A review of the literature indicates that the ability of the foetus to escape immunological rejection by the mother is due, at least in part, to alterations in maternal immunity during pregnancy. A variety of immunoregulatory mechanisms have been proposed, including the suppression of maternal immunity by pregnancy-associated proteins. One such pregnancy-associated protein, early pregnancy factor (EPF), is the subject of study in this thesis.
EPF has been detected in all species investigated, including mice, sheep, pigs and humans, and has been shown previously to affect lymphocyte activity both in vivo and in vitro. The present work establishes the biochemical properties of human EPF and provides further evidence to support .the suggestion that EPF may have an immunosuppressive role in pregnancy. Finally, production of EPF during pregnancy and by cancer patients is investigated, and the possible clinical applications are discussed.
EPF has been detected in both serum and urine during the early stages of human pregnancy. Gel filtration studies have shown that EPF from both sources occurs in several forms, of approximate molecular weights (MW) of 500,000 (500K), 250K, 100K, 60K, 40K, and 20K. The large M (500K and 250K} forms appear to be due to the association of EPF with a carrier molecule; EPF and carrier can be dissociated by passage through an ion exchange (DEAE-cellulose) column. It is suggested that dissociated EPF and carrier may pass (separately) through the glomerular membranes of the kidneys and then reassociate, thus explaining the presence of large MW EPF in the urine. Isoelectrofocussing of human urinary EPF produced two discrete fractions of EPF, at iso-electric points of approximately 6.5 and 6.7, possibly due to the test sample containing a mixture of free EPF and EPF associated with carrier. Treatment with 40% ammonium sulphate resulted in loss of EPF activity from both supernatant and precipitate fractions; activity was regained by recombination of the two fractions. This result was interpreted as indicating that human EPF could be separated into two components, apparently corresponding to EPF-A and EPF-B described previously in mice and sheep.
The biochemical properties of human EPF are similar to those of EPF from other species. Furthermore, immunological cross reactivity has been demonstrated, human EPF binding to antibodies prepared against mouse EPF and sheep EPF-B.
The properties of EPF differentiate it from most other pregnancy-associated proteins. However EPF has been detected as a contaminant of some commercial preparations of human chorionic gonadotrophin (hCG). EPF and hCG in these preparations could be separated by gel filtration, ion exchange chromatography, and affinity chromatography with Sepharose-coupled antibodies to EPF. Fractions derived by ion exchange chromatography were tested for immunosuppressive activity; only EPF-containing fractions suppressed the adoptive transfer by lymphoid cells of contact sensitivity from sensitised to naive mice. It was thus suggested that immunosuppressive activity previously attributed to hCG may have been due to the presence of EPF as a contaminant.
These results confirmed earlier reports concerning the immunosuppressive properties of EPF. Further studies in mice showed that the suppressive action of EPF was not direct, but mediated via the induction of an intermediary factor. This factor was found to suppress lymphocyte activity both in vitro and in vivo, and was thus termed a suppressor factor. Unlike EPF which is neither strain-restricted nor species-specific, the suppressor factor was found to be strain-restricted in its action. Experiments with immunodeficient mice indicated that at least two different cell types are involved in the system. The results suggested that EPF induces suppressor factor production by one cell type (possible non-T cells) and that this factor acts on a second population of cells (possibly T cells and/or natural killer cells). Preliminary studies showed that a similar mechanism operates with human lymphocytes.
The proposed mechanism of action for EPF provides a means by which its effect can be amplified. Furthermore it is suggested that suppression may be selectively directed towards cells responsible for allograft rejection.
Studies of EPF production in early human pregnancy have shown that EPF can be detected in the serum within 24 hours of fertilisation. In a group of 13 nulliparous women, EPF was detected after ovulation and intercourse in 18 of 28 cycles studied. However EPF production continued beyond 14 days in only 4 cases; successful pregnancy was maintained in 2 of these while in the other 2, disappearance of EPF preceded miscarriage. In the remaining 14 cases, EPF disappeared from the serum before the onset of menstruation, thus suggesting a high incidence of fertilisation and early embryonic loss.
EPF or an EPF-like substance (tEPF) has been detected also in patients with germ cell tumours of the testis. tEPF was found to be most frequently associated with tumours containing embryonic or chorionic elements; it was not detected in subjects with benign testicular disease, patients with other (non-germ cell) tumours, or in healthy males. It was shown that tEPF and pregnancy-derived EPF shared several properties, including gel filtration behaviour, reaction with antiserum to EPF, and interaction of free components, (tEPF-A binds with EPF-B and tEPF-B binds with EPF-A).
These studies indicate the possible clinical applications of EPF. Firstly with respect to pregnancy diagnosis, EPF provides a means of distinguishing fertilisation and implantation, and secondly, EPF may provide an additional marker for the germ cell tumours.