Note: Descriptions are shown in the official language in which they were submitted.
13399.~ 3
ISOFERRITIN AS A MARKER FOR PATHOLOGICAL PREGNANCY
TABLE OF CONTENTS
Page
l. Field of the Invention............................... 3
2. Background of the Invention.......................... 3
10 2.l. Ferritin..... ~.................................. 3
2.l.l. Placental Isoferritin.................. 4
- 2.2. Pathological Pregnancy......................... 5
2.2.l. Clinical Symptoms of Pathological
Pregnancies........................... 6
2.2.2. Immunosuppression and Pregnancy........ 7
3. Summary of the Invention............................. 9
4. Brief Description of the Figures..................... ll
5. Detailed Description of the Invention................ 12
5.l. Immunoassays........... ,......................................... l5
:20 5.2. Ferritin-Bearing Lymphocytes................... l9
5.3. Immunosuppression Therapy...................... 22
5.3.l. PLF Therapy and Pregnancy.............. 22
5.3.2. PLF and Transplants.................... 24
5.3.3. Therapeutic Regimens................... 25
25 6. Example......................................................... 27
6.l. Preparation of Monoclonal Antibodies...................... 27
6.l.l. Preparation of Oncofetal Ferritin.. 27
6.l.2. Preparation of Non-PLF Specific
Hybridomas....................................... 28
:30 6.l.3. Preparation of PLF-Specific
Hybridomas....................................... 32
:35
1~399~j~
6.2. PLF Levels and Preterm Deliveries.............. 33
6.3. P~F Levels and Toxemia of Pregnancy............ 39
-3~ 133995~
l. FIELD OF THE INVENTION
There is currently a widespread need for simple
yet accurate means of diagnosis for a number of conditions
which affect pregnant women. It is often difficult, if not
impossible, to determine in advance which pregnant women are
at risk for certain conditions which may prevent carrying a
chilcl to term, and/or which may endanger the life of both
mother and fetus. Early detection of problem pregnancies
enables rapid implementation of necessary therapy or
prophylactic measures. The present invention now provides a
serum marker which may be used for the diagnosis and
detection of several types of pathological pregnancies. The
marke!r which is indicative in all these conditions is
placental isoferritin (PLF; or oncofetal or embryonic
ferritin). Accurate detection of the presence or absence of
the marker is enabled by the discovery of a PLF-specific
monoclonal antibody, as well as a broadly cross-reacting
monoclonal antibody to isoferritin. The existence of these
antibodies has made possible the construction of
immunoassays which can accurately and rapidly facilitate
diagnosis of the aforementioned disease states.
2. BACKGROUND OF THE INVENTION
2.1. FERRITIN
Iron is known to be an essential element of the
makeup of every living organism, but may also become toxic
at physiological pH values by virtue of its tending to
oxidize, hydrolyze and precipitate as insoluble ferric oxide
polymers. The protein ferritin, found in all living cells,
is the body's means for ensuring that iron toxicity does not
occur. Ferritin functions by storing iron in the cells in a
soluble and readily available form. The iron stored in
cells may then be mobilized whenever needed by the body, for
example, for erythropoiesis.
-4- 13 3995~l
The name ~ferritin~ actually encompasses a num~er
of individual isomeric forms which are characteristic of
different tissue types. Each isoferritin has 24 subunits of
two distinct types, namely light subunits (L) and heavy
subunits (H~. These subunits differ in molecular weight,
the light subunit being about 18 kDa, and the heavy subunit
about 19-21 kDa. The isoferritins extracted from different
tissues or organs typically exhibit different isoelectric
points, with the isoelectric focusing pattern of human
tissues forming a continuous spectrum; those tissues
associated with high iron storage have ferritins at the
basic end of the spectrum (e.g. spleen and liver), while
iron poor tissues, (e.g. heart and placenta) and malignant
cells have acidic ferritins. (Drysdale, Ciba Found. Symp.
51:41, 1977). The difference in isoelectric point appears
to be related to the different distribution of light and
heavy subunits in each type. Specifically, heavy subunit-
rich ferritins are relatively acidic, and light chain rich
ferritins are relatively basic (Cosell, et al., in Ferritins
and Isoferritins as Biochemical Markers, p. 49-65, 1984,
Elsevier). Current studies indicate that the H and L
subunits are encoded by a complex group of genes, therefore
suggesting that there is an even greater heterogeneity of
ferritin molecules than had previously been expected.
- 25
2.1.1. PLACENTAL ISOFERRITIN
A specific type of acidic isoferritin has been
shown to be characteristic of neoplastic cells and placental
cells (Drysdale and Singer, Cancer Res. 44:3352, 1974).
This protein is also known as oncofetal ferritin or
placental isoferritin (PLF). Human placental ferritin has
been shown to be composed predominantly of a single subunit
type comigrating with a liver ferritin standard on SDS-PAGE
(Bro~n et al. Biochem. J. 182:763, 1979). However, an
immunoradiometric assay performed with anti-human spleen
_5_ 1339958
ferritin has shown tissue specific antigenicity for PLF.
(Brown et al., supra). A three subunit structure has been
revealed for PLF (Moroz et al. G.I. Pat. Clin. 1:17-23,
1986). In addition to the L and H subunits characteristic
of all ferritins, there is also a high molecular weight (43
kDa) subunit which appears to be unigue for human placenta,
and thus provides a potential site for identification of the
placental isoferritin molecule as distinguished from any
other type of ferritin.
2.2. PATHOLOGICAL PREGNANCY
While the majority of women who become pregnant
have no substantial difficulty in carrying a child to term,
there are certain conditions which commonly arise in
connection with problem pregnancies. Certain visible
abnormalities, such as bleeding, may be symptomatic of more
serious problems, but may also be a mere irregularity which
does not develop into a condition which threatens the normal
pattern of gestation. On the other hand, certain conditions
such as spontaneous abortion, may occur without any warning,
resulting in premature termination of the pregnancy and
often in the death of the fetus. Although some women, in
successive pregnancies, show a pattern of difficulty with
carrying a fetus to term, and can therefore be treated
accordingly in advance, the development of problems is not
uniformly predictable, particularly in a first pregnancy.
The early identification of a woman in a high risk category
would facilitate prescription of an appropriate treatment
regimen, thereby increasing the chances of full-term birth
of a healthy baby, as well as decreasing the possible
dangers to the mother's health. Some of the more commonly
occurring problems during pregnancy are toxemia, premature
contractions, premature delivery, missed abortion, and
spontaneous abortions or miscarriages.
- -6 1~39~5~
2.2.1. CLINICAL SYMPTOMS OF PATHOLOGICAL PREGNANCIES
A substantial number of pregnant women, perhaps
as h:igh as 7%, experience rapid weight gain, edema, and
elevation of blood pressure at some time during their
pregnancy. This condition, known as toxemia, results in a
decrease in blood flow and in glomerular filtration, a
situation which is completely the reverse of what is
observed in a normal pregnancy. Because of the reduction in
the glomerular filtration rate, the major problem is water
retention. A particularly severe form of toxemia,
ecla~psia, is characterized by extreme vascular spasticity
throughout the body, and clonic convulsion, followed by
great:ly decreased kidney output, hypertension and a general
toxic condition. Toxemia of course threatens the health of
the ietus, and in its more severe forms, may also threaten
the mother's life.
Other problems may also arise during pregnancy
which can be an indication of risk of premature delivery, or
of spontaneous abortions. Many women may experience
premature contractions at an early stage (as early as 16
weeks of gestation) during a pregnancy. The correlation of
this symptom with a high risk of preterm delivery has not
been clearly established. Similarly, abnormal bleeding can
fre~lently occur at any time during pregnancy. This symptom
may be an indication of an imminent miscarriage or a missed
abort:ion; on the other hand, in about 50~ of the cases it
does not develop into anything more serious. Clearly, an
early~ evaluation of such conditions, with some predictive
significance as to risk of premature delivery or spontaneous
abortion, would be of tremendous value to the clinician in
both a hospital and office setting.
~ _7_ 1339~
2.2.2. IMMUNOSUPPRESSION AND PREGNANCY
It has frequently been noted that something of an
anomaly exists in a pregnancy being carried to full term: a
fetal trophoblast which implants in the mother's uterus
carries major histocompatibility antigens from both the
mother and the father, and thus, except in rare
circum~tances, presents antigens to the maternal circulation
(and immune system) which must, under normal circumstances,
recognize the father's antigens as foreign. The embryo and
fetus therefore stand in the position of an allograft, and,
as yet, there has beerl no fully satisfactory explanation as
- to why the mother does not reject the fetus in much the same
manner as a foreign skin graft would be rejected. It is
clear that the embryo is protected in some way from the
action of the mother's immune system. A number of
mechanisms have been postulated to explain this phenomenon;
among them are low levels or absence of Class II antigens on
the syncytiotrophoblast which is in closest contact with the
placenta, making it more likely that paternal Class I
antigens will induce tolerance rather than a cytotoxic
response; protection of trophoblast cells against cytotoxic
lymphocytes by a barrier of negatively-charged
mucopolysaccharide, or a protective effect of the physical
barrier provided by the placenta.
Currently, one of the more popular theories of
fetal protection is the suggestion of a general suppression
of the mother's immune response. The placenta is known to
secrete a variety of different products into the maternal
circulation; these products include human chorionic
gonadotropin (HCG) as well as substance known to have an
immunoregulatory effect, such as estrogen, progesterone,
corticosteroids, and pregnancy-associated growth factors
(Caldwell et al. J. Immunol. 115:1249, 1975; Fabri~ et al.
Clin. Exp. Immunol. 28:306 r 1977; Baer et al. Ciba
Foundation Symp. Excerpts Medica 64:293, 1979; Suteri et al.
-8- 1~39958
Ann. NY Acad. Sci 286:384, 1977; Monse et al. J. Immunol.
128:218, 1982). Placental isoferritin is also known to be
secreted into the maternal circulation by the placenta
(Brown et al. Biochem. J. 182:763, 1979). It has been
suggested that some substance secreted by the placenta early
in development may be responsible for inducing a general
immunosuppression in the mother, thereby preventing
rejection of the embryo and allowing normal full-term
delivery.
In the same vein, a substantial number of
irregularities which can occur during pregnancy also remain
essentially unexplained as to cause. It has been suggested
that some, or many of these problems may be associated with
a mo1:her's mounting an immunological response to the
prese~nce of foreign antigens on the fetus. In other words,
in problem pregnancies, it is possible that the difficulties
arise because of the failure of this postulated temporary
immunosuppression to develop. The mother then reacts
normally in response to a foreign stimulus, and various
degrees of rejection of the fetus may then occur.
Evidence obtained in connection with the present
invention provides the basis for suggestion of a mechanism
for at least part of the physiological and immunological
basis for tolerance (or non-tolerance) of the ~'fetal
allograft~ in pregnant women. It has now been discovered
that a significant positive correlation exists between high
seruDl PLF levels in pregnant women, and successful full term
deli~rery. Similarly, abnormally low levels of PLF have been
shown to be consistently associated with premature delivery,
toxemia, and other pregnancy-related pathologies. Thus, low
levels or absence of PLF in pregnant women can 6erve as
marker for a potentially high risk pregnancy; detection of
this state is ideally achieved by use of a PLF-specific
monoclonal antibody in a novel immunoassay system, whereby
early~ detection and diagnosis of a pathological condition
- -9- 13399~8
can be made. This method is particularly well adapted for
use in monitoring women after their first trimester of
pregnancy, since levels of PLF, even in normal women, may be
too low to detect prior to that time. For diagnosis of
potentia~ problems in first trimester, however, it has been
discovered that detection of a relatively high level of
ferritin bearing lymphocytes (FBL's) indicative of the
occurrence of immunosuppression. Thus, pregnant women in
the first trimester having less than about 5% of FBLs can be
ident:ified as potentially being at risk for a problem
pregnancy.
The role of PLF in pregnancy does not appear to
be limited to its use as a convenient marker however; there
is also ample evidence to show that PLF plays an active role
in thle immunosuppression which appears to be a necessary
event to support a normal pregnancy. Based on the
obser~ation of an apparent cause-and-effect relationship
between PLF levels and immunosuppression, there is thus
provided a means for treatment and prevention of actual and
potentially pathological pregnancies, as well as a means for
inducing immunosuppressions in those clinical situations in
which production of a hyporesponsive immune state is
desirable, e.g., in organ or tissue transplantation, to
prevent rejection of the transplant.
- 25
3. SUMMARY OF THE INVENTION
The present invention relates to a method of
detecting a pathological pregnancy which comprises
contacting serum of a pregnant female with a first antibody
capable of reacting with placental isoferritin (PLF) or
normal serum ferritin, and a second antibody having
specificity for placental isoferritin, said second antibody
being linked to a reporter molecule capable of producing a
detectable signal; allowing time sufficient for an
antibody-PLF-antibody complex to form; observing the
133g95~
- -
presence or absence of the detectable signal; and
c~uant:ifying the result to determine the amount of placental
isoferritin in the serum. Lower than normal amounts of
placental isoferritin indicate risk of an abnormal
pregnancy. In a preferred embodiment, the antibodies are
both monoclonal antibodies. Examples of such useful
monoc:lonal antibodies are CM-G-8, a monoclonal antibody
which is capable of reacting with any type of ferritin, and
CM-H--9, a monoclonal antibody which is capable of reacting
only with placental isoferritin. As employed herein, the
phrase ~capable of reacting with placental isoferritin~ is
intended to encompass any antibody which can so react, e.g.,
either a general, relatively broadly cross-reacting antibody
which will react generally with all types of ferritins as
well as placental isoferritin-specific antibodies. The
present assay permits early diagnosis of problem
pregnancies, even before any symptoms appear; the early
diagnosis thereby allows immediate treatment of the
individual at risk, thus reducing the chances of a premature
termination of pregnancy, and possible death of the fetus.
The i.nvention also provides a method for detecting
pathological pregnancies in early stages of pregnancy,
before PLF levels are detectable even in normal pregnant
women, by observing the concentration of ferritin-bearing
lymphocytes in the blood. Low levels of these F~Ls is an
indic:ation of a poorly developed immunosuppressed statel and
thus, a potentially pathological pregnancy.
The present invention also provides a method for
induc:ing immunosuppression in a host by an initial
administration of alloantigens, to provoke an overall immune
response, and proliferation of T cells, in particular, a
speci.fic subset of ferritin-bearing lymphocytes: this is
followed by an infusion of PLF or an anti-PLF antibody, at
periodic intervals, i.e., throughout the period during which
immunsuppression is desired. The PLF aids in maintaining
133995~
the immunosuppression, once proliferation of the proper
subs~!t of s~ppressor T-cells has been achieved. A
therapeutic regimen oE this sort is particularly useful in
the t:reatment of pathological pregnancies, as well in
assoc:iation with organ and tissue transplants.
4. BRIEF DESCRIPTION OF THE FIGURES
FIGURE l shows a graphic depiction of the mean
serum PLF levels throughout the course of pregnancy in
normall pregnant women
FIGURE 2 shows a plot of serum PLF levels in
pregnant women relative to the timing of their delivery.
FIGURE 3 shows an analysis of the delivery times
of thle same subjects plotted in FIGURE 2; ~+~ indicates
normal level of PLF, and ~-~ indicates low levels or no PLF.
FIGURE 4 shows a comparison of PLF levels in
women, experiencing spontaneous first trimester, missed
abortions or incomplete abortions, with women having normal,
voluntarily terminatecl pregnancies (TOP).
FIGURE 5 shows a comparison of serum ferritin
levels in women having normal pregnancies with women
experiencing abnormal or pathological pregnancies. TOP =
termination of pregnancy (voluntary abortion); PET =
toxemia.
FIGURE 6 shows a comparison of PLF levels on the
third trimester of women having normal pregnancies with
women experiencing abnormal or pathological pregnancies; PET
5 toxemia.
FIGURE 7 shows levels of inhibition of immune
response caused by addition of PLF, and/or PLF specific
antibody to one way mixed lymphocyte cultures, compared with
inhibition by a non-PLF specific antibody (CM-G-8) and
normal ferritin (SPL-FER).
13399~8
-12
FIGURE 8 shows a comparison of mean serum
(normal) ferritin levels in women having a normal term
delivery, women with a normal pregnancy at 30-39 weeks of
gestation, and women with toxemia of pregnancy at 36 weeks
of gestation.
FIGURE 9 shows a comparison of mean serum PLF
levels among the same groups described in Figure 8.
FIGURE lO shows a comparison of the inhibitory
effect of untreated maternal serum, and maternal serum
treated with an anti-PLF antibody on proliferation of T-
cells in a one-way mixed lymphocyte reaction between
maternal and embryo lymphocytes.
FIGURE ll shows a comparison of the inhibitory
effects of PLF and anti-PLF antibody, on PHA induced
proliferation of lymphocytes, both before and after
transfusion.
5. DETAILED DESCRIPTION OF THE INVENTION
The present invention provides a means and a
method by which placental isoferritin can be readily
measured in serum. The observation that lower than normal
levels of PLF in serum of a pregnant woman appears to be
indicative of potential risk in pregnancy, in combination
with the discovery of PLF-specific antibodies, is the basis
for construction of immunoassays useful in the detection of
PLF.
The level of PLF in normal, non-pregnant
individuals is typically undetectable in their serum; on the
other hand, elevated levels of PLF are routinely found in
serum of healthy pregnant women (Moroz et al., Exp.
Haema1:ol. 15:258, 1987), from at least the 17th week of
gesta1:ion, and perhaps earlier, up through full term
delivery (Figure l). rhis level of PLF in pregnant women is
independent of the total ferritin level observed.
Unexpectedly, however, it has now been discovered that the
-13- 1339958
level.s of PLF in women exhibiting various types of
irreg~ularities or pathology in association with their
pregmancy is exceptionally low or undetectable. This
obser~ation has repeatedly been made in significant numbers
of wo,men exhib~ting clinical symptoms of toxemia, women who
have delivered prematurely, women who have undergone
spont.aneous or missed abortions~ and women who experience
prema.ture contractions and later deliver prematurely. The
low levels of PLF in a pregnant woman can thus be employed
as an. accurate marker of a potentially pathological
pregn.ancy, even before the onset of any clinical symptoms
which would otherwise indicate a problem existed.
- Initial studies performed in connection with the
present invention (Moroz et al. Clin. Exp. Immunol. 69:702-
706, 1987) examined levels of PLF in the sera of several
pregnant women, women at full-term, and preterm delivery,
and their newborns. In the course of this study, a number
of interesting observations came to light. Normal non-
pregnant adult females have an average non-PLF normal
ferritin level of about 50 + 59~8 ng/ml, but have very low
or undetectable levels of placental ferritin in their blood,
typically less than about 4.5 + 7.7 U/ml. The average
amount of non-PLF ferritin does not really change
significantly in pregnant women at any time during
pregnancy. However, PLF levels increase significantly in
pregnant women (see Table 1), reaching a peak just before
delivery, and dropping back to normal levels shortly
thereafter. Particularly surprising, however, was the
observation that in mothers who delivered preterm newborns,
the levels of PLF were significantly lower than observed in
mothers who delivered full term newborns (see Table 2).
Interestingly, the levels of PLF in both full-term and
preterm newborns did n.ot differ significantly.
-14- - 1339~8
Further clinical studies have now indicated that
this observed pattern of reduced PLF levels is common to a
numbe!r of other conditions characteristic of abnormal
pregn,ancy. Studies on women with preclamptic toxemia of
pregn,ancy tP.E.T.), when compared with non-toxemic pregnant
women, at similar stages of gestation, showed a pattern of
reduced or undetectable PLF throughout pregnancy. Again,
non-PLF ferritin levels were sub6tantially identical for
- both groups of pregnant women (Figures 5 and 6).
A third observation also correlates with the
above findings. Clinical testing of serum of women who
experience premature contractions early in pregnancy also
suggests that low levels of PLF may be predictive of the
occurrence of premature deliveries. Of the women selected
for the trial, all of whom were experiencing premature
contractions, a majority number showed PLF levels which were
below normal for pregnant women~ However, some of the women
did show normal levels of PLF. Followup studies on all
these women significantly showed that those showing normal
levels of PLF did in f'act carry their babies for the full
term, whereas those women experiencing premature
contractions with low PLF levels tended to deliver preterm
infants. Thus, in conjunction with the occurrence of
premature contractions, the level of PLF in pregnant women's
serum appears to be indicative of a risk of premature
delivery (see Figures 2 and 3). Finally, a significant
number of women who spontaneously abort, or abort
incom~pletely, have shown similar low levels of PLF (Figure
4), wlhile maintaining normal lev-ls of non-PLF normal
ferritin.
The foregoing observations establish a clear
correlation between a low (i.e., less than 10 U/ml) level of
PLF and high risk of an abnormal or pathological pregnancy.
1~39~
- -15-
Thus, PLF can be used as an accurate marker for prediction
of problem pregnancies, in many cases prior to the
appearance of other clinical symptoms of pathology.
These observations concerning PLF levels as a
prognostic indication have been successfully exploited in
combination with novel immunoassays utilizing a PLF-specific
monoclonal antibody. There are now available monoclonal
antibodies which will react with PLF, but which will not
react with any other type of ferritin. An example of a
hybridoma producing such an antibody is deposited in the
Institute Pasteur C.N.C.M. under Accession No. I-256, and
the antibody is referred to herein as CM-H-9. A method for
obtaining such antibodies is described in Section 6.2,
infra. These antibodies may be used alone in a variety of
different immunoassays in order to detect the levels of PLF
present in the serum of pregnant women. Alternately, the
PLF-specific antibodies may be used in combination with
ferritin-specific but non-PLF specific antibodies.
Monoclonal antibodies of this type are obtainable by the
methods described in Section 6.1.2, infra, and are referred
to herein as CM-G-8, or CM-OF-3; while these antibodies are
produced by different clones, they apparently have the same
specificity.
5.1. IMMUNOASSAYS
The present invention also provides a diagnostic
assay for the detection of potentially problem pregnancies.
The antibodies described above may be used as the basic
reagents of a number of different immunoassays to determine
the level of PLF in a pregnant woman's serum. Generally
speaking, the antibodies can be employed in any type of
immunoassay in which the result is quantifiable. This
incluldes both single site and two-site, or sandwich assays
of the non-competitive type, as well as in traditional
competitive binding assays.
-16- 1339~5~
Particularly preferred, for ease of detection, is
the sandwich assay, oi which a number of variations exist,
all of which are intended to be encompassed by the present
invention.
For example, in a typical forward assay,
unlabeled antibody is immobilized on a solid substrate and
the sample to be teste.d brought into contact with the bound
molecule. After a suitable period of incubation, for a
period of time sufficient to allow formation of an
antibody-antigen binary complex. At this point, a second
antibody, labelled with a reporter molecule capable of
inducing a detectable signal, is then added and incubated,
allowing time sufficient for the formation of a ternary
complex of antibody-antigen-labeled antibody. Any unreacted
material is washed away, and the presence of the antigen is
determined by observation of a signal, or may be quantitated
by comparing with a control sample containing known amounts
of antigen. Variation,s on the forward assay include the
simultaneous assay, in, which both sample and antibody are
added simultaneously to the bound antibody, or a reverse
assay in which the labelled antibody and sample to be tested
are first combined, incubated and added to the unlabelled
surface bound antibody. These techniques are well known to
those skilled in the art, and the possibility of minor
variations will be readily apparent. As used herein,
~sand1wich assay~ is intended to encompass all variations on
the basic two-site technigue.
For the immunoassays of the present invention,
the only limiting factor i8 that the labelled antibody be a
PLF-s~pecific antibody. Thus, a number of combinations are
possilble. The bound antibody and the labelled antibody can
both be PLF specific; alternately, the bound antibody will
be a mon-PLF, specific antiferritin antibody and the second
labelled antibody will be PLF specific.
133~95~
-17-
As a more specific example, in the typical
forward sandwich assay, a non-PLF specific antiferritin
antibody is either covalently or passively bound to a solid
surface. The solid surface is usually glass or a polymer,
the TlOSt commonly used polymers being cellulose,
polyalcrylamide, nylon, polystyrene, polyvinyl chloride or
polypropylene. The solid supports may be in the form of
tubes, beads, discs or microplates, or any other surface
suitable for conducting an immuno~s~-y. The binding
processes are well-known in the art. Following binding~ the
polyrner-antibody complex is washed in preparation for the
test sample. An aliquot of the serum sample to be tested is
then added to the solid phase complex and incubated at 25~C,
for a period of time sufficient to allow binding of any
ferr:Ltin present to the antibody. The incubation period
will vary, but will generally be in the range of about 2
minut:es - 16 hours. Following the incubation period, the
antibody-ferritin solid phase is washed and dried, incubated
with a second antibody specific for placental ferritin. The
second antibody is linked to a reporter molecule, the
visible signal of which is used to indicate the binding of
the second antibody to any PLF in the sample. By ~reporter
molec:ule~, as used in the present specification and claims,
is meant a molecule which by its chemical nature, provides
an analytically detectable signal which allows the detection
of Pl.F-bound antibody. Detection must be at least
relat:ively quantifiable, to allow determination of the
amount of PLF in the sample, this may be calculated in
abso]Lute terms, or may be done in comparison with a standard
(or series of standards) containing a known normal level of
PLF.
The most commonly used reporter molecules in this
type of assay are either enzymes, fluorophores or
racionuclide containing molecules. In the case of an enzyme
immunoassay an enzyme is conjugated to the second antibody,
-18- ~3399~8
usually by means of glutaraldehyde or periodate. As will be
readily recognized, however, a wide variety of different
ligation techniques exist, which are well-known to the
skilled artisan. Commonly used enzymes include horseradish
peroxidase, glucose o~idase, ~-galactosidase and alkaline
phosphatase, among others. The substrates to be used with
the specific enzymes are generally chosen for the
production, upon hydrolysis by the corresponding enzyme, of
a detectable color change. For example, p-nitrophenyl
pho~phate is suitable for use with alkaline phosphatase
conjugates; for peroxidase conjugates, 1,2-phenylenediamine
or tolidine are commonly used. It is also possible to
employ fluorogenic substrates, which yield a fluorescent
product rather than the chromogenic substrates noted above.
In all cases, the enzyme-labelled PLF-specific antibody is
added to the first antibody-ferritin complex, allowed to
bind to the complex, then the excess reagent is washed away.
A solution containing the appropriate substrate is then
added to the tertiary complex of antibody-PLF-labelled
antibody. The substrate reacts with the enzyme linked to
the second antibody, giving a qualitative visual signal,
which may be further quantitated, usually
spectrophotometrically, to give an evaluation of the amount
of PLF which is present in the serum sample.
Alternately, fluorescent compounds, such as
fluorescein and rhodamine, may be chemically coupled to
antibodies without altering their binding capacity. When
activated by illumination with light of a particular
wavelength, the fluorochrome-labelled antibody absorbs the
light energy, inducing a state of excitability in the
molecule, followed by emission of the light at a
characteristic longer wavelength. The emission appears as a
characteristic color visually detectable with a light
microscope. As in the EIA, the fluorescent labelled PLF-
specific antibody is allowed to bind to the first antibody-
-19- . 133g9~
ferritin complex. Afl:er washing off the unbound reagent,
the remaining ternary complex is then exposed to light of
the alppropriate wavelength, and the fluorescence observed
indicates the presence of interest. Immunofluorescence and
EIA technigues are bo1:h very well established in the art and
are particularly preferred for the present method. However,
other reporter molecules, such as radioisotopes,
chemiluminescent or bioluminescent molecules may also be
employed. It will be readily apparent to the skilled
artisan how to vary the procedure to suit the required use.
The present: immunoassays are quite useful in
- detection of abnormal PLF levels as early as the start of
the second trimester of pregnancy, i.e., about 12 weeks.
The immunoassay may even be extended to detect the presence
or absence of PLF in earlier stages of pregnancy, given a
suitably sensitive reporter molecule system. The diagnosis
of a potentially pathological pregnancy depends on
measurement of the amount of detectable PLF in the patient's
serum, and correlation of the reading with the standards of
PLF levels which are known to be normal for the patients'
particular stage of pregnancy. A graphic depiction of the
typical pattern of ri~e and fall of PLF in a normal
pregnancy is shown in Figure l. More specifically, a PLF
level of about 20 + 15 U/ml may be considered normal for a
woman in her twelfth week of pregnancy while a level of 40
+ 25 U/ml will be about normal for the twentieth week.
5.2. FERRITIN-BEARING LYMPHOCYTES
While the aforementioned assays, which detect PLF
levels in serum, are quite useful for identifying high risk
patients in the second and third trimesters of pregnancy,
the levels of PLF in t;he first trimester can be so low, even
in normal pregnant individuals, as to make diagnosis on this
basis less than completely reliable. However, it has now
been discovered that it is possible to make an accurate
' -20- 13:39958
determination of the level of immunosuppression by
observation of the numbers of ferritin-bearing lymphocytes
(FBL''s) found in the blood of the individual to be tested.
Initi.al studies indicate that immunosuppressed individuals
can also be identifie~i by the presence of this specific
subset of T-lymphocytes which will react with a PLF-specific
monoc:lonal antibody. These cells carry a PLF-receptor, and
bind with PLF in the serum. Such cells are present at a
- level of at least about 5%, and up to about 20% or moret of
all lymphocytes of a normal pregnant individual, but
const:itute a very low level, usually less than 1%, of the
lymphocytes of a normal individual. Although not yet
conclusively shown, it appears that expansion of this subset
early in pregnancy or even prior to pregnancy, is a
prerequisite to the establishment of an immunosuppressed
state. Cells of this type have also been observed to occur
in high proportions in other disease states in which
immunosuppression is a factor in pathology, i.e., breast
cancer, Hodgkins disease, and AIDS. Thus, the absence of
cells which do not stain when treated with anti-PLF antibody
early~ in pregnancy, i.e., the first trimester, is indictive
of a failure of initiation or breakdown of the mechanism
neceçsAry to permit the development of an immunosuppressed
state!. The identification of these cells makes it possible
to evaluate levels of PLF-production at a stage in which the
level is too low to be detected in serum, but will show up
on i~olated, anti-PLF antibody-labelled cells. Testing of
individuals in the early stages of pregnancy, i.e., the
first trimester, for a percentage of FBLs over about 5%,
there~fore provides an indicator of the state of
immunosuppression, and low levels, or complete absence, of
the F'BLs identifies an individual who may be at risk of a
problem pregnancy. Identification of this subset may be
made by isolation of lymphocytes from the blood of the
-21- 133~95~
individual to be tested by known methods, and conducting an
immun.oassay of the isolated lymphocytes by reaction with.
labelled PLF-specific monoclonal antibodies.
Without wishing to be bound by a particular
theory of operation, t:he following pattern of development of
the immunosuppressed ~tate, at least in connection with
pregnancy, is emerging from the data collected in connection
with the present invention. The presence of an embryo/fetus
in the mother's body exposes the maternal immune system to
alloantigens, in the form of the paternal contribution to
the fetus. This exposure to alloantigens, for example, HLA
antigens of the father, at least initially causes a
nonspecific proliferat:ion of all T-cell populations;
included among these i.s a CD8 subset of T-cells which cells
are either cytotoxic, or, more likely, suppressor, cells.
It is believed that the cells which stain when exposed t.o a
PLF-specific antibody represent expanded suppressor cell
populations. In a no~mal pregnancy, the placenta will
produce placental isoferritin (PLF) which acts as a ~signal~
to the suppressor T-cell population to continue
proliferation. The suppressor cells have the effect
generally of preventing the activity of helper T-cells, and
cytotoxic cells. Thus, the suppressive effect of these
cells aids in preventi.ng the rejection of the fetus in a
normal pregnancy. In a problem pregnancy, however, the
immunosuppressive mechanism fails; either the suppressor T-
cell population does not proliferate sufficiently, or the
placenta does not produce adequate levels of PLF, or both.
There is thus no prote.ction afforded to the fetus by way of
suppression of the mother's immune system, and the danger of
rejection is great. Ultimately the unsuppressed action of
the mother's immune system may cause premature labor or even
spontaneous abortion.
-22- 1339~5~
5.3. IMMUNOSUPPRESSION THERAPY
5.3.l. PLF THERAPY AND PREGNANCY
As noted above, it has often been postulated that
the development of certain types of problem pregnancies may
be associated with the failure of the pregnant woman to
deve].op a level of immunosuppression which will permit the
deve].oping fetus, with its foreign antigens, to coexist
safe].y within the mother's uterus. Beyond this speculation,
however, there have been no concrete answers as to the cause
of a suppression what should be the normal response of the
immune system.
As the foregoing paragraphs and the present
examples make clear, there is at the very least a
coincidental correlat.ion between very low levels of PLF and
several different types of pathological pregnancies, and
this correlation provides the basis for a useful marker
system. Further examination of the association of PLF with
problem pregnancies led to the question of whether there was
more than a coincidenc:e involved in the low PLF levels,
i.e., whether there was a cause-and-effect association here.
In the course of this study, it has now been unexpectedly
discovered that PLF has immunosuppressive properties, which
may indicate that the lack of its production throughout the
course of pregnancy may at least partially be responsible
for and directly relat:ed to a failure to carry a child to
full term in some cases.
In one-way mixed lymphocyte tests, designed to
show the effect a given compound has on the ability of
normal lymphocytes to respond to foreign antigens, PLF
proved to be quite effective in inhibiting the response of
lymphocytes against non-related l~ymphocytes in culture, as
shown in Figure 7. In brief, the mixed lymphocyte culture
is set up with two different types of unrelated lymphocytes
one s,et of which has been pre-inactivated by treatment with
Mitomycin C: under normal circumstances, normal lymphocytes
1339958
-23--
will respond to the presence of a foreign antigen by becoming
activLted and transferred into lymphoblasts. To test the
abilicy of any given compound to cause immunosuppression, the
compound of interest is added to the culture and observations
are mLde on the level of measurable response the functional
lymphocytes have against the foreign cells. As shown in Fiy-
ure 7, PLF in culture containing normal donor lymphocytes in
the presence of non-re:Lated cells did callse a measurable
level of inhibition of the expected response. However, the
resulcs observed in mixed culture having maternal lymphocytes
and cells derived from full term fetus are particularly in-
teresting: cultures iIl which PLF was placed showed an even
greater level of inhibition of maternal vs. embryo response
than was observed with normLl donor vs. non-related cells.
Thus, the presence of PLF has a ciemonstrated immunosuppres-
sive effect on maternaL lymphocytes in the presence of fetal
cells. Perhaps more surprising is the marked effect that the
PLF-specific antibody has on immunosuppression: the inhilc)i-
tion is even greater than that observed with PLF. Indee(i,
the response observed with a combination of both PLF and a
PLF monoclonal antibody is the highest of all, with over 60%
inhibition observed.
The results presented above show the utility of
PLF a, a useful marker to be detected for diagnostic pur-
poses in problem pregnancies; the apparent immunosuppres-
sive activity of PLF iIl mixed lymphocyte culture is also
strongly indicative of PLF playing an integral role in
causing immunosuppression. A third test, however, con-
firms the role that PLI' plays ln vivo. One way mixed lym--
phocyte cultures utilizing lymphocytes from mothers and
their infants (embryonic) were incubated with the maternal
serum presumably containing PLF. Reference to Figure 10
shows that maternal serum has as strong an inhibitory ef-
fect on lymphocyte activity as does PLF added alone to mixed
-24- 13 3g9 5~
lymphocytes. Even more interesting, however, is the effect
of treatment of maternal serum, prior to addition to the
~LC, on an immunosorbent column, with CM-H-9, an anti-PLF
specific antibody. Again, reference to Figure 10 shows
that, after the removal of P~F from the maternal serum by
the antibody, the inhibitory effect of the maternal serum is
virtually completely xemoved. This result supports the
conclusion of an active role of PLF in the immunosuppression
believed to be necessary in the maintenance of a full term
pregniancy.
5.3.2. PLF AND TRANSPLANTS
The immunological similarity between the
implantation of an embryo or fetus, and the grafting or
transplantation of al]ogeneic tissue to a host has often
been noted. It would be expected that, in both cases, the
host immune system would initiate a response against the
foreign antigens, and as the ultimate result, rejection of
the fetus or the graft: would eventually occur. As has
already been noted, the induction of an immunosuppressed
state at the start of pregnancy is almost certainly
responsible for the prevention of embryo rejection; it is
also recognized that i.mmune suppression plays a key role in
successful transplantation, although in the case of
transplants or grafts, immunosuppression is not, apparently r
a natural response, as it is in pregnancy. Therefore, it
has become routine practice to administer immunosuppressive
~drugs, such as cyclosporin A, prior to transplantation, in
order to ~prime~ the host~s immune system to receive, and
more importantly, tolerate, the transplanted tissue.
In recent years, clinical observations have shown
that blood transfusions decrease antigen induced and mitogen
induced proliferation of the cellular elements associated
with immunity, while enhancing suppression activity. In
many cases, one or more blood transfusions, prior to
- -25- 13399~8
transplantation, results in increased tolerance of the
allograft, and may minimize rejection of the foreign tissue~
This pattern, at least: superficially, resembles the pattern
proposed for normal pregnancy as well. It has been
suggested that the mechanism of induction of
immun,osuppression by t:ransfusion is related to an antigen-
nonspecific suppression, or idiotypic regulation (Woodruff
et al., Lancet l:1201--1203, 1983). More specifically, it is
thought that allogeneic lymphocytes present in the
transfusion may activate immunosuppression in certain
combinations in which the MHC of the blood donor differs
from that of the recipient at a particular HLA locus
- (Shearer et al., J. Exp. Med. 157:936-946, lg83).
Based on the physiological similarities between
the transplantation and embryo implantation, and the
desirability of immunosuppression in each, it is believed
that PLF may also serve as a therapeutic agent in prepping a
patient for transplant:ation or grafting, and a regimen is
proposed whereby the necessity for transfusion, particularly
multiple transfusions, is avoided. It has been observed,
however, that PLF alone will not be sufficient to cause
inhibition of lymphocyte proliferation. Figure 11 shows the
effect of PLF and anti.-PLF antibody on the inhibition of
phytohemaglutinnin (PHA) transformation of lymphocytes,
prior to transfusion. The same figure, however, shows that,
after transfusion, bot:h PLF and the antibody have a
significant effect on lymphocyte inhibition. Thus, a
combination of PLF infusion and immunization with
alloantigens is proposed as an alternative to the currently
used drug therapy, which may have undesirable side effects.
5.3.3. THERAPEUTIC REGIMENS
The aforeme.ntioned results strongly indicate the
therapeutic value of PLF, an~ the anti-PLF antibody in
maintaining a desirable level of immunosuppression in those
-26- 133995~
circumstances in which a hyporeactive immune state is
required. It is also shown that PLF probably requires the
presence of a particular subject of (presumed) suppression
T-ce].ls in order to exert its desired effect, while the
cell~i cannot maintain the suppressed state without the PLF
pres~nt in the serum as a ~signal~ for continued
prolLferation. The observation of these interrelationships
sugg~sts a valuable method of inducing the neceCcAry
response where the body's natural response is inadequate.
In pregnancy, for example, a woman testing negatively, by
way of the FBL test, in the first trimester, may be
stimulated by immunization with any alloantigen;
partiLcularly favored, however, is the use of allogeneic
lymphocytes, to initiate or supplement the proliferation of
suppressor T-cells. This type of treatment has frequently
been used to monitor the effect of exposure of pregnant
females to paternal antigens and any resulting effect it may
have on a normal term pregnancy (see, e.g., Netter, et al.,
Contracept. Fertil. Sex 15:542-549, 1987; Mowbray et al.,
Lance~t 1:941_943, 1985). Immunization with about 106-108
cells is typically sufficient to provoke a T-cell response.
The immunization is followed by parenteral administration of
PLF, or a PLF-specific antibody, preferably by infusion.
Therapeutic administration of PLF and/or the PLF antibody
may range from 1-100 mg, with about l-lO mg being a
preferred range. The patient can then be monitored
periodically by ELISA for her serum PLF levels, and, if
necessary, boosting doses of PLF/antibody may be
admini6tered as needed. For women in the second and third
trimesters, an immunization is not usually necessary, since
pregnancy would not continue to this point had an initial
level of suppression not already been established.
Therefore, the same parenteral administration as noted above
is generally adec~ate to maintain the necessary levels
' -27- 13399S8
throughout pregnancy to delivery. This can be verified,
however, by a simple 'ELISA to detect the presence of
ade~ate numbers of F'BLs.
With respect to pretreatment for transplantation
purposes, the individual should be immunized, again with
alloantigens, as noted above, preferably at least about 6
days prior to the proposed transplant. The immunization
will be followed by PLF and/or antibody infusion preferably
withiLn 24 hours of immunization. Following transplantation,
PLF ]Levels should be monitored and kept at a level of at
least: about 20 U/ml. Again, as with pregnancy PLF levels
can be readily monitored and boosting doses administered as
n~c~:s~ry.
6. EXAMPLE
6.1. PREPARATION OF MONOCLONAL ANTIBODIES
The following protocol describes the method of
preparation of both a PLF-specific monoclonal antibody and a
ferri.tin-specific, but non-PLF-specific, monoclonal
antibody.
6.1.1. PREPARATION OF ONCOFETAL FERRITIN
Ferritin was prepared from human placenta by a
modif'ication of the method used by Beamish et al. (J. Clin.
Path. 24:581, 1971). Placental tissue (500 gms) was sliced
and water added to a total volume of 2000 ml. After
homog~enization the tissue suspension was heated to 75~C for
20 minutes. The supernatant, after cooling and
centrifugation at 10,000 rpm for 15 minutes, was treated
with acetic acid to bring the pH to 4.6. The precipitated
protein was removed by centrifugation at 10,000 rpm for 15
minutes and a clear supernatant was adjusted to neutral pH
with dilute NaOH. When the clear brown supernatant was
ultracentrifuged at 100,000 g for 240 ~inutes the suspended
ferritin collected in a small button at the bottom of the
-28- 13~9958
tube The precipitate was redissolved in 0.9% saline and
further purified by passage through a Sephadex G200 column.
The i-erritin fraction from this column was passed through a
DEAE cellulose anion exchange resin using Tris-HCl buffer at
pH 7.5 and a G.C2-0.5 M gradient. Three protein peaks were
obtained, the most ac;idic peak pI=4.8(No.III) was collected
and used for analysis. Its purity was shown by isoelectric
focusing and immunoelectrophoresis against anti-ferritin
serum and anti-human whole serum. This was used for the
immunization of mice, as described below.
6.1.2. PREPARATION OF NON-PLF SPECIFIC HYBRIDOMAS
The myeloma cells used for hybridization were
from cell line PB/NSl,/l-Ag4-1; these were grown in RPMI-1640
with 20% Fetal Calf Serum (FCS). Spleen cells were obtained
from 4-6 week old Balb/c female mice. These mice immunized
with 3 weekly immunizations, of 50 ~g of acidic ferritin in
complete Freund's adjuvant. Hybridization was begun 3 days
after the last inject:ion of 10 ~g ferritin. Hyperimmune
mice were rested at least one month before the last
boost;ing.
Spleens were removed from the immunized mice in
RPMI-O, and rinsed twice in a petri dish with RPMI-O. The
splee!ns were teased apart in RPMI-O with 18 ga. needles, and
the resulting cell suspension transferred to a tube, wherein
large! chunks of tissue settled out. This single cell
suspe!nsion was removecl to a new tube spun at 800 RPM
(160 x g) for 5 minutes, and red blood cells were lysed with
0.83% NH4Cl, at pH 7. ej . The cells were washed three times
with RPMI-O, resuspencled in same and counted with Trypan
- Blue.
-29- 1339958
The myeloma cells were removed from culture
flasks with gentle pipetting into a 50 ml Falcon/Corning
tube, and spun down at 900 RPM (200 x g) for 5 minutes.
They were then washed once with RPMI-0, resuspended in same,
and counted with Trypan Blue.
The spleen and myeloma cells were combined in a
10:1 ratio in a single! 50 ml conical Falcon/Corning
disposable centrifuge tube, and the cells were pelleted at
900 RPM (200 x g) for 5 minutes; the medium was aspirated as
completely as possible. All solutions and media used from
this point were at room temperature. The tube with the cell
pellet was immersed in, a bath at 37~C, and 0.2 ml 33% PEG
1500 was added for 1 minute, accompanied by gentle stirring
and then centrifuged at 200 x g for 5 minutes. Cells were
resuspended and stirred gently for 1 minute followed by the
addition of 5 ml RPMI-0, with gentle stirring and then by
addition of of 5 ml RF'MI-0, and 20% Fetal Calf Serum. The
hybrid mixture looked like a poorly resuspended cell
suspension at this point, with many small clumps. The
mixture was pelleted at 200 xg 5 minutes, and the cells then
resuspended in RPMI-HY-HATD (at 37-C) at a concentration of
3x106/cc by squirting medium onto the cell pellet.
The hybrids were then plated out in flat bottom
96 well plates by adding 2 drops of cell suspension from a
5 ml pipet or with multi-pipettor using cut off tips (about
65 microliters), containing 100-120 RPMI-HY-HATD (approx.
2x10 cells). Control wells containing NS-l cells + RPMI-
HY-HATD at lxl06/ml were set up. Plates were cultured for 7
days. On day 8 and twice a week therefrom, half of the
culture medium was removed by careful aspiration and fed
with B0-100 microliters of RPMI-HY-HT medium. Positive
wells were screened for antibody production at 3 and 4 weeks
after hybridization.
The solutions and media used above are prepared
in th~e following manner:
~30- 1~39958
RPMI-0: no RPMI with Fetal Calf Serum
RPMI 1640-HY
500 ml sterile distilled water
55 ml 10 x RPMI-1640
6 ml 1.0 N Sodium Hydroxide
14 ml 7.5% Sodium Bicarbonate
6 ml Pen/st:rep
10 ml Glutamine ) + DMEM
86.5 ml FCS
RPMI-HY-HAl~D - day 0 --> day 7
For 100 ml of medium
95 ml RPMI - 1640 + 20% FCS
1.0 ml Pyruvate (lOOx)
2.0 ml 50 x HAT
2.0 ml 50 x deoxycytidine
RPMI-HY-HT - day 8 --> day 14
For 100 ml of medium
97 ml RPMI-1640 + 20% FCS
2.0 ml 50 x HT
1.0 ml Pyruvate (lOOx)
For Hybrids from day 15 onwards, use RPMI-1640 +
20% FCS and Pyruvate, or maintain in RPMI-HY-HT.
PEG 33 and 25% w/v
Must be odorless and white. For 100 ml autoclave
relevant wt in grams in glass bottle at 15 lbs
for 10-15 minutes. When bottle is cool enough to
hand hold (about 50~C) add RPMI 1640-0 to make up
to 100 ml, swirl to mix, store at room
temperature.
-31- 133995~
HATD - Final concentrations of reagents
H = Hypoxanthine 10 4M
A = Aminopterin 10 6M
T = Thymidine 2x105M
~ = Deoxyc~ytidine 2x106M
HT Stock 100x - 100 cc
Thymidine M.WT 242.33 - 0.04846 g
- Hypoxanthine M.Wt 136.2 - 0.1361 g
dd H2O up to 100 ml and warm to 60-70~C to
dissolve.
Readjust final volume with dd H2O. Dilute to 50x
and filter (0.2 ~) sterilize. Make 2 ml
aliquots, store at -20~C.
A Stock 1000x - 100 cc
Aminopterin F.Wt 440.4 (0.44 g)
Bring to 50 ml with dd H2O, add 0.1 N NaOH
dropwise until aminopterin dissolves. Bring
final volume to 100 ml with dd H2O. Adjust
volume to 100 ml. Filter (0.2 ~) sterilize.
Store at -.'0~C.
D Stock 100x - 100 cc
Deoxycytidine M.Wt 227.2 (0.00454 g)
Dissolve in dd H2O, adjust to 100 cc, dilute to
50x stock, filter (0.2 ~) sterilize and store at
-20~C.
HAT - 50x -- 200 ml
Combine 10CI ml 100 x HT with 10 ml 1000 x A ~ 90
ml dd H2O s 50 x HAT, filter (0.2 ~) sterilize,
make 2 ml aliquots and freeze at -20~C.
-32- 1~39958
Screening and determination of the specificity of
the monoclonal antiboclies was performed by a
hemagglutination test. Embryonic placenta and adult spleen
ferritin were coupled to 0x red blood cells (0x RBC) by
CrCl2. 50 ~l of incre.asinq dilutions (starting at l:l0 of
hybridoma culture medium supernatant were mixed with l0 ~l
of adult or embryonic ferritin Ox RBC and hemagglutination
determined. Supernatants of clones giving a
hemagglutination titer of at least l:lOOO were selected. A
clone~ designated CM-0~-3 was selected. The antibody
produced by the clone CM-OF-3 is specific for ferritin and
it cross-reacts with both adult and placental ferritin. The
monoclonal antibody obtained, CM-OF-3 was used to block the
cross-reactive determi.nants of placental and adult ferritin,
in order to produce a different monoclonal antibody, CM-OF-
H9, which is directed to a specific fetal determinant of
PLF.
6.l.3. PREPAR~TION OF PLF-SPECIFIC HYBRIDOMAS
AND ANTIBODIES
Placental lerritin (PLF) isolated from human
placenta, a protein oi pI 4.8, was reacted with monoclonal
anti~odies CM-OF-3 in the following ratio: PLF (90 ~g in
PBS) was mixed with ascites fluid from BALB/c mouse
containing CM-OF-3 ant:iferritin monoclonal antibodies (l0
mg/ml).
The mixture was incubated for 30 min at 37~C
followed by overnight incubation at 4-C. The mixture was
centrifuged at l0000 G;, the precipitate formed was
discarded, and the supernatant was used for immunization.
Each BALB/c mouse was immunized with the above supernatant
mixed with complete Freund's adjuvant, injected
intradermally once a week for 3 weeks. A booster
immunization of one-fi.fth of the above dose was injected
intraperitoneally.
' -33- - 133~958
After 3 days from boost mouse spleen was
asepltically removed and fusion was performed by incubating
108 spleen cells with 107/P3-NSI/1-Ag4 myeloma cells as set
out above ~n the hybridization procedure and the same
subsequent procedures for positive clone identification was
followed. Positive clones were also tested against liver
and spleen ferritin to confirm a lack of cross-reactivity.
Thus a clone designated as CM-OF-H9 was obtained which
produces antibodies CM-H-9. A biologically pure culture of
the ~-OF-H9 hybridoma has been deposited with the Institute
Pasteur C.N.C.M. under Accession No. I-256.
Characteristics of the CM-OF-H9 monoclonal
antibodies include: the CM-H9 monoclonal antibody belongs
to the IgG class; it does not form precipitates with
ferriLtin it binds rabbit complement. In the ascitic fluid
obtained the antibody content was about 7 mg per ml. One
ml of ascitic fluid binds about 2 mg of embryonic ferritin
and none of adult spleen or liver ferritin.
6.2. PLF I~VELS AND PRETERM DELIVERIES
Twenty-five full-term newborns (14 males and 11
females) aged 38-41 weeks of gestation (mean weight 3350 +
3000 g) 25 preterm newborns (12 males and 13 females) aged
26-3~i weeks of gestat:ion (mean weight 1820 + 503 g) and
their respective mothers were examined. The gestational age
was c:alculated from the first day of the mother's menstrual
bleecling preceding pregnancy and was confirmed by clinical
exami.nation (Dubowitz scoring). Twenty-three women at 17-22
weeks of gestation who underwent amniocentesis because of
their age and fifteen women at 30-39 weeks of normal
pregnancy were investigated. Forty healthy volunteers ~16
females and 24 males) aged 20-40 years served as controls.
Blood was withdrawn from the umbilical cord of
preterm and full-term newborns after ligation of the cord
from t~e newborn side of the placenta. Concomitantly venous
-34- 13~9~
blood was obtained f:rom the respective mothers. Amniotic
fluid was wlthdrawn Erom pregnant women at 17-22 weeks of
pregnancy and venous blood was collected from all other
groups as shown in Table 1.
Monoclona:L antibodies CM-G-8 and CM-H-9 were
prodluced against human placental ferritin as previously
described (Moroz et al., 1985). Antibodies were obtained
frorl ascites fluid following precipitation with 50~
saturated ammonium sulphate solution. Placental ferritin
used for standard was obtained following purification on
DEAE'-cellulose column as described previously (Moroz et al.,
Clin. Chem. Acta 148 111, 1985). Liver ferritin standards
were obtained from McELISA ferritin kits (Elias Medizin-
technik GmbH, D-7800" Freiburg). The amount of placental
ferritin which bound 250 pg of alkaline phosphatase-
conjugated CM-H-9 Mol~ was considered as 10 units of PLF.
The enzyme-linked immunosorbent assays measuring
the serum ferritin and PLF in the serum (MoELISA type A and
MoELISA type B respec:tively) have been described (Moroz et
al., Exp. Hemat. I5:.'58, 1987),. In both assays, the
monoclonal antibody C'M-G-8 which binds to all ferritins was
coupled to the solid phase. For the second site CM-G-~
MoAb-enzyme conjugate was used in MoELISA type A and CM-H-9
MoAb-enzyme conjugate in MoELISA type B.
The MoELI';A type A and B were performed as
follows: The wells of a microtitre plate (Dynatech m-129B~
were coated with 150 ~1 CM-G-8 MoAb (100 ~g/ml phosphate-
buffered saline (PBS), pH 7.2) and incubated overnight at
4~C. The plate was washed three times with PBS-Tween (PBS,
0.05% Tween 20) and $,haken dry.,
Test sera (100 ~1) diluted 1:2 in MoELISA type A
and 1:4 in MoELISA type B in PBS-Tween 0.025%, were added ln
duplicates to the wells. Serum diluent and ferritin
standards were also added in duplicates to the wells. Serum
dilu,ent and ferritin standards were also added in
* Trade-mark
133~8
duplicates. A serum siample with an elevated ferritin
concentration was diluted in the diluent to determine
recovery at high dilut.ion. The plates were incubated at 4~C
for 1 h in MoELISA A and overnight in MoELISA B, washed
three times with P~S-I'ween and then 100 ~1 of alkaline
phosphatase (AP) MoAb conjugate (0.4 ~1) was added to each
well. The plate was i.ncubated further for 120 min at room
temperature and washed~ again three times. The enzyme
substrate (p-nitrophenylphosphate 1 mg/ml of diethanolamine
buffer, pH 8.0, 0.5 m~[ MgC12) was added and the reaction was
stopped after 10-30 mi.n by addition of 0.05 ml of 2 M NaOH.
The amount of colored product was measured by absorbance at
405 nm.
Statistical analysis was performed according to
the Mann-Whitney U-test (non-parametric).
The mean concentration of total serum ferritin i~
pregnant women and in women at delivery was similar to that
measured in the sera of adult blood bank female donors
(Table 1). The mean concentration of ferritin ranged from
46 to 63 ng/ml as dete!rmined by MoELISA type A (Table 1).
In comparison, pregnan,t women exhibited elevated serum
concentration of PLF, as measured by MoELISA type B
significantly higher t.han controls (P ~ 0.025, Table 1). It
is noteworthy that 70% of the sera of the normal female
testeld contained no detectable PLF.
In this stu.dy it was found that high PLF levels
were Idetected in the serum of women as early as 17 weeks of
gestation (45.5 + 52.9 u/ml). PLF levels did not change
significantly during 30-39 weeks of gestation (81.6 + 89.3)
and w~ere high at full-term delivery (54.8 + 53). However,
in women who had pretenm delivery the concentration of serum
PLF was significantly lower (15.8 + 15.7 u/ml) than that
measured in the sera of pregnant women at similar gestation
perio~d and in women at. te~m delivery (P = 0.02) (Table 1).
Measurement of ferritin concentrations in the amniotic fluid
~ -36- 1339~
at 17-22 weeks of gestation revealed that the total ferritin
concentration (86.4 + 78.5 ng/ml) was not significantly
different from that of the serum (63.7 + 48.9 ng/ml) whereas
the ]PLF level (19.4 + 8.2 u/ml was significantly lower than
that measured in the serum (45.5 + 52.g u/ml, P = 0.02).
Table 1
The Levels of ~erritin and PLF During Pregnancy
Fema:Le Ferritin PLF
blood donor n (ng/ml) (u/ml)
Adu11: females 16 50.3+59.84.5+ 7.7
Pregnant 17-22 weeks 23 63.7+48.9*45.5+52.9t
Pregnant 30-39 weeks 15 50.0+39.4*81.6+89.3t
Term delivery 25 46.1+37.3*54.8+53.0t
Preterm delivery 18 61.2+29.4*15.8+15.7t
Amniotic fluid 17-22 lweeks 15 86.4+78.519.4+ 8.2
*~ot significantly different from adult female.
t';ignificantly higher than adult female (P = 0.0025)~
';ignificantly lower than in pregnant women and women at
term delivery (P = 0.02).
The results are expressed as mean + s.d.
Determinations of ferritin and PLF in the sera of
term and preterm newborns. The mean concentration of total
ferri.tin in the sera of full-term and preterm newborns
(289.4+167.5, 208.2+191.5 respectively) was significantly
higher than that measl~red in the sera of their mothers
(46.~.+37.3, 61.2+29.4 respectively, P-0.0005, Table 2). Yet
no di.fference in ser~n ferritin concentration was observed
between term and preterm newborns (Table 2).
-37~ 399~8
High PLF levels were detected in the sera of term
newborns (20.3+25.8 u/'ml) as compared to normal healthy
adult.s (8.1+14.8 u/ml~l. However, the levels of PLF were
signi,ficantly lower than that of their mothers (Table 2,
P50. 0075). In contrast in preterm newborns the serum PLF
level.s (9.4+15.1) were low, similar to those of healthy
adult.s but were signii-'icantly lower than those of their
mothe!rS (P50. 018). The PLF levels in preterm newborns were
not s,ignificantly lower than those found in term newborns
(P=0.26).
-38- 1339958
Table 2
The revels of Ferritin and PLF in the Sera of
Full-term anc' Preterm Newborns and Their Mothers.
Ferritin PLF
Group n(ng/ml) (u/ml)*
Full term newborns25 289.4+167.5 P=0.0000 20.3+25.8 P=0.0075
Mothers 2546.1+ 37.3 54.8+53.0
Pretenn newborns 25 208.2+191.5 ~-0.0005 9.4+15.1 P=0.018
~thers 1861.2+ 29.4 15.8+15.7
Blood bank ~onors 40 135.3~ 65.9 8.1+14.8
Male 24108.0+ 58.0 10.0+10.0
Fem~le 1650.3~ 59. 8 4.5~ 7.7
15 *The results as mean + s.d.
Significantly higher than PLE in the sera of mothers of
prete~m newborns and adult female (P=0.02, P=0.0025
respectively).
-39- 1~399~8
6.3. PLF LE:VELS AND TOXEMIA OF PREGNANCY
Twenty patients with toxemia of pregnancy (PET)
who delivered their babies in the mid to late third
trimester, were selected prospectively on the basis of
standard clinical criteria. None of these women had
eclampsia. Five women had severe pre-eclampsia, two had
gestational diabetes ~with mild PET, one had chronic
hype~.tension with superimposed PET the rest of the women had
mild PET (see Table 1). All the patients in the two control
groups had normal uncomplicated pregnancies and deliveries.
Blood for chemical analysis was drawn into glass
colle~ction tubes. Serum was collected into glass tubes,
immediately frozen and stored at -20~C until analyzed.
Total serum ferritin lwas measured independently in two
different laboratorie,s. One used a monoclonal antibody
ELISA method. The second laboratory used a polyclonal
antiserum ELISA. The anti-human ferritin-enzyme conjugate
was obtained from DAKO and ferritin standards were obtained
from serono diagnostics. The results obtained from these
two laboratories were compatible. The measurement of the
serum PLF was performed by a MoELISA as previously
described.
Statistically analysis was performed using T test
for ~Inpaired observations. Differences were considered
signi.ficant at p<0.05.
Table 3 exhibits the clinical data of the 20 PET
patients. It is noteworthy that the mean gestational week
of the toxemic patients was 36+3.2 The haemoglobin level
ranged between 8.8 to 13.9 gr% with a mean of 11.3+1.45 gr%.
~40- 13399~8
Table 3
Clinical Data of the Pre-Eclamptic Toxemia Group (N=20)
N Range Mean S.d.
Maternal age (years) 20 22-38 30 5.25
Primipara 12
Multipara 7
Grand Multipara
Gesta1:ion age (weeks) 20 29-42 36 3.2
Oedema 120
Album:inuria (1-4+) 20
Haemoglobin (9%) 20 -8.8-13.9 11.35 1.45
Systo:Lic B.P. (on admittance) 20 130-200 153 l9
Diastolic B.P. (on admittance) 20 80-110 98 10.7
Gest. diabetes 2
Chron.Lc hypertension and
superimposed PET
Mode of delivery C,/S 7
V~g. 13
Birth weight in grams 20 700-3700 2676 800
Ante-partum fetus death
The mean serum concentration of normal ferritin
in to~temic patients was 38.55+55 (ng/ml). This level was
not si.gnificantly different from that level found in normal
pregnant women at 30-39 weeks of gestation (50.0+39.4 ng/ml)
or from the mean serum level in normal women at term
30 delivery (46.1+37.3 ng,/ml) (Fig. 8).
As seen in ~Fig. 9, the mean PLF level in the sera
of to~:emic patients was 7.5+23 u/ml. This level was
significantly lower (p~0.001) than the level of PL~F
(81.6~89.3) in normal pregnant women at similar gestational
age (3l0-39 weeks) as well as from the level of PLF measured
-41~ 1339~58
in wcmen at normal term delivery (54.8+53 u/ml). It is
noteworthy that 17/20 toxemic patients were completely
lacking PLF in their serum and only 1/20 had a PLF level
compatible with the level observed in the normal pregnancies
of similar gestationa]. age.
