Detailed Assessment of Big Big Prolactin in Women with Hyperprolactinemia and Normal Ovarian Function

IAN S. FRASER, ZHUANG GUANG LUN, JIAN PING ZHOU, ADRIAN C. HERINGTON, GAY McCARRON, IAN CATERSON, KRIS TAN, AND Downloaded from https://academic.oup.com/jcem/article/69/3/585/2653406 by guest on 29 September 2021 ROBERT MARKHAM Department of Obstetrics and Gynecology, University of Sydney, New South Wales 2006, the Division of Obstetrics and Gynecology and Regional Endocrine Laboratory, Royal Prince Alfred Hospital, Camperdown, New South Wales 2050; and the Medical Research Centre, Prince Henry's Hospital, Melbourne, Victoria 3004, Australia

ABSTRACT. Six women with elevated circulating levels of big in PRL (by 93%), but not BBPRL, 30 min after suckling. TRH big PRL (BBPRL) and apparently normal ovarian function were caused a brisk rise in PRL (by 363 ± 116%) but only a sluggish variously studied through a menstrual cycle and menstruation, rise in BBPRL (by 17.5 ± 7.4%; n = 3). Bromocriptine rapidly through pregnancy and suckling, and during stimulation tests suppressed PRL (by 81.8 ± 34.4%), but only slowly suppressed with TRH and suppression with bromocriptine. BBPRL (by 21.0 ± 8.7% after 6 h; n = 3). Plasma binding studies No significant changes in monomeric PRL were demonstrated did not demonstrate any evidence of a circulating specific PRL- during the menstrual cycle, but all subjects showed a small and binding protein. significant rise in BBPRL during the preovulatory phase. High These data indicate that plasma concentrations of BBPRL PRL levels were present in day 1 menstrual plasma, but BBPRL may vary under the influence of a number of factors, but are was only present in low concentrations. All subjects (n = 5) much less sensitive to TRH stimulation, bromocriptine suppres- demonstrated a rise in both PRL and BBPRL during pregnancy, sion, pregnancy, and suckling than PRL. The occurrence of with a consistent tendency for PRL to increase to a proportion- BBPRL does not seem to be due to a specific circulating binding ately greater extent than BBPRL. One subject exhibited a rise protein. (J Clin Endocrinol Metab 69: 585, 1989)

T I ^HERE have been several recent reports of women capable of secreting such forms (7). Treatment with JL with apparently normal ovarian function who have bromocriptine can suppress circulating levels of both by chance been found to have substantial hyperprolac- polymeric and monomeric forms, the monomeric form to tinemia (1-5). In a high proportion of these women PRL such an extent that defective luteal function may occur was found to be circulating in polymeric forms [big (4). prolactin (BPRL) and big big PRL (BBPRL)]. The mol We have identified six women with hyperprolactine- wt of the different forms are approximately 22,000 for mia, normal ovarian function, and excessive circulating monomeric PRL (PRL; the main biologically active hor- levels of BBPRL. These women have been investigated mone), 50,000 for BPRL (probably a mixture of dimeric further during normal menstrual cycles, menstruation, and trimeric forms), and greater than 100,000 for and pregnancy and during bromocriptine suppression BBPRL (of uncertain structure). The polymeric forms and TRH stimulation tests. have substantially less biological activity than PRL, and BBPRL has less than 15% the receptor activity of PRL Materials and Methods (6). In normal women PRL usually accounts for at least 80-90% of the total circulating hormone. Clinical Material Little is known about factors or underlying disorders Six women were found to have elevated circulating PRL that may influence the secretion of these polymeric levels, as measured by immunoradiometric assay, during inves- forms. Most of these women have not had demonstrable tigations for various menstrual or endocrine disorders. These pituitary adenomas, although adenomas are sometimes women were unusual in that excessive elevations of PRL were found in the presence of fairly regular menstrual cycles. At the PRL levels found, marked irregularity of the cycle or even Received January 4,1989. Address all correspondence and requests for reprints to: Dr. I. S. amenorrhea would have been expected. For this reason, at a Fraser, Department of Obstetrics and Gynecology, University of Syd- later date all six had assays for the polymeric forms of PRL ney, New South Wales 2006, Australia. carried out. Two other women with moderate hyperprolactine-

585 586 FRASER ET AL. JCE & M • 1989 Vol 69 • No 3 mia and regular cycles were found to have a low concentration was measured in menstrual and peripheral plasma in subjects of BBPRL. All six subjects had detailed investigations of 3 and 5. plasma gonadotropins, ovarian and adrenal steroids, thyroid function, and high resolution pituitary computed tomographic Hormone assays scans. Clinical details are shown in Table 1. More detailed Serum PRL levels were measured using the Tandem immu- investigation with serial plasma progesterone measurements noradiometric assay kit (manufactured by Hybritech, Inc. (San and an endometrial biopsy confirmed that each subject had Diego, CA). Inter- and intraassay variations were less than 5%. entirely normal cyclical ovarian function, and all four who The normal range for this assay in our laboratory in women in presented with short term infertility (9-18 months) achieved a the reproductive age group is 2-15 Mg/L- Estradiol and proges- successful pregnancy without treatment. Subsequently, all six terone were measured by RIAs manufactured by Diagnostic were found to have markedly elevated circulating BBPRL Products Corp. (Los Angeles, CA) and Farmos Diagnostica levels. A series of detailed investigations was carried out, as (Turku, Finland), respectively. Downloaded from https://academic.oup.com/jcem/article/69/3/585/2653406 by guest on 29 September 2021 described below. These studies were approved by the Royal Prince Alfred Hospital and University of Sydney Medical Eth- Recovery and dilution ical Review Committees, and informed consent was obtained A serum sample containing elevated PRL concentrations from all subjects. was serially diluted with the zero calibrator and assayed in Five women (subjects 1-5) agreed to have three times weekly duplicate. The results in Table 2a showed a complete parallel- blood sampling throughout one menstrual cycle for measure- ism between serum sample and the standard. In view of possible ment of PRL, estradiol, and progesterone. BBPRL was meas- interference in the PRL assay by enzymes in the menstrual ured in early follicular (day 2-4) preovulatory (day 12-17), and plasma, serial dilutions were carried out. Because of the high midluteal (day 21-28) samples. Samples were generally col- viscosity of the menstrual plasma, the dilution was performed lected between 1100 and 1600 h after subjects had rested for with the column buffer. The results of the menstrual plasma 10-15 min. Four women (subjects 2, 4, 5, and 6) had samples dilution study were shown in Table 2b. These results also collected during pregnancy, and subject 4 also had samples showed a complete parallelism between menstrual plasma and collected before and during suckling after lactation was well the standard. established. Three subjects (subjects 1, 3, and 5) underwent TRH stim- Gel chromatography ulation tests where samples were collected at 0, 20, 40, and 60 Sephacryl S-200 gel (Pharmacia Fine Chemical, Uppsala, min, and TRH (200 /xg; Ayerst Co., Rouses Point, NY) was Sweden) was equilibrated with 0.5 M phosphate buffer contain- administered iv at 0 min. Total PRL was measured in all ing 0.25% BSA, pH 6.5, at room temperature, and a column of samples, and BBPRL in baseline and peak samples. The same 50 X 0.8 cm was set up. Serum (1.0 mL) was applied to the three subjects (no. 1, 3, and 5) underwent a bromocriptine column and run at a flow rate of 6.0 mL/h. One-milliliter suppression test where blood samples were collected at 0, 2, 4, fractions were collected by using an automatic fraction collec- and 6 h, and a single 2.5-mg bromocriptine tablet (Sandoz Co., tion (Redirac 2112, LKB, Bromma, Finland). Because of high Hanover, NJ) was administered at 0 h. Total PRL was meas- viscosity of menstrual plasma, it had to be diluted 1:5 or 1:10 ured in all samples, and BBPRL in baseline and nadir samples. with column buffer before its application to the column. All Three subjects (subjects 2, 3, and 5) collected several men- fractions were kept at 4 C before assay. The column was strual blood samples on days 1 and 2 of menstruation using a calibrated with the following molecular size markers blue dex- menstrual cup (manufactured by Shanghai New Asia Medical tran (mol wt, -2,000,000), [125I]human albumin (mol wt, Rubber, Shanghai, China). Total menstrual plasma PRL con- -69,000), [125I]FSH (mol wt, -24,000). The protein profile was centrations were measured in all samples. Simultaneous pe- also measured using a spectrophotometer at a wavelength of ripheral plasma samples were assessed for total PRL. BBPRL 280 nm.

TABLE 1. Clinical details of six women with elevated plasma BBPRL levels

Subject no 1 2 3 4 5 6 Age (yr) 37 38 35 32 33 24 Menstrual cycle Regular Transiently Transient ir- Regular Regular Regular irregular regular Disorder Dysmenorrhea Premenstrual Dysmenorrhea Previous Premenstrual Hirsutism syndrome anorexia amenorrhea syndrome Galactorrhea Parity Nulliparous One child One child One child Nulliparous One child Progesterone, luteal peak 39.0 54.4 67.2 59.2 39.5 (nmol/L) PRL (Mg/L) 82-112 34-50 130-188 90-140 40-78 45.5 BIG BIG PRL 587

TABLE 2a. Measured concentrations of serum PRL during serial di- azide, and 0.1% BSA. Bound and free [125I]hPRL were sepa- lution rated on minicolumns of Ultrogel AcA 54 (0.6 X 22 cm) essentially as described for the assay of the serum binding protein PRL cone. (Mg/L) % for human GH (hGH) (10). As a control, [125I]hGH binding to Observed Expected Recovery the same serum samples was determined using the same meth- Undiluted 92 odology (10). 1:2 48 46 104 1:4 23 23 100 1:6 15 17 113 Results Serum was diluted with a zero calibrator. Changes during the menstrual cycle

TABLE 2b. Measured concentrations of total PRL in menstrual plasma Downloaded from https://academic.oup.com/jcem/article/69/3/585/2653406 by guest on 29 September 2021 during serial dilution (n = 2) Circulating concentrations of total PRL, BBPRL, PRL, and the percentage of PRL in subjects 1-5 at three PRL cone. (Mg/L) % different stages during the menstrual cycle are shown in Dilution Observed Expected Recovery Table 3. Plasma BBPRL levels were significantly higher Undiluted 175 ± 2.55° in the preovulatory (mean ± SD, 82.4 ± 43.9 ng/L) than 1:2 89.5 ± 1.85 87.5 103 in the early follicular (67.8 ± 39.3; t = 4.086; P < 0.02) 1:4 48.0 ± 5.45 43.8 110 and midluteal phases (63 ± 38.9; t = 3.68; P < 0.025). 1:8 22.0 ± 0.95 21.8 101 Plasma PRL levels and the percentage of PRL did not 1:16 11.0 ± 0.85 10.9 101 vary significantly during the cycle. Plasma levels of es- Menstrual plasma is diluted with column buffer. ° Mean ± SD. tradiol and progesterone were entirely within normal limits for each stage of the cycle. Estimation of BBPRL PRL concentrations in each fraction were determined by Changes during pregnancy immunoradiometric assay as described above. Since the activity of antigen-antibody complex is proportional to the concentra- In subject 4, blood samples were collected at intervals tion of the hormone measured, the counts per min of the throughout pregnancy, and measurements of total PRL complex were plotted against the fraction number. BBPRL was were made in each sample (Fig. 1 and Table 4). BBPRL recovered in fractions 15-21, BPRL in fractions 21-24, and was measured at least once in each trimester. Total PRL PRL in fractions 24-30. The area under the curve was estimated increased steadily during pregnancy as expected, and the by cutting out and weighing the chromatogram. The weight of proportion of PRL to BBPRL increased substantially paper from fractions 15-30 represents the total area under the curve (hence, the total PRL concentration), and the weight of and progressively as pregnancy advanced beyond the first paper from fractions 15-21 represents the area under the curve trimester. BPRL was not detected in any of these sam- of BBPRL. Hence, the percentage of BBPRL can be calculated ples. In subjects 2, 5, and 6 a smaller number of samples as follows: (weight of paper between fractions 15-21 •*• by the was collected, but the same trends were noted in all weight of paper representing total area under the curve) X three. 100%. To obtain the absolute value of BBPRL in micrograms per L, the percentage of BBPRL is multiplied by the total PRL concentration. A similar calculation was carried out for BPRL. TABLE 3. Changes in BBPRL, PRL, and percentage of PRL during the menstrual cycle

Subject no. Measurement of specific binding of PRL by serum Mean ± SD Human PRL (NIDDK hPRL 1-6), a gift of the NIDDK (Bethesda, MD), was iodinated to a specific activity of about BBPRL (fig/L) 40 iiCi/ng by the Iodogen method (8). Before use [125I]hPRL F 76.6 12.5 98.3 107.3 44.5 67.8 ± 39.3 M 93.6 23.8 106.5 134.6 53.3 82.4 ± 43.9 was purified on an Ultrogel Ac A 54 column (1 X 55 cm). The L 81.0 14.4 94.2 97.2 28.1 63.0 ± 38.9 suitability of the tracer for receptor-like binding studies was PRL (Mg/L) assessed using classical PRL receptor preparations, rabbit F 8.4 14.5 15.0 7.2 33.6 15.7 ± 10.6 mammary gland membranes, or cytosols (9) and was found to M 18.7 26.6 19.3 5.8 15.9 17.3 ± 7.5 be normal (data not shown). L 10.2 25.2 12.4 6.7 22.7 15.4 ± 8.0 To determine the PRL-binding capacity of serum, 100-/xL %PRL aliquots of serum were incubated for 2 h at 21-23 C with [125I] F 9.8 53.6 13.3 6.3 43.0 25.2 ± 21.6 hPRL (20,000 cpm: ~0.20 ng) in the presence or absence of M 16.4 52.3 15.4 4.2 22.9 22.2 ± 18.0 excess unlabeled hPRL (to assess nonspecific binding) in 25 L 11.2 63.6 11.6 6.5 44.7 27.5 ± 25.3 mM HEPES buffer containing 10 mM MgCl2 0.02% sodium F, Early follicular; M, midcycle; L, midluteal. 588 FRASER ET AL. JCE & M • 1989 Vol 69 • No 3

prolactin(ng/ml)

prolactin

Big-big prolactin

Monomaric prolactin FIG. 1. Plasma levels of total PRL, BBPRL, and PRL through one menstrual cycle, at different stages during pregnancy, and before and during suckling in subject 4. •, BBPRL; D, PRL monomer. Downloaded from https://academic.oup.com/jcem/article/69/3/585/2653406 by guest on 29 September 2021

14 21 12 15 20 24 7 vaaks poatpartun Days of menstrual cycla Haeka of gastatlon lactation

TABLE 4. Changes in serum PRL concentrations (micrograms per L) during a menstrual cycle before pregnancy and during pregnancy in four subjects

Subject Type of Menstrual cycle days Pregnancy weeks no. PRL 2-4 14-17 21-28 10-12 13-15 24 32-38 2 Total 27 50.4 39.6 70.9 72.5 131.4 BBPRL 12.5 (46.4) 23.8 (47.7) 14.4 (36.4) 30.5 (43) 26.3 (36.4) 38.4 (29.2) PRL 14.5 (53.6) 26.6 (52.3) 25.2 (63.6) 40.4 (57) 46.2 (63.6) 93.0 (70.8)

4 Total 114.6 140.5 103.9 395 541 1168 1072.5 BBPRL 107. (93.7) 134.6 (95.8) 97.2 (93.6) 371 (93.9) 504 (93.6) 995 (85.2) 858 (80) PRL 7.2 (6.3) 5.8 (4.2) 6.7 (6.4) 24 (6.1) 37 (6.4) 173 (14.8) 214.5 (20)

5 Total 78.1 69.2 50.8 85.5 BBPRL 44.5 (57) 53.3 (77.1) 28.1 (53.3) 63.4 (74.2) PRL 33.6 (43) 15.9 (22.9) 22.7 (44.7) 22.1 (25.8)

6 Total 45.4 337.3 BBPRL 37.2 (81.9) 178 (52.8) PRL 8.2 (18.1) 159 (47.2) Percentages are given in parentheses.

Changes during breast feeding TABLE 5. TRH stimulation tests in three women with excessive circulating levels of BBPRL In subject 4 blood samples were obtained immediately before and after 30 min of suckling at 7 weeks postpartum Subject no. Mean ± SD (Fig. 1). Total PRL increased to 1083 fig/L from 938 /*g/ L, and PRL increased from 137.9 /tg/L (14.7%) to 267 BBPRL (Mg/L) fig/L (25.5%). No significant changes were noted in Baseline 107.0 109.5 26.3 80.9 ± 47.3 BBPRL. Peak 118.8 127.0 33.0 93.0 ± 52.0 PRL (Mg/L) Baseline 16.0 14.4 9.8 12.5 ±3.2 Peak 53.6 71.0 55.0 60.0 ± 9.7 TRH stimulation tests %PRL Baseline 12.9 18.5 27.5 19.6 ± 7.4 In subjects 1,3, and 5 the effects of TRH stimulation Peak on total PRL, BBPRL, and PRL are summarized in 31.7 35.8 62.0 43.1 ± 16.4 Table 5. BBPRL increased by only 17.5%, whereas PRL % Increase in BBPRL 11.0 16.0 25.5 17.5 ±7.4 increased spectacularly by 363.1 ± 116.0% (mean ± SD) % Increase in PRL to a peak plasma concentration of 60.0 ±9.7 /*g/L. 235.0 393.1 461.2 363.1 ± 116.0 BIG BIG PRL 589

Bromocriptine suppression tests TABLE 7. Measurements of PRL and BBPRL in menstrual and peripheral plasma in three women with elevated plasma BBPRL levels In subjects 1, 3, and 5 the effects of a single 2.5-mg dose of bromocriptine on total PRL, BBPRL, and PRL Total Day of PRL BBPRL Subject no. BPRL PRL are summarized in Table 6. Levels of BBPRL decreased menses (/xg/ (Mg/L) in all three subjects, but only by a mean of 21%, whereas L) levels of PRL decreased much more dramatically by 81.8% to a mean nadir of 3.3 ± 2.2 Menstrual 1 42.3 2 16.3 PRL polymers in menstrual plasma Peripheral 1 . 34.9

2 34.2 Downloaded from https://academic.oup.com/jcem/article/69/3/585/2653406 by guest on 29 September 2021 Total PRL levels were measured in menstrual and o peripheral plasma samples collected from three subjects. Menstrual 1 623 46.7 (7.5) 567 (92.5) Simultaneous peripheral and menstrual blood samples 2 98 60.4 (61.6) 37.6 (38.4) could be assayed for BBPRL only in subjects 3 and 5. There were no significant changes in total PRL and Peripheral 1 164.2 136.2 (83) 28 (17) 2 133.2 117.6 (88.3) 15.6 (11.7) BBPRL in the peripheral samples collected on days 1 c and 2 of menstruation. However, total PRL and PRL O Menstrual 1 84 15.5 (18.5) 9.53 (11.3) 58.9 (70.1) levels in menstrual plasma were substantially higher 2 23 13.4 (58.2) 3.10 (13.8) 6.4 (27.8) than peripheral levels on day 1 and subsequently decreased on day 2 (Table 7 and Fig. 2). By contrast, Peripheral 1 40 29.3 (73) 10.7 (27) BBPRL levels were substantially lower in menstrual 2 31.1 22.6 (72.5) 8.5 (27.4) than peripheral plasma on both days 1 and 2. It is of Percentages are given in parentheses. interest that menstrual levels of BBPRL were higher than PRL levels on day 2, although not nearly so high protein may have been fully saturated with endogenous as those in peripheral plasma (Fig. 2). PRL. The apparent absence of a PRL-binding protein contrasts with the presence, in all sera, of essentially normal levels of circulating GH-binding protein; between PRL-binding capacity of serum 8-17% of the [125I]hGH added was specifically bound by PRL-binding capacity was determined in serum from 100-jiL aliquots of serum (normal range, 12-20%). PRL patients 1, 3, 4, and 5 (Table 8). A pool of serum collected does not cross-react with the binding protein for hGH from more than 200 normal blood donors (courtesy of (10), and therefore, the BBPRL present in the patients' the Red Cross Blood Bank, Melbourne, Australia) was sera is unlikely to be due to binding to either a circulating used as a normal control sample. Specific binding less PRL-binding protein or the GH-binding protein. than 2% of the [125I]hPRL added was detected in the serum assayed. This was true for serum across a broad Discussion range of immunoreactive PRL levels, arguing against, but not dismissing, the possibility that a PRL-binding It has been recognized for well over 10 yr that PRL may exist as variants of different mol wt in plasma and TABLE 6. Bromocriptine suppression tests in three women with exces- the pituitary (11-13). There was also early evidence for sive circulating levels of BBPRL pituitary secretion of large wt forms (14), sometimes Subject no. from pituitary tumor tissue (7,15). It was also recognized Mean ± SD that the larger mol wt forms (BPRL and BBPRL) had 1 2 3 reduced biological activity and reduced binding to PRL BBPRL (Mg/L) receptors (6, 14). Baseline 115.1 78.3 40.9 78.1 ± 37.1 Nadir 102.2 56.4 31.2 63.3 ± 36 A small number of women were discovered, by chance, to have substantial hyperprolactinemia in the presence PRL (Mg/L) Baseline 18.1 17.0 18.1 17.7 ± 0.6 of continued normal ovulation and fertility (1-5), an Nadir 2.8 1.3 5.7 3.3 ± 2.2 occurrence that was most unexpected because of the %PRL usual known effects of hyperprolactinemia in causing Baseline 13.6 17.9 30.5 20.7 ± 8.8 anovulation, amenorrhea, and infertility. On further in- Nadir 2.7 2.2 15.5 6.8 ± 7.5 vestigation most of these women have been found to % Decrease in BBPRL 11.2 28.0 23.7 21.0 ± 8.7 possess an unusual form of hyperprolactinemia where the majority of circulating PRL is in the big big form % Decrease in PRL 84.5 92.4 68.5 81.8 ± 34.4 and, hence, is biologically relatively inactive. It is not 590 FRASER ET AL. JCE & M • 1989 Vol69«No3

Menstrual blood sample(Day 2) Menstrual blood sample(Day 1) dilution 1 in 10 with buffer dilution 1 In 5 with buffer

big-big CPM CPM Monomeric prolactln 10000 -l prolactln 4000 -I 61.6t 92.S%

8000 -

FIG. 2. Gel chromatography patterns from menstrual plasma samples collected from subject 3 on day 1 (leftpanel) and day 2 (rightpanel) of menses, show- Downloaded from https://academic.oup.com/jcem/article/69/3/585/2653406 by guest on 29 September 2021 6000 3000 - ing different proportions of PRL monomer and polymer (note differences in scale). Sample in left panel was diluted 1:10, and sample in right panel 1:5.

1 2000 2000 -J I 10 IS 20 25 30 35 10 IS 20 25 30 35 Fraction number Fraction number

TABLE 8. Specific binding of PRL by serum of four women with excessive circulating BBPRL levels

Day of Fractions % Specific binding of PRL Subject no. menstrual BBPRL BPRL PRL 125 125 cycle [ I]hPRL [ I]hGH

1 15 112.0 83.4 16.6 1.7, 2.7 16.6 23 91.6 88.8 11.2 2.0 3 12 125.9 84.6 15.4 2.1 13.9 26 106.7 88.4 11.6 1.9 4 14 140.5 95.8 6.5 2.95, 1.3 8.1 24 weeks 1168.2 85.2 14.8 (pregnancy) 3 78.1 57.0 43 0.51, 1.1 15.5 13 69.2 77.2 22.9 0.51 Normal pool serum <2 12.20 known what leads to these excessive levels of an abnor- the cycle. Plasma concentrations and relative propor- mal PRL form, but the great majority of these women tions of PRL did not change during the cycle. The reason do not have a detectable pituitary tumor or other under- for this preovulatory rise in BBPRL, which was seen in lying pathological process and do not seem to suffer any all five subjects, is not known. Three of these five women untoward sequelae. and one other were also studied in pregnancy, where total Little is known about factors that may control plasma PRL levels rose, as expected, with increasing gestation; levels of polymeric forms of PRL, although anecdotal after the first trimester a proportionately greater rise evidence exists that levels can be suppressed by bromo- occurred in PRL (6- to 30-fold increase). Nevertheless, criptine (16). Although BBPRL may be released by TRH a 3- to 8-fold rise in absolute levels of BBPRL also and metoclopramide, PRL appears to be released much occurred by the third trimester. During suckling in well more efficiently (6, 14,16). established lactation in one subject PRL levels doubled In the present study we have followed five women with (from a high baseline), while BBPRL remained un- serial plasma samples through one menstrual cycle and changed at about the same high level as at the end of found significantly higher plasma levels of BBPRL in pregnancy. the preovulatory phase compared with other stages of During stimulation with TRH or suppression with BIG BIG PRL 591 bromocriptine PRL was much more rapidly and pro- this is a scientific curiosity in which normal reproductive foundly responsive than BBPRL. Nevertheless, BBPRL and mammary function is maintained without obvious did rise in all three subjects by a mean of 17.5% with pathological disturbance of the function of any organ. TRH and was suppressed by bromocriptine in all three Nevertheless, this is a poorly understood condition, and subjects by a mean of 21.0%. we know nothing of its natural history. These patients In all women investigated in this study plasma PRL should probably continue to have an annual follow-up to was within normal limits and responded as would be keep plasma PRL levels under review, with a further expected for normal women to a range of physiological high resolution computed tomographic scan if levels rise and pharmacological stimuli. Plasma BBPRL generally significantly. Although this condition appears to be rare, moved in the same direction, but much more sluggishly the subjects have no recognizable symptoms and have been detected by chance measurement of total PRL in and to a much lesser degree. The only situation where Downloaded from https://academic.oup.com/jcem/article/69/3/585/2653406 by guest on 29 September 2021 BBPRL appeared to be different was a significant preo- plasma. They have mainly been detected in centers that vulatory rise during the menstrual cycle which was not carry out large numbers of PRL assays in infertility paralleled by a rise in monomer. However, other re- patients and many patients with minor menstrual dis- searchers have reported midcycle PRL elevations in some turbances. Therefore, it seems probable that it is more normal women, which are thought to be stimulated by frequent than currently realized in the normal population the endogenous preovulatory estradiol rise. and, hence, merits detailed study. The existence of this The finding of markedly elevated PRL levels in day 1 condition needs to be more widely recognized and inves- menstrual plasma is of more than passing interest and tigated, in spite of the laborious nature of the chroma- fits with the known ability of decidualized endometrium tography-based assay, because of the clinical implica- to secrete PRL (17). Even in these women the endome- tions of hyperprolactinemia due to the monomer alone trium did not seem to secrete PRL polymers, although it and the investigations and treatment that may be entered is possible that enzymes in menstrual endometrium could into inadvertently. have broken polymers down into monomer. The finding that BBPRL levels in day 2 menstrual plasma were References higher than PRL levels makes this explanation unlikely. 1. Whittaker PG, Wilcox T, Lind T. Maintained fertility in a patient There has been concern that enzymes in menstrual with hyperprolactinemia due to 'big-big' prolactin. J Clin Endocri- nol Metab. 1981;53:863-6. plasma might interfere with the assay, but this has been 2. Soong YK, Fergusson KM, McGarrick G, Jeffcoate SL. Size het- shown not to be true. It seems probable that BBPRL in erogeneity of immunoreactive prolactin in hyperprolactinaemia menstrual blood comes directly from peripheral blood serum. Clin Endocrinol (Oxf). 1982;16:259-65. 3. Anderson AN, Pedersen H, Djursing H, Anderson BN, Friesen and is merely diluted by an endometrial transudate (18). HG. Bioactivity of prolactin in a woman with an excess of large This study has clarified some of the responses of molecular size prolactin, persistent hyperprolactinemia and spon- BBPRL in physiological and pharmacological situations taneous conception. Fertil Steril. 1982;38:625-8. 4. Andino NA, Bidot C, Valdes M, Machade AJ. Chromatographic in women with hyperprolactinemia due mainly to exces- pattern of circulatory prolactin in ovulatory hyperprolactinemia. sive levels of the polymeric form. The sluggish responses Fertil Steril. 1985;44:600-5. of BBPRL suggest that its plasma levels are not under 5. Larrea F, Villaneuva C, Cravioto MC, Escorza A, del Realo. Further evidence that big-big prolactin is preferentially secreted in women such sensitive control as PRL, either through slow secre- with hyperprolactinemia. Fertil Steril. 1985;44:25-30. tion or slow removal from the circulation. It is not even 6. Farkouh NH, Packer MG, Frantz HG. Large molecular size pro- certain that BBPRL (or, when it is present, BPRL) is lactin with reduced receptor activity in human serum: high proportion in basal state and reduction after thyrotropin-releasing secreted in significant amount in this form. It is likely hormone. J Clin Endocrinol Metab. 1979;48:1026-32. that BBPRL is cleared more slowly, as has been dem- 7. Fang VS, Refetoff S. Heterogenous human prolactin from a giant onstrated for the big big forms of GH (19). The possibility pituitary tumour in a patient with panhypopituitarism. J Clin Endocrinol Metab. 1978;47:780-7. that BBPRL represents specific complexes of PRL with 8. Salacinski PRP, McLean C, Sykes JEC, Clement-Jones VV, Lowry serum binding proteins was considered a strong possibil- PJ. Iodination of proteins, glycoproteins, and peptides using a ity given the now unequivocal demonstration that the solid-phase oxidizing agent l,3,4,6-tetrachloro-3a-6a-diphenyl gly- coluril (Iodogen). Anal Biochem. 1981;117:136-46. heterogeneous circulating forms of GH can be explained 9. Ymer SI, Herington AC. Binding and structural characteristics of by the presence in serum of a high affinity, receptor-like, a soluble lactogen-binding protein from rabbit mammary-gland GH-specific binding protein (10, 20). However, no evi- cytosol. Biochem J. 1986;237:813-20. 10. Herington AC, Ymer S, Stevenson J. Identification and character- dence for a similar PRL-binding protein in serum was isation of specific binding proteins for growth hormone in normal obtained. Thus, the origin and structure of BBPRL human sera. J Clin Invest. 1986;77:1817-23. remain unclear. 11. Suh HK, Frantz AG. Size heterogeneity of human prolactin in plasma and pituitary extracts. J Clin Endocrinol Metab. What are the biological implications of this type of 1974;39:929-35. hyperprolactinemia, and what does one tell a patient who 12. Guyda HJ. Heterogeneity of human growth hormone and prolactin has the condition? All available evidence suggests that secreted in vitro: immunoassay and radioreceptor assay correla- 592 FRASER ET AL. Vol69-No3 tions. J Clin Endocrinol Metab. 1975;41:953-67. the Endocrine Society of Australia; 1986. 13. Kiefer KA, Malarkey WB. Size heterogeneity of human prolactin 17. Maslar IA, Riddick DH. Prolactin production by human endome- in CSF and serum; experimental conditions that alter gel filtration trium during the normal menstrual cycle. Am J Obstet Gynecol. patterns. J Clin Endocrinol Metab. 1978;46:119-24. 1979;135:751-5. 14. Gamier PE, Aubert ML, Kaplan SL, Grumbach MM. Heteroge- 18. Fraser IS, McCarron G, Markham R, Resta T. Blood and total fluid content of menstrual discharge. Obstet Gynecol. 1985;65:194- neity of pituitary and plasma prolactin in man: decreased affinity 98. of "big" prolactin in a radioreceptor assay and evidence for its 19. Baumann G, Amburn Kd, Buchanan TA. The effect of circulating secretion. J Clin Endocrinol Metab. 1978;47:1273-81. growth hormone-binding protein on metabolic clearance, distribu- 15. Rogol AD, Rosen SW. Prolactin of apparent large molecular size: tion and degradation of human growth hormone. J Clin Endocrinol the major immunoreactive component in plasma of a patient with Metab. 1987;64:657-60. a pituitary tumor. J Clin Endocrinol Metab. 1974;38:714-7. 20. Baumann G, Amburn K, Shaw MA. The circulating growth hor- 16. Sweefong NG, Richards RJ, Ward GJ, Junze HE, Arving MJ, mone (GH)-binding protein complex: a major constituent of plasma Cameron DP. Big-big prolactin; what is it? [Abstract 136] Proc of GA in man. Endocrinology. 1988;122:976-84. Downloaded from https://academic.oup.com/jcem/article/69/3/585/2653406 by guest on 29 September 2021

Problem Solving in Gynecologic Endocrinology and Infertility December 8-9, 1989 The Knickerbocker Hotel Chicago, Illinois The program will focus on the latest developments impacting on the diagnosis and management of infertility, repetitive abortion, and endocrinopathies including: male infertility, sexually-transmitted dis- eases, vaginal ultrasound, assisted reproductive technologies, laser endoscopy, endocrinology of pregnancy, and new treatment for androgen excess. New advances in estrogen therapy and breast cancer markers will also be addressed. These are exciting times in our specialty with new treatments and techniques. We will discuss these changes, pointing out their potential for improved patient care, and the pitfalls and limitations to their use. Dr. Howard Judd, a world renowned expert in this field, will be our guest lecturer. For further information, please contact: The University of Chicago Center for Continuing Medical Education 5841 South Maryland Avenue-Box 139 Chicago, Illinois 60637 Telephone: (312) 702-1056