- Email: [email protected]
An analysis of the variation of plasma concentrations of placental protein 14 in artificial cycles
Vol. 57, No.4, April 1992
FERTILITY AND STERILITY Copyright" 1992 The American Fertility Society
Printed on acid-free paper in U.S.A.
An analysis of the variation of plasma concentrations of placental protein 14 in artificial cycles
Tin-Chiu Li, Ph.D.*t Caroline Dalton, B.Sc.:\: Anthony E. Bolton, D.Sc.:\:
Elizabeth Ling, B.Sc.* Alistair Warren, Ph.D.§ Ian D. Cooke, F.RC.O.G.*
University of Sheffield, Sheffield City Polytechnic, and Jessop Hospital for Women, Sheffield, United Kingdom
Objective: To analyze the factors affecting the variation of plasma concentration of placental protein 14 (PPI4) in artificial cycles. Design: The effects of different hormone replacement therapy (HRT) regimens were examined in a crossover design. Setting: Jessop Hospital for Women, Sheffield, United Kingdom. Patients: Eighteen women with premature ovarian failure: 6 associated with Turner's syndrome and 12 with idiopathic premature ovarian failure. Interventions: Four different HRT regimens; 36 study cycles. Main Outcome Measures: Plasma PP14 concentrations on days 1, 15, 19, and 29 ofthe artificial cycles. Results: In cycles treated with a standard HRT, the levels were similar to those of the natural cycle. Subjects with Turner's syndrome did not have elevated PP14 levels, whereas the majority [75%]) of those with idiopathic premature ovarian failure had elevated levels on day 29 of the cycle. Levels of PP14 were reduced when either the doses of estradiol valerate were reduced to t or the doses of progesterone (P) were reduced to ~ of the standard HRT. Conclusions: Plasma levels ofPP14 are dependent not only on P stimulation but also on adequate estrogen priming. Fertil Steril 1992;57:776-82
(f2
Key Words: Placental protein, artificial cycles
The endometrial protein placental protein 14 (PP14) (1), also known as progestogen-dependent endometrial protein (2), a-uterine protein (3), and pregnancy-associated endometrial a-2-globulin (4), is one of the principal proteins secreted by endometrial glands (5) in the luteal phase of the menstrual cycle. The production of this protein is thought to be dependent on progesterone (P) (2). The concentration of PP14 in the plasma and in
Received March 25, 1991; revised and accepted December 6, 1991. * Department of Obstetrics and Gynaecology, University of Sheffield. t Reprint requests: Tin-Chiu Li, Ph.D., Jessop Hospital for Women, Sheffield S3 7RE, United Kingdom. :\: Department of Biomedical Sciences, Sheffield City Polytechnic. § Department of Biomedical Science, University of Sheffield.
776
Li et al.
PP14 levels in artificial cycles
endometrial tissue has been found to increase from the early luteal phase to reach a maximum in the late luteal phase of a normal menstrual cycle (6, 7). In pregnancy, the level of PP14 continues to rise to a maximum at approximately 8 to 10 weeks of gestation (8). Recently, it has been found that PP14 inhibits lymphoproliferation (9, 10), suggesting that it may play an important role in the process of human implantation by protecting the implanting embryo from rejection by the maternal immune response. In addition, the successful determination of the N-terminal amino-acid sequence of PP14 has revealed homology with the ,a-lactoglobulin group, indicating a further possible role of this protein in the transport of retinol to the implanting blastocyst (11). In a previous study, we found a large, random variation of the levels of PP14 in different parts Fertility and Sterility
Table 1
Doses ofE and P in the Various HRTs Used in the Present Study* Standard regimen
Day of cycle
E2
lto5 6to 9 10 to 13 14 15 to 16 17 18 to 26 27 to 28
1 2 6 2 2 2 4 1
Fixed I-mg E regimen
Low P regimen P
E2
25 50 50 25
1 2 6 2 2 2 4 1
P
E2
Fixed 2-mg E regimen P
E2
P
25 50 50 25
2 2 2 2 2 2 2 2
25 50 50 25
1TIII
5 10 10 5
1 1 1 1 1 1 1 1
* The standard HRT was identical to that proposed by Navot et al. (15).
of the endometrium (mean coefficient of variation = 73%) (12), casting doubt on the measurement of PP14 levels in the endometrial tissue as being of much diagnostic value. On the other hand, the measurement ofPP14 concentration in plasma has been considered by some investigators to be of value in assessing ovulation (7) and the secretory function of the endometrium (13). In this prospective study, we employed the artificial cycle as an experimental model to examine the variation of plasma concentrations of PP14 among different individuals receiving the same hormone replacement therapy (HRT), in different cycles of the same individual treated with various forms of HRT, and in relation to a number of clinical observations.
MATERIALS AND METHODS Subjects
All subjects included in this study have premature ovarian failure, defined as amenorrhoea of 6 months or more, associated with persistently low plasma estradiol (E 2) «100pmol/L) and elevated plasma follicle-stimulating hormone (FSH) (>20 IU/L) in a woman <40 years of age (14). A total of 18 subjects were recruited, of whom 12 had normal chromosomal analysis, absent antiovarian antibody, and no obvious cause for the premature ovarian failure and so were considered to have idiopathic premature ovarian failure. The remaining 6 subjects had an abnormal karyotype, either XOjXO (Turner's syndrome, n = 4) or XO/XX (Turner's mosaic, n = 2). The mean (±SD) age ofthe subjects (n = 18) was 30.2 ± 3.3 years. The mean (±SD) duration of premature ovarian failure, excluding those with Turner's syndrome/mosaic, was 5.4 ± 2.8 years. Vol. 57, No.4, April 1992
Hormone Replacement Therapy
Each subject recruited underwent a study cycle in which a standard HRT, previously reported to be capable of supporting implantation in a donor-oocyte program (15), was administered. In addition, each subject underwent one or two additional study cycles in which the HRT was modified according to one of the following three protocols: fixed 1 mg E 2, fixed 2 mg E 2, and low P (Table 1). In the low P regimen, the doses of E2 valerate administered were the same as the standard HRT, whereas the doses of P in the low P regimen were ! of those of the standard. In the fixed E2 dose regimens (1 mg or 2 mg), a fixed daily dose of E2 was administered, in contrast to the variable E2 doses of the standard HRT; the doses of P were identical to those of the standard HRT. There were six subjects in each of the three modified HRT regimens. Thus, a total of 36 cycles was studied. In the cycle immediately preceding each study cycle, the HRT was standardized to the use of cycloprogynova (Schering, Sussex, United Kingdom), 2 mg starting 28 days before and finishing 7 days before the commencement of the study cycle. Observations
In each study cycle, plasma samples were obtained on days 1, 15, 19, and 29 for assay of PP14. All blood samples were collected in heparinized tubes. Pelvic ultrasonography was performed on day 29 to measure endometrial thickness. In addition, an outpatient endometrial biopsy was obtained on day 19 of the cycle for histologic assessment. Placental Protein 14 Assay
Placental protein 14 was measured using a modification of the radioimmunoassay (RIA) described Li et al.
PP14 leuels in artificial cycles
777
by Bolton et al. (16). In brief, PP14 was iodinated by the chloramine-T method and the resulting tracer purified using a column of Con A Sepharose. For the assay, 100 p.L of standards (200, 100, 50, 25, 12.5, 6.25, 3.13, 1.56 ng/mL) prepared from purified PP14 were incubated with 100 p.L of 1 ng/mL PP14 tracer and 100 p.L of antiserum at a dilution to bind 45% of the added tracer for either 14 hours at room temperature or 72 hours at 4°C. The antibody-bound tracer was separated from the unbound using Amerlex-M (Amersham International PLC, Buckinghamshire, United Kingdom) magnetic separating reagent. The sensitivity of the assay was 3 ng/mL. The intra-assay and interassay coefficients of variation (CVs) (within a working range of 5 to 100 ng/mL) was <10% (16). Samples with results> 100 ng/mL were diluted to bring the concentrations to within the working range. Hormone Assay
Plasma E2 measurement was carried out by nonextraction, solid phase RIA using 1251 as a marker (Diagnostic Products Limited, Abingdon, Oxon, United Kingdom). The sensitivity of the assay was 29 pmol/L and the mean intra-assay and interassay CVs were 5.4% and 6.4%, respectively. Plasma FSH concentration was measured by immunoradiometric assay incorporating two high-affinity monoclonal antibodies using 1251 as a marker (Serono Diagnostics Limited, Surrey, United Kingdom). The sensitivity of the assay was 0.25 IU/L. The mean intra-assay CV was 2.0%. The mean interassay CV was 3.3%. The calibration was based on the Second International Reference Preparation, code 78/549. Ultrasonography
This was performed by a transabdominal realtime scanner (Combison 320, Kretztechnik, Zipf, Austria). The maximal thickness of the endometrium on longitudinal section was measured three times, and the mean value obtained was used for subsequent calculation. The ultrasonography was performed by two observers, with an interobserver CV of approximately 8%. The endometrium was measured from the echogenic interface of the junction of the endometrium and myometrium, and the result therefore represents the two layers of endometrium. Endometrial Biopsy
Endometrial biopsies were obtained on day 19 of the cycle without the use of anaesthesia, using a 778
Li et al.
PP14 levels in artificial cycles
Sharman's curette (Down Surgical Ltd., Sheffield, United Kingdom). A single endometrial specimen was obtained from the fundus and upper part of the body of the uterus. Each specimen was fixed immediately in 2% (vol/vol) glutaraldehyde in sodium cacodylate (0.1 M) buffer (pH 7.4) for 4 to 6 hours. It was then washed in buffer, dehydrated, and embedded in JB-4 (Polysciences, Inc., Warrington, PA), a plastic polymer. The specimen was'sectioned at 2 p.m, and the sections were stained with acid fuchsin and toluidine blue for examination by light microscopy. All th~ endometrial biopsies were dated histologically according to the traditional dating criteria (17). Ethical Aspects
Informed consent was obtained from each woman participating in the study. Approval for this study had been obtained from the Ethical Committee of the former Southern District ofthe Sheffield Health Authority. Statistical Analysis
Although the production of PP14 is considered to be dependent on P, significant levels of PP14 were found in the plasma of the first half of the cycle: on day 1 (7 days after the cessation of any HRT) and day 15 (14 days after the administration of E2 but not any P). This represents the basal secretion of PP14, independent of P. To evaluate whether or not the administration of defined doses of P results in a significant rise in the plasma concentration of PP14, the 90th percentile (22.0 ng/mL) of the concentration on days 1 and 15 of the cycle (basal plasma PP14 concentrations) was taken as the arbitrary cutoff point between the presence or absence of a significant rise. The comparison of paired results for subjects who underwent two separate study cycles was performed by Wilcoxon's matched-pair nonparametric test. This and other statistical analyses (Fisher's exact and Student's t-test) were performed with the Statistical Package for Social Sciences, available within the IBM mainframe computer of the University of Sheffield. RESULTS Variation Throughout The Cycle
The concentrations of PP14 (geometric mean) in the plasma on day 1 (17.6 ng/mL, n = 18), day 15 (16.3 ng/mL, n = 18), and day 19 (16.6 ng/mL, Fertility and Sterility
Table 2 Relationship Between High or Low Plasma Concentrations of PP14 on Day 29 of the Cycle Treated With the Standard HRT and the Underlying Diagnosis of Premature Ovarian Failure Idiopathic premature ovarian failure
Turner's syndrome/ mosaic Turner *
9
3
2 4
Idiopathic premature ovarian failure
Turner's syndrome §
High PP14t Low PP14*
High PP14 LowPP14
9 3
o 4
* Not significant, Fisher's exact. Mosaic Turner includes those with XO/XX. t ~22.1 ng/mL. * :522.0 ng/mL. § P < 0.02, Fisher's exact.
pathic premature ovarian failure, 1 Turner mosaic) received the standard HRT in one cycle and the 2-mg fixed dose regimen in a subsequent cycle. The paired results of the plasma concentrations of PP14 on day 29 of these cycles are shown in Figure 1. Progesterone
Similarly, six subjects (all idiopathic premature ovarian failure) received the standard HRT in one cycle and the low P regimen in a separate cycle. The paired results ofthe plasma concentrations ofPP14 on day 29 of these cycles are shown in Figure 1. Association With Other Observations
The relationship between plasma concentrations of PP14 on day 29 of the cycle and (1) endometrial 100
n = 18) were found to be similar, whereas those of day 29 (48.5 ngjmL, n = 18) were significantly higher (P < 0.05) than those of days 1, 15, and 19 (MannWhitney nonparametric test). The ranges of the concentrations of PP14 were: day 1: 8.1 to 29.1 ngjmL; day 15: 7.7 to 26.9 ngjmL; day 19: 8.6 to 28.0 ngjmL; day 29: 10.7 to 130 ngjmL.
30
::J' E
The variability of plasma concentrations ofPP14 among subjects receiving the standard HRT (n = 18) tnay be expressed in terms of the CV. The results for each of the 4 days studied are: day 1: 34%; day 15: 26%; day 19: 30%; day 29: 80%. The intersubject variation of the plasma concentration of PP14 on day 29 of cycles treated with the standard HRT were further analyzed with reference to (1) whether or not the concentration was above the 90th percentile of the basal concentration and (2) whether the underlying diagnosis of premature ovarian failure was idiopathic or Turner's syndrome. The results are summarized in Table 2. All 4 subjects with Turner's syndrome had low PP14 levels, whereas only 3 of 12 subjects with idiopathic premature ovarian failure had low PP14 levels (P < 0.02).
1«
.s
~
100 (b)
0
tilS
~
cu .... 'E
II>
30
CJ
======----= E
c: 0
CJ
.... 0-
o:t
0-
10
oc:::::::::::::
::
1 '( standard
cu E en cu
ii:
fixed 1mg E
standard
Cl c:
Variation Between Subjects
10
(a)
fixed 2mg E
(e)
100
30
10 1L standard
lowP
Hormone replacement therapy
Variation in Response to Various HRTs Estrogen
Six subjects (5 idiopathic premature ovarian failure, 1 Turner) received the standard HRT in one cycle and the I-mg fixed dose regimen in a subsequent cycle, whereas a further six subjects (5 idioVol. 57, No.4, April 1992
Figure 1 A comparison of the paired results of plasma concentrations of PP14 on day 29 of the artificial cycles in subjects who received the standard HRT (15) in one cycle and one of three modified HRT regimens in another cycle (see Table 1). (a), Standard HRT versus fixed 1-mg E2 valerate. (b), Standard HRT versus fixed 2-mg E2 valerate. (c), Standard HRT versus low P. The paired results are compared by Wilcoxon's matched-pair nonparametric test. *, Significantly different between the pairs (P < 0.05); NS, not significantly different between the pairs. Li et al.
PP14 levels in artificial cycles
779
thickness on day 29 and (2) histologic dating of endometrial samples obtained on day 19 of the cycle are analyzed in Table 3. Whereas those with high PP14 levels had significantly (P < 0.02) thicker endometrium on day 29 than those with low PP14 levels, there was no difference in histologic dating results of specimen taken on day 19 between those with high or low PP14 levels. DISCUSSION
In this study, we analyzed the variation of the plasma concentration of PP14 on different days of the artificial cycle, in response to various doses of estrogen (E) and P, among individuals with different underlying causes of their premature ovarian failure, and its relation to a number of other clinical observations in the cycle. The artificial cycle, produced by the exogenous administration of steroid hormones in women with nonfunctioning ovaries, was used as an experimental model to permit the response of the endometrium to precisely defined doses of steroid hormones to be examined. This would not have been possible in the natural cycle. Although a number of endometrial proteins have been identified, PP14 is of particular interest because it has been the most widely studied human endometrial protein: the variation of plasma concentration ofPP14 throughout the natural cycle has been examined by a number of investigators (6, 7), the complete amino acid sequence and the nucleotide sequence of the complementary deoxyribonucleic acid encoding the protein have been reported (11), and PP14 has been found to possess immunosuppressive activity, potentially playing an important role in protecting the implanting blastocyst from immunorejection (9, 10). In addition, the measurement of plasma levels of PP14 has been considered to be of value in monitoring endometrial function (7, 13). Variation Throughout The Cycle
The pattern of variation of the plasma concentration of PP14 observed in the artificial cycle appears to be similar to that observed in natural, spontaneously ovulatory cycles (6, 7): with low levels in the early luteal phase (day 19) and high levels in the late luteal phase (day 29). In addition, the mean basal plasma level of PP14 in the artificial cycles (approximately 17 ng/mL) was similar to that of the natural cycle reported by Julkunen et al. (20 ng/mL) (7), although the basal level reported by Wood et al. (6) was somewhat lower (11.4 ng/mL). Furthermore, the PP14 level in the late luteal phase (48.5 ng/mL) 780
Li et al.
PP14 levels in artificial cycles
Table 3 A Comparison of Endometrial Thickness and the Results of Histologic Dating of Endometrial Samples Between Those With High or Low Plasma Concentration of PP14 on Day 29 of the Artificial Cycle *
Endometrial thickness on day 29 (mm) Histologic dating!7 of endometrial specimen taken on day 19
Low PP14t (n = 20)
High PP14t (n = 16)
Student's t·test
8.0 ± 0.6
9.9 ± 0.6
P < 0.02
3.8 ± 0.2
4.2 ± 0.1
NS§
* Values are means ± SEM. t ~22.0 ng/mL. t ;;,:22.1 ng/mL. § NS, not significant.
of artificial cycles was 2.8 times that of the basal level, which is in good agreement with that observed in the natural cycle (6, 7). A notable difference between the natural and artificial cycles is the PP14 level on day 1 of the cycle. In the former, the start of a cycle immediately follows on from the end of the previous one, and, given the relatively long half-life (42 hours) (7) of plasma PPI4, the level on day 1 of the cycle is significantly higher than the basal one. It returns to the basal level approximately 7 days afterward (midproliferative phase). In the present study, artificial cycles had all been preceded by a week without any HRT. Thus, the level of PP14 on day 1 is similar to that of day 15 (basal level}. Variation Between Subjects
Our results indicate that, given the same HRT, the plasma concentration of PP14 in women with Turner's syndrome (XO) is significantly different from those of women with idiopathic premature ovarian failure. On day 29 of the cycle treated with the standard HRT, none of the 4 subjects with Turner's syndrome had plasma concentration of PPI4> 22.0 ng/mL, whereas the majority (fz [75%]) of subjects with idiopathic premature ovarian failure had plasma concentration of PP14 > 22.0 ng/mL (90th percentile of basal concentrations) (Table 2). The association of low plasma concentrations of PP14 and Turner's syndrome has been previously described in a case report by Critchley et al. (18). In that case, the plasma concentrations of PP14 remained subnormal throughout the first trimester, but the pregnancy apparently progressed normally to term. Thus, the significance of subnormal concentrations of plasma PP14 remains uncertain: it will be of interest to correlate plasma concentrations Fertility and Sterility
of PP14 in natural and artificial cycles to implantation rate and pregnancy outcome in these cycles. In contrast, the two subjects with Turner's mosaic (XOjXX) appeared to have elevated plasma concentration of PP14 on day 29 of the cycle and, in this respect, behave more like subjects with idiopathic premature ovarian failure (normal chromosomes) than those with Turner's syndrome (XO). The precise explanation for these observations is unclear, but it seems likely that the X chromosome may influence the responsiveness of the endometrium to steroid hormones. Work is currently in progress in our departments to examine the messenger ribonucleic acid concentrations of E receptor, P receptor, and PP14 in the endometrium of subjects with idiopathic premature ovarian failure and Turner's syndrome. The finding in this study that the plasma concentration of PP14 varies widely among different subjects (CV on day 29 = 80%) despite the same HRT suggests that the endometrium of these subjects may have responded differently to the steroid hormones. This is consistent with previous reports that examined the endometrial response to steroid hormones in menopausal women by histologic analysis of endometrial biopsy specimens (19, 20). Variation in Response to Various HRTs Estrogen
On day 29 of the cycle, the plasma concentrations of PP14 in cycles treated with fixed 1-mg doses of E2 valerate were significantly lower than those treated with the standard HRT (average daily dose of E2 valerate = 3.0 mg). This provides evidence to suggest that the secretion of PP14 by the endometrium is dependent not only on P but also on an adequate priming of the endometrium by E 2 • This is in keeping with a previous report by Seppala et a1. (21) that a significant correlation (r = 0.55, P = 0.01) existed between plasma concentrations of E2 in the follicuhu phase and PP14 in the luteal phase. On the other hand, the concentration of those treated with fixed 2-mg doses of E2 valerate was not significantly different from those treated with the standard HRT, indicating that the administration of E2 in an incremental, sequential manner as in the standard HRT may not be essential for adequate priming of the endometrium: fixed daily doses are equally effective. Second, an average daily dose of 2 mg or 3 mg E2 valerate appeared to produce rather similar results. Perhaps adequate priming of the endometrium may be achieved by average daily doses of at least 2 mg but not 1 mg of E2 valerate. Vol. 57, No.4, April 1992
Progesterone
On day 29 of the cycle, the plasma concentration of PP14 in those treated with low P regimen was consistently lower than those treated with the standard HRT. This confirms the fact that the secretion of PP14 by the endometrium is dependent on an adequate amount of P. Histologic Dating and Plasma PP14
In a previous study, Joshi et a1. (22) found that the serum progestogen -dependent endometrial protein (serologically identical to PP14 (23» levels correlated with the histologic dating results of the endometrium obtained in the late luteal phase. In the present study, we found that the plasma concentration of PP14 on day 29 of the cycle did not bear a close relationship to the results of histologic dating of specimens obtained 10 days earlier in the cycle, i.e., day 19. This may be due to a number of reasons. First, the result of histologic dating represents an overall assessment of the maturity of the endometrium and takes into consideration many different histologic features in the glandular and stromal components. In contrast, the plasma concentration of PP14 reflects the synthesis and release into the plasma of a specific (one of many) protein produced by the glandular component of the endometrium. Second, the results of histologic dating referred to specimens obtained in day 19 of the cycle, but the results of plasma concentration of PP14 referred to day 29 of the cycle. On day 19, the endometrial glands have good evidence of active cellular secretory activity, but the corresponding plasma concentration of PP14 is low. Previous immunohistochemical studies (13, 24) suggested that PP14 is not present in significant amounts in the endometrial tissue at this stage of the cycle. It seems, therefore, that the two measurements (histologic dating on day 19 and plasma concentration ofPP14 on day 29) reflect different activities of the endometrium at two different stages of the cycle. Third, it has been suggested that plasma levels of PP14 may not necessarily reflect local endometrial production of the protein as evaluated by immunohistochemistry (6). It remains to be seen whether or not the two measurements are complementary to each other in the evaluation of endometrial function. Endometrial Thickness and Plasma PP14
The results in Table 3 suggest that high plasma concentrations of PP14 on day 29 were associated with a significantly greater endometrial thickness. The association between endometrial thickness and Li et al.
PP14 levels in artificial cycles
781
endometrial secretion of PP14 is perhaps not surprising, although this has not been previously documented. A thicker endometrium may be more mature in development than the one which is thinner, or, given that the stage of maturity/development is the same, a thicker endometrium represents a larger amount of endometrial tissue, leading to a greater amount of PP14 being produced. To conclude, we have used the artificial cycle as an experimental model to analyze the various factors affecting the plasma concentration of PP14. It seems likely that the measurement of the concentration of plasma PP14 in the late luteal phase may provide a useful, noninvasive means of assessing endometrium function.
10.
11.
12.
13.
14. Acknowledgments. We are grateful for the help of Elizabeth A. Lenton, Ph.D., Peter Dockery, Ph.D., Samuel S. Ramsewak, M.R.C.O.G., and Lucas Klentzeris, M.R.C.O.G., in this study. The secretarial assistance of Mrs. Marilyn Thorpe is appreciated.
15.
16. REFERENCES 1. Bohn H, Kraus W, Winckler W. New soluble placental tissue proteins: their isolation, characterization and quantification. Placenta 1982;4:67-81. 2. Joshi SG, Ebert KM, Smith RA. Properties of the progestagen-dependent protein of the human endometrium. J Reprod Fertil1980;59:287-96. 3. Sutcliffe RG, Joshi SG, Paterson WF, Bank JF. Serological identity between human alpha uterine protein and human progestagen-dependent endometrial protein. J Reprod Fertil 1982;65:207-9. 4. Bell SC, Patel S, Hales MW, Kirwan PH, Drife JO. Immunochemical detection and characterization of pregnancy-associated endometrial at, and a2-globulins secreted by the human endometrium. J Reprod FertiI1985;74:261-70. 5. Julkunen M, Koistinen R, Suikkari AM, Seppala M, Janne OA. Identification by hybridization histochemistry of human endometrial cells expressing mRNA encoding a uterine betalactoglobulin homologue and insulin-like growth factorbinding protein-I. Mol EndocrinoI1990;4:700-7. 6. Wood PL, Walker RA, Bell SC. Serum levels of pregnancyassociated endometrial a2-globulin (a2-PEG) during normal menstrual and combined oral contraceptive cycles and relationship to immunohistological localization. Hum Reprod 1989;4:140-6. 7. Julkunen M, Apter D, SeppUla M, Stenman U-H, Bohn H. Serum levels of placenta protein 14 reflect ovulation in non conceptional menstrual cycles. Fertil Steril1986;45:47-50. 8. Julkunen M, Rutanen E-M, Koskimies A, Ranta T, Bohn H, Seppalii M. Distribution of placental protein 14 in tissues and body fluids during pregnancy. Br J Obstet Gynaecol 1985;92:1145-51. 9. Bolton AE, Pockley AG, Clough KJ, Mowles EA, Stoker RJ, Westwood OMR, et al. Identification of placental protein 14
782
Li et al.
PP14 levels in artificial cycles
17. 18.
<
19.
20.
21.
22.
23.
24.
as an immunosuppressive factor in human reproduction. Lancet 1987;1:593-5. Pockley AG, Mowles EA, Stoker RJ, Westwood OMR, Chapman MG, Bolton AE. Suppression of in vitro lymphocyte reactivity to phytohemagglutinin by placental protein 14. J Reprod ImmunoI1988;13:31-9. Julkunen M, Seppala M, Janne OA. Complete amino acid sequence of human placental protein 14: a progesterone-regulated uterine protein homologous to beta-Iactoglobulins. Proc Nat! Acad Sci USA 1988;85:8845-9. Li TC, Pockley AG, Bolton AE, Cooke !D. The variation of endometrial protein PP14 in different parts of the human endometrium. Int J Gynaecol Obstet 1991;34:257-60. Waites GT, Bell SC, Walker RA, Wood PL. Immunohistological distribution of the secretory endometrial protein, pregnancy-associated endometrial a2-globulin, a glycosylated tJ-Iactoglobulin homologue, in the human fetus and adult employing monoclonal antibodies. Hum Reprod 1990;5:105-11. Mattison DR, Evans MI, Schwimmer WB, White BJ, Jensen B, Schulman JD. Familial premature ovarian failure. Am J Hum Genet 1984;36:1341-8. Navot D, Laufer N, Kopolovic J, Rabinowitz R, Birkenfeld A, Lewin A, et al. Artificially induced endometrial cycles and establishment of pregnancies in the absence of ovaries. N Engl J Med 1986;314:806-11. Bolton AE, Chapman MG, Stoker RJ, Andrew CE, Wass D, Bohn H. The radioimmunoassay of human placental protein 14 (PP14). Clin Chim Acta 1983;135:283-91. Noyes RW, Hertig AT, Rock J. Dating the endometrial biopsy. Fertil Steril1950;1:3-25. Critchley HOD, Chard T, Lieberman BA, Buckley CH, Anderson DC. Serum PP14 levels in a patient with Turner's syndrome pregnant after frozen embryo transfer. Hum Reprod 1990;5:250-4. Fraser D, Whitehead MI, Endacott J, Morton J, Ryder TA, Pryse-Davies J. Are fixed-dose oestrogen/progestogen combinations ideal for all HRT users? Br J Obstet Gynaecol 1989;96:776-82. Li TC, Dockery P, Ramsewak SS, Klentzeris L, Lenton EA, Cooke ID. The variation of endometrial response to a standard hormone replacement therapy in women with premature ovarian failure. An ultrasonographic and histological study. Br J Obstet Gynaecol 1991;98:656-61. Seppala M, Martikainen H, Ronnberg L, Riittinen L, Kauppila A. Suppression of prolactin secretion during ovarian hyperstimulation is followed by elevated serum levels of endometrial protein PP14 in the late luteal phase. Hum Reprod 1989;4:389-91. Joshi SG, Rao R, Henriques EE, Raikar RS, Gordon M. Luteal phase concentrations of a progestagen-associated endometrial protein (PEP) in the serum of cycling women with adequate or inadequate endometrium. J Clin Endocrinol Metab 1986;63:1247-9. Julkunen M, Raikar RS, Joshi SG, Seppala M. Placental protein 14 and progestagen-dependent endometrial protein are immunologically indistinguishable. Hum Reprod 1986;1: 7-8. Manners CV. Endometrial assessment in a group of infertile women on stimulated cycles for IVF: immunohistochemical findings. Hum Reprod 1990;5:128-32.
Fertility and Sterility
FERTILITY AND STERILITY Copyright" 1992 The American Fertility Society
Printed on acid-free paper in U.S.A.
An analysis of the variation of plasma concentrations of placental protein 14 in artificial cycles
Tin-Chiu Li, Ph.D.*t Caroline Dalton, B.Sc.:\: Anthony E. Bolton, D.Sc.:\:
Elizabeth Ling, B.Sc.* Alistair Warren, Ph.D.§ Ian D. Cooke, F.RC.O.G.*
University of Sheffield, Sheffield City Polytechnic, and Jessop Hospital for Women, Sheffield, United Kingdom
Objective: To analyze the factors affecting the variation of plasma concentration of placental protein 14 (PPI4) in artificial cycles. Design: The effects of different hormone replacement therapy (HRT) regimens were examined in a crossover design. Setting: Jessop Hospital for Women, Sheffield, United Kingdom. Patients: Eighteen women with premature ovarian failure: 6 associated with Turner's syndrome and 12 with idiopathic premature ovarian failure. Interventions: Four different HRT regimens; 36 study cycles. Main Outcome Measures: Plasma PP14 concentrations on days 1, 15, 19, and 29 ofthe artificial cycles. Results: In cycles treated with a standard HRT, the levels were similar to those of the natural cycle. Subjects with Turner's syndrome did not have elevated PP14 levels, whereas the majority [75%]) of those with idiopathic premature ovarian failure had elevated levels on day 29 of the cycle. Levels of PP14 were reduced when either the doses of estradiol valerate were reduced to t or the doses of progesterone (P) were reduced to ~ of the standard HRT. Conclusions: Plasma levels ofPP14 are dependent not only on P stimulation but also on adequate estrogen priming. Fertil Steril 1992;57:776-82
(f2
Key Words: Placental protein, artificial cycles
The endometrial protein placental protein 14 (PP14) (1), also known as progestogen-dependent endometrial protein (2), a-uterine protein (3), and pregnancy-associated endometrial a-2-globulin (4), is one of the principal proteins secreted by endometrial glands (5) in the luteal phase of the menstrual cycle. The production of this protein is thought to be dependent on progesterone (P) (2). The concentration of PP14 in the plasma and in
Received March 25, 1991; revised and accepted December 6, 1991. * Department of Obstetrics and Gynaecology, University of Sheffield. t Reprint requests: Tin-Chiu Li, Ph.D., Jessop Hospital for Women, Sheffield S3 7RE, United Kingdom. :\: Department of Biomedical Sciences, Sheffield City Polytechnic. § Department of Biomedical Science, University of Sheffield.
776
Li et al.
PP14 levels in artificial cycles
endometrial tissue has been found to increase from the early luteal phase to reach a maximum in the late luteal phase of a normal menstrual cycle (6, 7). In pregnancy, the level of PP14 continues to rise to a maximum at approximately 8 to 10 weeks of gestation (8). Recently, it has been found that PP14 inhibits lymphoproliferation (9, 10), suggesting that it may play an important role in the process of human implantation by protecting the implanting embryo from rejection by the maternal immune response. In addition, the successful determination of the N-terminal amino-acid sequence of PP14 has revealed homology with the ,a-lactoglobulin group, indicating a further possible role of this protein in the transport of retinol to the implanting blastocyst (11). In a previous study, we found a large, random variation of the levels of PP14 in different parts Fertility and Sterility
Table 1
Doses ofE and P in the Various HRTs Used in the Present Study* Standard regimen
Day of cycle
E2
lto5 6to 9 10 to 13 14 15 to 16 17 18 to 26 27 to 28
1 2 6 2 2 2 4 1
Fixed I-mg E regimen
Low P regimen P
E2
25 50 50 25
1 2 6 2 2 2 4 1
P
E2
Fixed 2-mg E regimen P
E2
P
25 50 50 25
2 2 2 2 2 2 2 2
25 50 50 25
1TIII
5 10 10 5
1 1 1 1 1 1 1 1
* The standard HRT was identical to that proposed by Navot et al. (15).
of the endometrium (mean coefficient of variation = 73%) (12), casting doubt on the measurement of PP14 levels in the endometrial tissue as being of much diagnostic value. On the other hand, the measurement ofPP14 concentration in plasma has been considered by some investigators to be of value in assessing ovulation (7) and the secretory function of the endometrium (13). In this prospective study, we employed the artificial cycle as an experimental model to examine the variation of plasma concentrations of PP14 among different individuals receiving the same hormone replacement therapy (HRT), in different cycles of the same individual treated with various forms of HRT, and in relation to a number of clinical observations.
MATERIALS AND METHODS Subjects
All subjects included in this study have premature ovarian failure, defined as amenorrhoea of 6 months or more, associated with persistently low plasma estradiol (E 2) «100pmol/L) and elevated plasma follicle-stimulating hormone (FSH) (>20 IU/L) in a woman <40 years of age (14). A total of 18 subjects were recruited, of whom 12 had normal chromosomal analysis, absent antiovarian antibody, and no obvious cause for the premature ovarian failure and so were considered to have idiopathic premature ovarian failure. The remaining 6 subjects had an abnormal karyotype, either XOjXO (Turner's syndrome, n = 4) or XO/XX (Turner's mosaic, n = 2). The mean (±SD) age ofthe subjects (n = 18) was 30.2 ± 3.3 years. The mean (±SD) duration of premature ovarian failure, excluding those with Turner's syndrome/mosaic, was 5.4 ± 2.8 years. Vol. 57, No.4, April 1992
Hormone Replacement Therapy
Each subject recruited underwent a study cycle in which a standard HRT, previously reported to be capable of supporting implantation in a donor-oocyte program (15), was administered. In addition, each subject underwent one or two additional study cycles in which the HRT was modified according to one of the following three protocols: fixed 1 mg E 2, fixed 2 mg E 2, and low P (Table 1). In the low P regimen, the doses of E2 valerate administered were the same as the standard HRT, whereas the doses of P in the low P regimen were ! of those of the standard. In the fixed E2 dose regimens (1 mg or 2 mg), a fixed daily dose of E2 was administered, in contrast to the variable E2 doses of the standard HRT; the doses of P were identical to those of the standard HRT. There were six subjects in each of the three modified HRT regimens. Thus, a total of 36 cycles was studied. In the cycle immediately preceding each study cycle, the HRT was standardized to the use of cycloprogynova (Schering, Sussex, United Kingdom), 2 mg starting 28 days before and finishing 7 days before the commencement of the study cycle. Observations
In each study cycle, plasma samples were obtained on days 1, 15, 19, and 29 for assay of PP14. All blood samples were collected in heparinized tubes. Pelvic ultrasonography was performed on day 29 to measure endometrial thickness. In addition, an outpatient endometrial biopsy was obtained on day 19 of the cycle for histologic assessment. Placental Protein 14 Assay
Placental protein 14 was measured using a modification of the radioimmunoassay (RIA) described Li et al.
PP14 leuels in artificial cycles
777
by Bolton et al. (16). In brief, PP14 was iodinated by the chloramine-T method and the resulting tracer purified using a column of Con A Sepharose. For the assay, 100 p.L of standards (200, 100, 50, 25, 12.5, 6.25, 3.13, 1.56 ng/mL) prepared from purified PP14 were incubated with 100 p.L of 1 ng/mL PP14 tracer and 100 p.L of antiserum at a dilution to bind 45% of the added tracer for either 14 hours at room temperature or 72 hours at 4°C. The antibody-bound tracer was separated from the unbound using Amerlex-M (Amersham International PLC, Buckinghamshire, United Kingdom) magnetic separating reagent. The sensitivity of the assay was 3 ng/mL. The intra-assay and interassay coefficients of variation (CVs) (within a working range of 5 to 100 ng/mL) was <10% (16). Samples with results> 100 ng/mL were diluted to bring the concentrations to within the working range. Hormone Assay
Plasma E2 measurement was carried out by nonextraction, solid phase RIA using 1251 as a marker (Diagnostic Products Limited, Abingdon, Oxon, United Kingdom). The sensitivity of the assay was 29 pmol/L and the mean intra-assay and interassay CVs were 5.4% and 6.4%, respectively. Plasma FSH concentration was measured by immunoradiometric assay incorporating two high-affinity monoclonal antibodies using 1251 as a marker (Serono Diagnostics Limited, Surrey, United Kingdom). The sensitivity of the assay was 0.25 IU/L. The mean intra-assay CV was 2.0%. The mean interassay CV was 3.3%. The calibration was based on the Second International Reference Preparation, code 78/549. Ultrasonography
This was performed by a transabdominal realtime scanner (Combison 320, Kretztechnik, Zipf, Austria). The maximal thickness of the endometrium on longitudinal section was measured three times, and the mean value obtained was used for subsequent calculation. The ultrasonography was performed by two observers, with an interobserver CV of approximately 8%. The endometrium was measured from the echogenic interface of the junction of the endometrium and myometrium, and the result therefore represents the two layers of endometrium. Endometrial Biopsy
Endometrial biopsies were obtained on day 19 of the cycle without the use of anaesthesia, using a 778
Li et al.
PP14 levels in artificial cycles
Sharman's curette (Down Surgical Ltd., Sheffield, United Kingdom). A single endometrial specimen was obtained from the fundus and upper part of the body of the uterus. Each specimen was fixed immediately in 2% (vol/vol) glutaraldehyde in sodium cacodylate (0.1 M) buffer (pH 7.4) for 4 to 6 hours. It was then washed in buffer, dehydrated, and embedded in JB-4 (Polysciences, Inc., Warrington, PA), a plastic polymer. The specimen was'sectioned at 2 p.m, and the sections were stained with acid fuchsin and toluidine blue for examination by light microscopy. All th~ endometrial biopsies were dated histologically according to the traditional dating criteria (17). Ethical Aspects
Informed consent was obtained from each woman participating in the study. Approval for this study had been obtained from the Ethical Committee of the former Southern District ofthe Sheffield Health Authority. Statistical Analysis
Although the production of PP14 is considered to be dependent on P, significant levels of PP14 were found in the plasma of the first half of the cycle: on day 1 (7 days after the cessation of any HRT) and day 15 (14 days after the administration of E2 but not any P). This represents the basal secretion of PP14, independent of P. To evaluate whether or not the administration of defined doses of P results in a significant rise in the plasma concentration of PP14, the 90th percentile (22.0 ng/mL) of the concentration on days 1 and 15 of the cycle (basal plasma PP14 concentrations) was taken as the arbitrary cutoff point between the presence or absence of a significant rise. The comparison of paired results for subjects who underwent two separate study cycles was performed by Wilcoxon's matched-pair nonparametric test. This and other statistical analyses (Fisher's exact and Student's t-test) were performed with the Statistical Package for Social Sciences, available within the IBM mainframe computer of the University of Sheffield. RESULTS Variation Throughout The Cycle
The concentrations of PP14 (geometric mean) in the plasma on day 1 (17.6 ng/mL, n = 18), day 15 (16.3 ng/mL, n = 18), and day 19 (16.6 ng/mL, Fertility and Sterility
Table 2 Relationship Between High or Low Plasma Concentrations of PP14 on Day 29 of the Cycle Treated With the Standard HRT and the Underlying Diagnosis of Premature Ovarian Failure Idiopathic premature ovarian failure
Turner's syndrome/ mosaic Turner *
9
3
2 4
Idiopathic premature ovarian failure
Turner's syndrome §
High PP14t Low PP14*
High PP14 LowPP14
9 3
o 4
* Not significant, Fisher's exact. Mosaic Turner includes those with XO/XX. t ~22.1 ng/mL. * :522.0 ng/mL. § P < 0.02, Fisher's exact.
pathic premature ovarian failure, 1 Turner mosaic) received the standard HRT in one cycle and the 2-mg fixed dose regimen in a subsequent cycle. The paired results of the plasma concentrations of PP14 on day 29 of these cycles are shown in Figure 1. Progesterone
Similarly, six subjects (all idiopathic premature ovarian failure) received the standard HRT in one cycle and the low P regimen in a separate cycle. The paired results ofthe plasma concentrations ofPP14 on day 29 of these cycles are shown in Figure 1. Association With Other Observations
The relationship between plasma concentrations of PP14 on day 29 of the cycle and (1) endometrial 100
n = 18) were found to be similar, whereas those of day 29 (48.5 ngjmL, n = 18) were significantly higher (P < 0.05) than those of days 1, 15, and 19 (MannWhitney nonparametric test). The ranges of the concentrations of PP14 were: day 1: 8.1 to 29.1 ngjmL; day 15: 7.7 to 26.9 ngjmL; day 19: 8.6 to 28.0 ngjmL; day 29: 10.7 to 130 ngjmL.
30
::J' E
The variability of plasma concentrations ofPP14 among subjects receiving the standard HRT (n = 18) tnay be expressed in terms of the CV. The results for each of the 4 days studied are: day 1: 34%; day 15: 26%; day 19: 30%; day 29: 80%. The intersubject variation of the plasma concentration of PP14 on day 29 of cycles treated with the standard HRT were further analyzed with reference to (1) whether or not the concentration was above the 90th percentile of the basal concentration and (2) whether the underlying diagnosis of premature ovarian failure was idiopathic or Turner's syndrome. The results are summarized in Table 2. All 4 subjects with Turner's syndrome had low PP14 levels, whereas only 3 of 12 subjects with idiopathic premature ovarian failure had low PP14 levels (P < 0.02).
1«
.s
~
100 (b)
0
tilS
~
cu .... 'E
II>
30
CJ
======----= E
c: 0
CJ
.... 0-
o:t
0-
10
oc:::::::::::::
::
1 '( standard
cu E en cu
ii:
fixed 1mg E
standard
Cl c:
Variation Between Subjects
10
(a)
fixed 2mg E
(e)
100
30
10 1L standard
lowP
Hormone replacement therapy
Variation in Response to Various HRTs Estrogen
Six subjects (5 idiopathic premature ovarian failure, 1 Turner) received the standard HRT in one cycle and the I-mg fixed dose regimen in a subsequent cycle, whereas a further six subjects (5 idioVol. 57, No.4, April 1992
Figure 1 A comparison of the paired results of plasma concentrations of PP14 on day 29 of the artificial cycles in subjects who received the standard HRT (15) in one cycle and one of three modified HRT regimens in another cycle (see Table 1). (a), Standard HRT versus fixed 1-mg E2 valerate. (b), Standard HRT versus fixed 2-mg E2 valerate. (c), Standard HRT versus low P. The paired results are compared by Wilcoxon's matched-pair nonparametric test. *, Significantly different between the pairs (P < 0.05); NS, not significantly different between the pairs. Li et al.
PP14 levels in artificial cycles
779
thickness on day 29 and (2) histologic dating of endometrial samples obtained on day 19 of the cycle are analyzed in Table 3. Whereas those with high PP14 levels had significantly (P < 0.02) thicker endometrium on day 29 than those with low PP14 levels, there was no difference in histologic dating results of specimen taken on day 19 between those with high or low PP14 levels. DISCUSSION
In this study, we analyzed the variation of the plasma concentration of PP14 on different days of the artificial cycle, in response to various doses of estrogen (E) and P, among individuals with different underlying causes of their premature ovarian failure, and its relation to a number of other clinical observations in the cycle. The artificial cycle, produced by the exogenous administration of steroid hormones in women with nonfunctioning ovaries, was used as an experimental model to permit the response of the endometrium to precisely defined doses of steroid hormones to be examined. This would not have been possible in the natural cycle. Although a number of endometrial proteins have been identified, PP14 is of particular interest because it has been the most widely studied human endometrial protein: the variation of plasma concentration ofPP14 throughout the natural cycle has been examined by a number of investigators (6, 7), the complete amino acid sequence and the nucleotide sequence of the complementary deoxyribonucleic acid encoding the protein have been reported (11), and PP14 has been found to possess immunosuppressive activity, potentially playing an important role in protecting the implanting blastocyst from immunorejection (9, 10). In addition, the measurement of plasma levels of PP14 has been considered to be of value in monitoring endometrial function (7, 13). Variation Throughout The Cycle
The pattern of variation of the plasma concentration of PP14 observed in the artificial cycle appears to be similar to that observed in natural, spontaneously ovulatory cycles (6, 7): with low levels in the early luteal phase (day 19) and high levels in the late luteal phase (day 29). In addition, the mean basal plasma level of PP14 in the artificial cycles (approximately 17 ng/mL) was similar to that of the natural cycle reported by Julkunen et al. (20 ng/mL) (7), although the basal level reported by Wood et al. (6) was somewhat lower (11.4 ng/mL). Furthermore, the PP14 level in the late luteal phase (48.5 ng/mL) 780
Li et al.
PP14 levels in artificial cycles
Table 3 A Comparison of Endometrial Thickness and the Results of Histologic Dating of Endometrial Samples Between Those With High or Low Plasma Concentration of PP14 on Day 29 of the Artificial Cycle *
Endometrial thickness on day 29 (mm) Histologic dating!7 of endometrial specimen taken on day 19
Low PP14t (n = 20)
High PP14t (n = 16)
Student's t·test
8.0 ± 0.6
9.9 ± 0.6
P < 0.02
3.8 ± 0.2
4.2 ± 0.1
NS§
* Values are means ± SEM. t ~22.0 ng/mL. t ;;,:22.1 ng/mL. § NS, not significant.
of artificial cycles was 2.8 times that of the basal level, which is in good agreement with that observed in the natural cycle (6, 7). A notable difference between the natural and artificial cycles is the PP14 level on day 1 of the cycle. In the former, the start of a cycle immediately follows on from the end of the previous one, and, given the relatively long half-life (42 hours) (7) of plasma PPI4, the level on day 1 of the cycle is significantly higher than the basal one. It returns to the basal level approximately 7 days afterward (midproliferative phase). In the present study, artificial cycles had all been preceded by a week without any HRT. Thus, the level of PP14 on day 1 is similar to that of day 15 (basal level}. Variation Between Subjects
Our results indicate that, given the same HRT, the plasma concentration of PP14 in women with Turner's syndrome (XO) is significantly different from those of women with idiopathic premature ovarian failure. On day 29 of the cycle treated with the standard HRT, none of the 4 subjects with Turner's syndrome had plasma concentration of PPI4> 22.0 ng/mL, whereas the majority (fz [75%]) of subjects with idiopathic premature ovarian failure had plasma concentration of PP14 > 22.0 ng/mL (90th percentile of basal concentrations) (Table 2). The association of low plasma concentrations of PP14 and Turner's syndrome has been previously described in a case report by Critchley et al. (18). In that case, the plasma concentrations of PP14 remained subnormal throughout the first trimester, but the pregnancy apparently progressed normally to term. Thus, the significance of subnormal concentrations of plasma PP14 remains uncertain: it will be of interest to correlate plasma concentrations Fertility and Sterility
of PP14 in natural and artificial cycles to implantation rate and pregnancy outcome in these cycles. In contrast, the two subjects with Turner's mosaic (XOjXX) appeared to have elevated plasma concentration of PP14 on day 29 of the cycle and, in this respect, behave more like subjects with idiopathic premature ovarian failure (normal chromosomes) than those with Turner's syndrome (XO). The precise explanation for these observations is unclear, but it seems likely that the X chromosome may influence the responsiveness of the endometrium to steroid hormones. Work is currently in progress in our departments to examine the messenger ribonucleic acid concentrations of E receptor, P receptor, and PP14 in the endometrium of subjects with idiopathic premature ovarian failure and Turner's syndrome. The finding in this study that the plasma concentration of PP14 varies widely among different subjects (CV on day 29 = 80%) despite the same HRT suggests that the endometrium of these subjects may have responded differently to the steroid hormones. This is consistent with previous reports that examined the endometrial response to steroid hormones in menopausal women by histologic analysis of endometrial biopsy specimens (19, 20). Variation in Response to Various HRTs Estrogen
On day 29 of the cycle, the plasma concentrations of PP14 in cycles treated with fixed 1-mg doses of E2 valerate were significantly lower than those treated with the standard HRT (average daily dose of E2 valerate = 3.0 mg). This provides evidence to suggest that the secretion of PP14 by the endometrium is dependent not only on P but also on an adequate priming of the endometrium by E 2 • This is in keeping with a previous report by Seppala et a1. (21) that a significant correlation (r = 0.55, P = 0.01) existed between plasma concentrations of E2 in the follicuhu phase and PP14 in the luteal phase. On the other hand, the concentration of those treated with fixed 2-mg doses of E2 valerate was not significantly different from those treated with the standard HRT, indicating that the administration of E2 in an incremental, sequential manner as in the standard HRT may not be essential for adequate priming of the endometrium: fixed daily doses are equally effective. Second, an average daily dose of 2 mg or 3 mg E2 valerate appeared to produce rather similar results. Perhaps adequate priming of the endometrium may be achieved by average daily doses of at least 2 mg but not 1 mg of E2 valerate. Vol. 57, No.4, April 1992
Progesterone
On day 29 of the cycle, the plasma concentration of PP14 in those treated with low P regimen was consistently lower than those treated with the standard HRT. This confirms the fact that the secretion of PP14 by the endometrium is dependent on an adequate amount of P. Histologic Dating and Plasma PP14
In a previous study, Joshi et a1. (22) found that the serum progestogen -dependent endometrial protein (serologically identical to PP14 (23» levels correlated with the histologic dating results of the endometrium obtained in the late luteal phase. In the present study, we found that the plasma concentration of PP14 on day 29 of the cycle did not bear a close relationship to the results of histologic dating of specimens obtained 10 days earlier in the cycle, i.e., day 19. This may be due to a number of reasons. First, the result of histologic dating represents an overall assessment of the maturity of the endometrium and takes into consideration many different histologic features in the glandular and stromal components. In contrast, the plasma concentration of PP14 reflects the synthesis and release into the plasma of a specific (one of many) protein produced by the glandular component of the endometrium. Second, the results of histologic dating referred to specimens obtained in day 19 of the cycle, but the results of plasma concentration of PP14 referred to day 29 of the cycle. On day 19, the endometrial glands have good evidence of active cellular secretory activity, but the corresponding plasma concentration of PP14 is low. Previous immunohistochemical studies (13, 24) suggested that PP14 is not present in significant amounts in the endometrial tissue at this stage of the cycle. It seems, therefore, that the two measurements (histologic dating on day 19 and plasma concentration ofPP14 on day 29) reflect different activities of the endometrium at two different stages of the cycle. Third, it has been suggested that plasma levels of PP14 may not necessarily reflect local endometrial production of the protein as evaluated by immunohistochemistry (6). It remains to be seen whether or not the two measurements are complementary to each other in the evaluation of endometrial function. Endometrial Thickness and Plasma PP14
The results in Table 3 suggest that high plasma concentrations of PP14 on day 29 were associated with a significantly greater endometrial thickness. The association between endometrial thickness and Li et al.
PP14 levels in artificial cycles
781
endometrial secretion of PP14 is perhaps not surprising, although this has not been previously documented. A thicker endometrium may be more mature in development than the one which is thinner, or, given that the stage of maturity/development is the same, a thicker endometrium represents a larger amount of endometrial tissue, leading to a greater amount of PP14 being produced. To conclude, we have used the artificial cycle as an experimental model to analyze the various factors affecting the plasma concentration of PP14. It seems likely that the measurement of the concentration of plasma PP14 in the late luteal phase may provide a useful, noninvasive means of assessing endometrium function.
10.
11.
12.
13.
14. Acknowledgments. We are grateful for the help of Elizabeth A. Lenton, Ph.D., Peter Dockery, Ph.D., Samuel S. Ramsewak, M.R.C.O.G., and Lucas Klentzeris, M.R.C.O.G., in this study. The secretarial assistance of Mrs. Marilyn Thorpe is appreciated.
15.
16. REFERENCES 1. Bohn H, Kraus W, Winckler W. New soluble placental tissue proteins: their isolation, characterization and quantification. Placenta 1982;4:67-81. 2. Joshi SG, Ebert KM, Smith RA. Properties of the progestagen-dependent protein of the human endometrium. J Reprod Fertil1980;59:287-96. 3. Sutcliffe RG, Joshi SG, Paterson WF, Bank JF. Serological identity between human alpha uterine protein and human progestagen-dependent endometrial protein. J Reprod Fertil 1982;65:207-9. 4. Bell SC, Patel S, Hales MW, Kirwan PH, Drife JO. Immunochemical detection and characterization of pregnancy-associated endometrial at, and a2-globulins secreted by the human endometrium. J Reprod FertiI1985;74:261-70. 5. Julkunen M, Koistinen R, Suikkari AM, Seppala M, Janne OA. Identification by hybridization histochemistry of human endometrial cells expressing mRNA encoding a uterine betalactoglobulin homologue and insulin-like growth factorbinding protein-I. Mol EndocrinoI1990;4:700-7. 6. Wood PL, Walker RA, Bell SC. Serum levels of pregnancyassociated endometrial a2-globulin (a2-PEG) during normal menstrual and combined oral contraceptive cycles and relationship to immunohistological localization. Hum Reprod 1989;4:140-6. 7. Julkunen M, Apter D, SeppUla M, Stenman U-H, Bohn H. Serum levels of placenta protein 14 reflect ovulation in non conceptional menstrual cycles. Fertil Steril1986;45:47-50. 8. Julkunen M, Rutanen E-M, Koskimies A, Ranta T, Bohn H, Seppalii M. Distribution of placental protein 14 in tissues and body fluids during pregnancy. Br J Obstet Gynaecol 1985;92:1145-51. 9. Bolton AE, Pockley AG, Clough KJ, Mowles EA, Stoker RJ, Westwood OMR, et al. Identification of placental protein 14
782
Li et al.
PP14 levels in artificial cycles
17. 18.
<
19.
20.
21.
22.
23.
24.
as an immunosuppressive factor in human reproduction. Lancet 1987;1:593-5. Pockley AG, Mowles EA, Stoker RJ, Westwood OMR, Chapman MG, Bolton AE. Suppression of in vitro lymphocyte reactivity to phytohemagglutinin by placental protein 14. J Reprod ImmunoI1988;13:31-9. Julkunen M, Seppala M, Janne OA. Complete amino acid sequence of human placental protein 14: a progesterone-regulated uterine protein homologous to beta-Iactoglobulins. Proc Nat! Acad Sci USA 1988;85:8845-9. Li TC, Pockley AG, Bolton AE, Cooke !D. The variation of endometrial protein PP14 in different parts of the human endometrium. Int J Gynaecol Obstet 1991;34:257-60. Waites GT, Bell SC, Walker RA, Wood PL. Immunohistological distribution of the secretory endometrial protein, pregnancy-associated endometrial a2-globulin, a glycosylated tJ-Iactoglobulin homologue, in the human fetus and adult employing monoclonal antibodies. Hum Reprod 1990;5:105-11. Mattison DR, Evans MI, Schwimmer WB, White BJ, Jensen B, Schulman JD. Familial premature ovarian failure. Am J Hum Genet 1984;36:1341-8. Navot D, Laufer N, Kopolovic J, Rabinowitz R, Birkenfeld A, Lewin A, et al. Artificially induced endometrial cycles and establishment of pregnancies in the absence of ovaries. N Engl J Med 1986;314:806-11. Bolton AE, Chapman MG, Stoker RJ, Andrew CE, Wass D, Bohn H. The radioimmunoassay of human placental protein 14 (PP14). Clin Chim Acta 1983;135:283-91. Noyes RW, Hertig AT, Rock J. Dating the endometrial biopsy. Fertil Steril1950;1:3-25. Critchley HOD, Chard T, Lieberman BA, Buckley CH, Anderson DC. Serum PP14 levels in a patient with Turner's syndrome pregnant after frozen embryo transfer. Hum Reprod 1990;5:250-4. Fraser D, Whitehead MI, Endacott J, Morton J, Ryder TA, Pryse-Davies J. Are fixed-dose oestrogen/progestogen combinations ideal for all HRT users? Br J Obstet Gynaecol 1989;96:776-82. Li TC, Dockery P, Ramsewak SS, Klentzeris L, Lenton EA, Cooke ID. The variation of endometrial response to a standard hormone replacement therapy in women with premature ovarian failure. An ultrasonographic and histological study. Br J Obstet Gynaecol 1991;98:656-61. Seppala M, Martikainen H, Ronnberg L, Riittinen L, Kauppila A. Suppression of prolactin secretion during ovarian hyperstimulation is followed by elevated serum levels of endometrial protein PP14 in the late luteal phase. Hum Reprod 1989;4:389-91. Joshi SG, Rao R, Henriques EE, Raikar RS, Gordon M. Luteal phase concentrations of a progestagen-associated endometrial protein (PEP) in the serum of cycling women with adequate or inadequate endometrium. J Clin Endocrinol Metab 1986;63:1247-9. Julkunen M, Raikar RS, Joshi SG, Seppala M. Placental protein 14 and progestagen-dependent endometrial protein are immunologically indistinguishable. Hum Reprod 1986;1: 7-8. Manners CV. Endometrial assessment in a group of infertile women on stimulated cycles for IVF: immunohistochemical findings. Hum Reprod 1990;5:128-32.
Fertility and Sterility