of the treatments, reducing
levels of both total fi~ronectin and FFN to about half
that of the corresponding controls by 5 day~, and to
about 90% by 13 days. This result confirms that the
stromal cells possess functional estrogen receptors.
Although MPA also inhibited outputs of fibronectin and
FFN, adding E2 + ~PA to the cultures unaccountably
elevated levels of total fibronectin to control values,
but inhibited FFN midway between E2 and MPA alone. In
contrast to the significant levals of fibronectin and
FFN measured in the conditioned madium, neither protein
was detected in the harvested cell pellets.
.'`:
''"''''''''' '
-. ~ .
W092/10199 2~753:~Q PCT/US91/08986
-~2-
Isolated endometrial glands were distributed among
16 Millicell (Millipore) chambers containing collagen I
gels in BM + ITS++2% SCS and~ grown under polarizing
conditions. These were divided into control (+ 0.1%
ethanol) and experimental (lOnM E2 + l ~M MPA) groups
and, placed in a 37C incubator. After a 7 day
incubation, the media were collected from all 16
chambers and pooled into control subgroups A and B, and
experimental groups A and B, then concentrated and
frozen. The cells in the A subgroups were harvested by
collagenase I and the cell pellets stored frozen,
whereas in the B subgroups, the medium was replenished
with corresponding fresh medium and the cultures
returned to the incubator. After an additional 7 day
incubation, the medium from the B subgroup was
collected and the cells harvested as above. The
collected media were analyzed for FFN and total
fibronectin contant and the cell pellets were assayed
for protein content (after subtracting collagen blank
values (<8% of total cell protein)), FFN and total
fibronectin content.
In contrast to the results in stromal cell
cultures (proliferative phase endometrial epithelial
cells cultured under polarizing conditions on collagen
(see Table II) release of FFN into the media and it
appears that; l) FFN îs the primary and perhaps only
fibronectin released by these primary gland cultures;
and that 2) estradiol plus MPA inhibits this production
(or shuts off FFN synthesis completely).
v~ ~J U~
WO~2/1~199 PCT/US91/08986
-43-
TABLE II
PRODUCTION OF FFN BY PRIMARY
ENDOMETRIAL GLANDULAR_CULTURES OBTAINED
FROM THE PROL~FERATIVE PHASE
Experimental Conditions
~in ng/ml of media/mg cell protein)
._____ - ~ _ . . _
Control E2 + MPA
. . _
Days/Culture [FFN] [tFN] [FFN] [tFN] 0-6*
: _ .... .
(A) 0-7* 722 587 < 5 3.2 j~.
. . . .__. . _ . ._.
(B) 0-7* 864 602 < 5 5.7
. .... ___ . . ..
(B) 7-12* 1893 1836 < 5 _ 6 0 _ : ~
. Uncond Medla** 5 0 < 5 0.0 ~ .
~, Thus, the endometrial glandular epithelial cells
from proliferative phase human endometrium release FFN,
but virtually no adult fibronectin, into the media.
This xelease was completely inhibited by E2 + MPA. :.
'
,~
'
- ' '
.
