0021-972X/90/7101-0179$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright© 1990 by The Endocrine Society
Vol. 71, No. 1 Printed in U.S.A.
Evidence for the Presence of Human Chorionic Gonadotropin (hCG) and Free /?-Subunit of hCG in the Human Pituitary R. HOERMANN, G. SPOETTL, R. MONCAYO, AND K. MANN Medical Department II, Klinikum Grosshadern, University of Munich, D-8000 Munich 70 and Medical Department I, University of Ulm (R.M), D-7900 Ulm, West Germany
ABSTRACT. Previous studies have indicated that the pituitary gland may produce free a-subunit and small quantities of hCG in addition to other glycoprotein hormones. Since synthesis of holo-hCG requires the presence of both subunits, we have investigated the occurrence in human pituitary of free /3-subunit of hCG, in addition to intact holo-hCG. We processed a pituitary extract by fractionated ammonium sulfate precipitation followed by sequential chromatography on Sephadex G-100 and Ultrogel AcA 44. The fractions obtained were assessed for their reactivities with a panel of polyclonal and monoclonal antibodies specific for holo-hCG, /3-subunit of hCG, a-subunit, or hCG/LH. In addition to the expected LH and a-subunit, we detected materials which eluted from the column in positions very similar to those of cochromatographed 125I-hCG tracer and hCG-|8 (NIH CR123-/8), and which showed immunoreactivity in specific immunoradiometric assays for holo-hCG and hCG-/3, respectively. Holo-hCG and hCG-|8 material derived from the urine of a postmenopausal woman showed behaviors on the column similar to the pituitary forms. Both the pituitary holo-hCG- and free hCG-0-subunit activity could be enriched (~500 times) by affinity chromatography on an hCG antibody-coupled Sepharose column. When subjected to isoelectric focusing in granulated gel holo-hCG and hCG-/3-subunit of pituitary origin were focused in the pi-range of pregnancy hCG and pregnancy hCG-jS-subunit, respectively. Like pregnancy hCG, most (75%) of the pituitary hCG was bound to a column of Con A-Sepharose; however, the Con A-nonbinding hCG fraction (~25%) was much
higher than that found in pregnancy hCG. On the basis of immunoreactivites, the content of holo-hCG in our pituitary extract was estimated to be 60 Mg/g. arid that of free /3-subunit 45 Mg/g; for comparison, LH was approximately 20 mg/g, and free a-subunit 1.6 mg/g. In addition, we could demonstrate the presence of both holo-hCG- and free hCG-jS-subunit-like immunoreactivity in NaCl-extracts from single pituitaries of two postmenopausal women. In these studies a second hCG-/3-immunoreactive material eluting far behind the hCG-/3-position was found. Chromatography of purified LH-/3-subunit, which crossreacts 1.56% in the hCG-/3 IRMA, yielded an elution pattern clearly distinguishable from that of the hCG-jS-immunoreactive substances. Spiking of the pituitary extract with hCG-/3 and a /3-fragment of hCG similar to the so-called /3-core fragment resulted in a rise of the first and second hCG-/?-activity peak, respectively.Taken together, the immunological and the various chromatographic studies provide strong evidence for the existence of both free hCG-/?-subunit and holo-hCG in the human pituitary gland. The properties of hCG- and hCG-/3-like materials found in the human pituitary compared favorably with those of the respective purified standard preparations of pregnancy sources and of urinary material from a postmenopausal woman. Apparently, the human pituitary is able to produce both an a-subunit, which has been well recognized, and an hCG-/3subunit as described in this report, and to assemble the two subunits to produce holo-hCG. {J Clin Endocrinol Metab 7 1 : 179-186, 1990)
I
MMUNOREACTIVE hCG-like material, apart from its abundance in human placenta, has been found in relatively small concentrations in a number of other tissues including testis, liver, colon, and pituitary (1-5). Recent advances in assay techniques for the measurement of hCG, particularly the development of sensitive immunoradiometric assays (IRMAs) have allowed detection of very low concentrations of hCG-like material in the serum of healthy normal women and men (6-8). Serum hCG, like LH, is elevated in postmenopausal women, and its serum concentration rises upon stimu-
lation by GnRH in both postmenopausal women and normal men (7, 8). These observations have pointed to the pituitary gland as the source of hCG present in serum of non pregnant individuals. However, direct evidence for secretion of the intact hCG molecule by the human pituitary is lacking and no information is presently available on pituitary production of the free hCG-/?-subunit. Only the synthesis and secretion of free a-subunit, which is common to all the pituitary glycoprotein hormones, by both the normal human pituitary and pituitary adenomas is well known (9-11). In the present studies we have therefore undertaken to further elucidate the immunological and biochemical properties of hCG-like material derived from human pituitary extract and to look for evidence of the presence
Received March 3,1989. Address requests for reprints to: Dr. Rudolf Hoermann, Medical Department II, Klinikum Grosshadern, University of Munich, Marchionistrasse 15, D-8000 Munich 70, West Germany 179
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
HOERMANN ET AL.
180
of hCG-/?-subunit in the pituitary. A portion of these findings has been presented in abstract form (12). Materials and Methods Highly purified preparations of intact hCG (CR121), FSH (HSl) a-subunit of hCG (hCG-a, CR121a), and /?-subunit of hCG (hCG-j8, CR123/3) were supplied by the Hormone Distribution Program of the NIDDK. TSH, /3-subunit of TSH (TSH-/3), /?-subunit of FSH (FSHfi), and j8-subunit of LH (LH-/?) were obtained from Calbiochem (San Diego, CA); hLH was obtained from Serono (Freiburg, FRG). The fragment of the hCG-/?-subunit used in some chromatography studies had been isolated from crude hCG and resembles in its properties /3-core fragment described by others (13), as we have reported previously (14). Monoclonal antibodies 11/6 (against hCG and LH), 3/6 (hCG and hCG-/3-subunit), 12/17 (hCG and hCG-/?-subunit), and 2/6 (specific for the /?subunit of hCG) were raised and purified by Dr. K. Siddle (University of Cambridge, Cambridge, UK, as previously described (15, 16). Polyclonal anti-hCG/LH antiserum, which recognizes both hCG and LH, was obtained by immunization of a rabbit with hCG (17). The titer of this antiserum was 1:40,000, yielding a bound to total ratio (B/T) of 20%. The polyclonal antibody against a-subunit was produced in a rabbit by immunization with a-subunit. It was used in a titer of 1:1,500 (B/T 21%) (17). Immunoassays for hCG IRMA for hCG (holo-hCG). For the determination of holo-hCG we have developed a specific IRMA, which has recently been described in detail (16). This assay, which employs two monoclonal antibodies (3/6 as solid phase antibody and 11/6 as tracer) measures holo-hCG with negligible cross-reactivity with free hCG-/3-subunit ( as compared to approximately 20 mg/g of LH and 1.6 mg/g of a-subunit. Gel chromatography of NaCl extract of a single pituitary gland on a column of AcA 44 A 200-JUI aliquot of the NaCl extract obtained from a single pituitary gland of a postmenopausal woman (see Materials and Methods) was chromatographed together with 125I-hCG (20,000 cpm) on a column of Ultrogel AcA 44. The fractions (0.9 ml each) eluted with 0.01 M TRISHC1, 0.15 M NaCl, pH 7.3, were assessed by various immunoassays. Holo-hCG-like activity measured by 3/ 6-11/6-IRMA eluted in Fr. 57 to Fr. 72 with a maximum in Fr. 63, which corresponded well with the position of 125 I-hCG. For comparison, LH-activity (hCG/LH-RIA) ranged from Fr. 70 to Fr. 81 with a maximum rise in Fr. 73 (Fig. 3). Testing the same fractions in the hCG-/3 assay (2/6-12/17-IRMA) revealed two distinct peaks of the activity, one in Fr. 62 to Fr. 75 (maximum in Fr. 68) and the other in Fr. 87 to Fr. 99 (maximum in Fr. 92; Fig. 4B). Since some degree of cross-reactivity of LH-/3 had to be considered in this assay the elution pattern of purified LH-/3 (2.5 ng) on the same column was studied by measuring the eluted fractions in the hCG-/3-assay: the activity was found in Fr. 75 to 90 with a maximum in Fr. 83 and did not correspond with either of the two hCG-/3-immunoreactivity peaks of pituitary origin (Fig. 4, C and B). In another experiment, we added a defined
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
HOERMANN ET AL.
182
FlG. 1. Chromatography of urinary extract (50 mg) obtained from a postmenopausal woman on a column of Ultrogel Ac A 44. The column was eluted with 0.1 M TRIS-C1, 0.15 M NaCl, pH 8.0, and the fractions (2 ml) obtained were assessed for their content of holo-hCG (hCG), free hCG-0-subunit (hCG-/?), asubunit, and hCG/LH by the use of IRMAs and RIAs of different specificities described in Materials and Methods. The elution positions from the column of 125IhCG tracer and 12SI-a-subunit are indicated by arrows.
hCG-6 (pg/l) 0.4 T
hCG (pg/1) 41
12S
JCE & M • 1990 Vol71«Nol 12S
l-hCG
hCG-a (pg/t)
!-hCGi
I
0.3 .
hCG/LH-RIA (pg/1)
r 40
[40
30
" 30
20
• 20
10
•
-hCG/LH
0.2 hCG-or
0.1
10
100 Fraction (No.)
hCG-a (\igfl) hCG/LH-RIA (mg/l)
hCG (pg/i) hCG-6 (pg/l)
r 24
12-
10-
18 8-
12
6-
4-
40 holo-hCGactivity (\ig/l) 20 i
2-
60
B 10 -
50
60
70
80
90 Fraction (No.)
FlG. 2. Chromatography of pituitary extract (21 mg) on a column of ultrogel AcA 44 preequilibrated and eluted with 0.1 M TRIS-C1, 0.15 M NaCl, pH 8.0. The fractions (2 ml) were measured in various RIAs and IRMAs with different specificities for hCG and its subunits: hCG/LHRIA, specific holo-hCG-IRMA (hCG), specific hCG-/3-subunit-IRMA (hCG-0), and a-subunit-RIA.
amount of hCG-/3 (100 ng) to the pituitary extract before the chromatography of the material on AcA 44. This led to a rise of hCG-/?-immunoreactivity eluting in a position virtually identical to the first peak described above, while the second peak was apparently unaffected (Fig. 4D). Chromatography of the pituitary extract supplemented by jft-fragment of hCG (100 ng) resulted in an elevation of the second hCG-/?-immunoreactive peak without a change of the first peak (Fig. 4E).
40 SO LHactivity (\igfl)
60
70
80
90
100
110
120
60
70
80
90
100
110
120
20.000 1
10,000 -
40
50
Fraction No.
Affinity purification of pituitary hCG on hCG-antibodycoupled Sepharose
FIG. 3. Chromatography of a NaCl extract obtained from a single pituitary gland of a postmenopausal woman on a column of ultrogel AcA 44. The fractions (0.9 ml each) eluted with 0.1 M TRIS-C1, 0.15 M NaCl, pH 8.0, were assessed for holo-hCG- (B) and hCG/LH-immunoreactivities (C), as shown here. 125I-hCG was cochromatographed (A).
The pituitary extract (270 mg in 5 ml 0.15 M Na2HPO4, 0.2 M NaCl, pH 9.0) was affinity-purified on a column of cyanbromide-activated Sepharose coupled with the monoclonal antibody 3/6 that reacts with both holo-hCG and hCG-jff-subunit (Fig. 5). As expected, both holo-hCG-
and hCG-jS-immunoreactivity were adsorbed to the column and could subsequently be eluted from the column by 2 M KSCN, pH 5.0. In contrast, a-subunit- and LHactivity were apparently not adsorbed to the column and
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
HOLO-hCG AND hCG-/3 IN HUMAN PITUITARY
183
hCG (yg/l) hCQ-0 (yg/1)
hCG/LH-RIA (mg/1) hCG-a (mg/1)
200 -I
150 -
120
400
r 40
• 300
30
200
• 20
100
- 10
2 M KSCN, pH 5.0
hCG
100 -
50 -
0
10
20
30 Fraction (No.)
FIG. 5. Immunopurification of hCG from pituitary extract. Pituitary extract (270 mg) was applied to a column of Sepharose coupled with monoclonal hCG/hCG-/3-antibody 11/6 (see Materials and Methods). The column was preequilibrated with 0.15 M Na2HPO4, 0.2 M NaCl, pH 9.0, and eluted up to fraction 15 with the same buffer. The elution proceeded with adding 2 M KSCN, pH 5.0. Determination of the various immunoreactivities, i.e. holo-hCG (hCG), hCG-^-subunit (hCG-/3), asubunit, and hCG/LH in the fractions revealed a selective adsorption of holo-hCG- and hCG-/S subunit-activity by the column and the recovery of the adsorbed hCG-material from the column, as shown here. hCG-6 (pg/l)
r3
40 Fraction No.
FIG. 4. Chromatography of a NaCl extract obtained from a single pituitary gland of a postmenopausal woman on a column of ultrogel AcA 44. The fractions (0.9 ml each) eluted with 0.1 M TRIS-C1, 0.15 M NaCl, pH 8.0, were assessed for hCG-jS-immunoreactivity in a specific IRMA for the /3-subunit of hCG, as shown here. 125I-hCG was cochromatographed and the elution patterns were very similar in the various experiments (A). The pattern of the pituitary extract alone is shown in B. For comparison in C, the behavior of a large quantity of LH-/3 (2.5 Mg) was measured by the hCG-/3-IRMA. In other experiments, the pituitary extract was supplemented before chromatography on the same column with hCG-/3 (NIH CR123/3 100 ng, D) or a fragment of the 0subunit of hCG (100 ng, E).
eluted with the void volume (Fig. 5). This procedure resulted in an approximately 500-fold enrichment of hCG/hCG/3-like activity per gram protein.
40 Fraction FIG. 6. Con-A-Sepharose column chromatography of 20 mg pituitary extract. The column was preequilibrated and eluted with 10 mM TRISCl, pH 7.5, 0.15 M NaCl, 1 mM CaCl2, 1 mM MnCl2,1 mM MgCl2 up to fraction 10; elution then continued with the addition of 1 M a-methylglycoside. Fractions of 2 ml each were collected and tested for holohCG (hCG)- and hCG-/?-subunit (hCG-/3)-immunoreactivities.
Chromatography of pituitary extract on Con A-Sepharose When pituitary extract (20 mg) was chromatographed on a column of Con A-Sepharose as described in Materials and Methods, the fractions (2 ml) obtained before and after treating the column with 1 M a-methyl-glucoside were assayed for holo-hCG and free hCG-/3-subunit (Fig. 6). Approximately 75% of the total holo-hCG-immunoreactivity bound to the Con A-column and 25% was unbound. The ratio of the bound to unbound hCG-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
HOERMANN ET AL.
184
0-subunit immunoreactivity by the lectin was approximately 62%-38%. When a portion of the nonbinding fractions was rechromatographed in another experiment, no significant binding to the column was found. For LHactivity measured by RIA, the bound/unbound ratio was 84.6%/15.4%. Isolectric focusing of pituitary extract in granulated gel
Pituitary extract (220 mg) was subjected to preparative isoelectric focusing in granulated gel (pH 3-6). Thirty sections (0.5 cm each) were eluted with distilled water and measured for their immunoreactivities in specific assays for holo-hCG and hCG-/?-subunit. The peak of holo-hCG activity was found around pi 4.5, while the peak of hCG-/8 activity appeared slightly ahead of the holo-hCG peak at pi 3.5-5. In addition, there was a rise in hCG-/? activity close to the cathode. For comparison, in the same experiment LH-immunoreactivity focused also in a pi-range greater than 6 (Fig. 7). In other experiments, purified ^-fragment of hCG was subjected to electrofocusing and showed a pi about 9 (data not shown).
Discussion Although many studies have suggested secretion of hCG by the human pituitary (1-8), direct evidence is still hCG (pg/1) hCG-6 (M9/1) LH ^ 40-]
30"
-2
20 Fraction No.
30
FIG. 7. Isoelectric focusing of pituitary extract in granulated gel (12.5 X 26 cm). A linear pH gradient from pH 3-6 was applied. After focusing, the gel was sectioned into 30 x 0.5-cm slices eluted in distilled water. The fractions were measured in a specific holo-hCG-IRMA (hCG) and an IRMA specific for the free hCG-/3-subunit (hCG-/3). For comparison, LH-activity determined by RIA is also shown.
JCE & M • 1990 Vol 71 • No 1
lacking. Studies using serum and urine are ambiguous because of the possibility of hCG secretion by tissues other than the pituitary. Studies with pituitary extracts are subjected to doubt due to a lack of specificity of the hCG antibodies used. The expression of the a-subunit gene in the human pituitary has been well recognized (911); however, there are several genes for the hCG-j8subunit, and their regulation in the placenta is poorly understood; regulation of hCG-/3 genes in nontrophoblastic tissues is not understood at all (20). Obviously, the synthesis of intact hCG requires the presence of both an a- and a /3-subunit of hCG. In the present studies we have therefore focused our attention on demonstrating the presence of free hCG-/?-subunit in the human pituitary as well as further characterizing the immunological and physical properties of pituitary holo-hCG. To this
end, hCG- and hCG-/3-like materials in pituitary extracts were purified and characterized by means of gel filtration on Ultrogel AcA 44, affinity chromatography on Con ASepharose, affinity chromatography on hCG-antibodycoupled Sepharose, and preparative electrofocusing in granulated gel. The immunoreactivities of the fractions obtained were determined in a panel of assays for holohCG, hCG/LH, hCG/hCG-^-subunit, free hCG-,3-subunit, and a-subunit. With respect to the presence of holo-hCG, these studies provide several lines of evidence that the human pituitary contains intact holo-hCG rather than an hCGrelated peptide: first, the behavior of pituitary hCG during gel chromatography on AcA 44 is indistinguishable from that of cochromatographed 125I-hCG; second, the material exhibits immunoreactivity in a monoclonal IRMA with a high specificity for the holo-hCG molecule and negligible cross-reactivity with free hCG-/?-subunit and LH, since there was only negligible holo-hCG-immunoactivity in the fractions containing LH-peak-activity (Fig. 3); third, use of an hCG-antibody-coupled Sepharose column resulted in differential adsorption and selective enrichment of the material compared with LH; fourth, electrofocusing of the material occurred in a plrange very similar to that we and others have previously reported for pregnancy hCG (19, 21-23); and, fifth, the ability of most of the material (75%) to bind Con A Sepharose, similar to the Con A-binding properties of pregnancy hCG. In addition, there were other variant forms of pituitary holo-hCG that are apparently not bound by Con A-Sepharose. These appear to be considerably more prevalent in pituitary hCG than in pregnancy hCG, which contains only a minute amount of material not bound by Con A (17, 24). Thus, pituitary hCG exhibits microheterogeneity similar to that of hCG of tumorous origin from trophoblastic diseases or testicular cancer, which contains high proportions of Con Anonbinding forms (17). The content of hCG per gram
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
HOLO-hCG AND h.CG-,3 IN HUMAN PITUITARY pituitary extract was roughly estimated to be 61 fig, which was about 0.3% the content of LH. Our evidence for the presence of the free hCG-j8subunit in the human pituitary is the demonstration of material showing an immunoreactivity with a specific monoclonal antibody against hCG-|3-subunit. Interestingly, in the experiments with NaCl-extracts of single glands we found yet another peak of hCG-jS-immunoreactive material, eluting in a much lower molecular size range than that expected for hCG-0. The hCG-/3-IRMA used was sufficiently specific to distinguish pituitary hCG-jQ-immunoreactivity from that of the holo-hCG and other pituitary hormones and subunits (LH, FSH, TSH, FSH-/3, TSH-/?), except for LH-jS-subunit, which showed a cross-reactivity of 1.56%. Therefore, we applied LH-jSsubunit in a relatively high dosage to the same column, and measured the eluted fractions by hCG-0 assay. The single peak of immunoreactivity obtained with LH-/3 differed sharply in its position and pattern from pituitary hCG-/3-activity, its elution being clearly retarded compared to the first hCG-/3-activity peak. In another experiment, we spiked the pituitary extract with hCG-/3 (NIH CR123/3). This led to a rise of the hCG-,3-immunoreactivity that coincided with the first pituitary-hCG-/3-peak. Next, together with holo-hCG-, pituitary hCG-0-activity was differentially removed from LH/LH-/3-immunoreactivity and immunopurified by a Sepharose-coupled antibody that recognizes hCG-/?-subunit in both the holohCG and the free form. When subjected to isoelectric focusing, the pituitary hCG-/3-activity exhibited similar characteristics to those previously described for the hCGiS-subunit of pregnancy urine (21-23). Further, like holohCG of pituitary origin, pituitary free hCG-/? differed somewhat from its pregnancy counterpart in that it contained a relatively higher fraction of Con A-nonbinding variant forms (25%). The majority of the material, however, was bound by Con A. Taken together, the immunological and the various chromatographic studies provide convincing evidence for the existence of free hCG-j8-subunit, in addition to holohCG, in the human pituitary gland. The properties of hCG- and hCG-/5-like materials found in the human pituitary compared favorably with those of the respective purified standard preparations and of urinary material from a postmenopausal woman. Less clear at this point, is the nature of the second form of hCG-/5-like immunoreactivity whose molecular size on AcA 44-chromatography is much smaller than hCG-/?. Interestingly, the elution position of this material was the same as that of beta-fragment of hCG added to the pituitary NaCl-extract. A rise of hCG-/3 activity in the alkaline pi-range during electrofocusing of pituitary extract is also consistent with the assumption of the presence of a /?fragment. Further, we demonstrated a considerable
185
cross-reactivity of ^-fragment in the hCG-(8-IRMA. The preparation and properties of our /3-fragment have been described previously (14) and are similar to those reported by others for so-called j8-core fragment (13). One may speculate from these data on the presence of /3-core in the human pituitary, which would fit well with another recent report on the occurrence of /3-core fragment in serum (25). Further investigations, including measurement of pituitary extracts by specific immunoassays for 0-core, are warranted. Because of the possibility that the hCG we detected in the pooled pituitary extract was due to contamination by a specific specimen that contributed large amounts of hCG to the pool, NaCl-extracts obtained from single pituitary glands of two postmenopausal women (data of the second case similar, but not shown) were also investigated. These experiments indeed confirmed the presence of holo-hCG- and hCG-/?-subunit-like material in individual glands. Acknowledgment The authors are grateful to Dr. Robert E. Wehmann, National Institutes of Health, Bethesda, for his assistance in improving the style of the manuscript.
References 1. Braunstein GD, Kamdar V, Rasor J, Swaminathan N, Wade ME. Widespread distribution of a chorionic gonadotropin-like substance in normal human tissues. J Clin Endocrinol Metab. 1979;49:917-25.
2. Chen HC, Hodgen GD, Matsuura S, Lin LJ, Gross E, Reichert Jr LE, Birken S, Canfield RE, Ross GT. Evidence for a gonadotropin from nonpregnant subjects that has physical, immunological, and biological similarities to human chorionic gonadotropin. Proc Natl Acad Sci USA. 1976;73:2885-9. 3. Robertson DM, Suginami H, Hernandez Montes H, Puri CP, Choi SK, Diczfalusy E. Studies on a human chorionic gonadotropin-like material present in nonpregnant subjects. Acta Endocrinol (Copenh). 1978;89:492-505. 4. Yoshimoto Y., Wolfsen AR, Hirose F, Odell WD. Human chorionic gonadotropin-like material: Presence in normal human tissue. Am J Obstet Gynecol. 1979;134:729-33. 5. Stockell Hartree A, Shownkeen RC, Stevens VC, Matsuura SM, Ohashi M, Chen HC. Studies of the human chorionic gonadotropinlike substance of human pituitary glands in its significance. J Endocrinol 1983;96:115-26. 6. Odell WD, Griffin J. Pulsatile secretion of human chorionic gonadotropin in normal adults. N Eng J Med. 1987;317:1688-91. 7. Borkowski A, Puttaert V, Gyling M, Muquardt C, Body JJ. Human chorionic gonadotropin-like substance in the plasma of normal nonpregnant subjects and women with breast cancer. J Clin Endocrinol Metab. 1984;58:1171-8. 8. Stenman UH, Altthan H, Ranta T, Vartiainen E, Jalkanen J, Sepala J. Serum levels of human chorionic gonadotropin in nonpregnant women and men are modulated by gonadotropin-releasing hormone and sex steroids. J Clin Endocrinol Metab. 1987;64:7306. 9. Kourides IA, Weintraub BD, Ridgway EC, Maloof F. Pituitary secretion of free alpha and beta subunit of human thyrotropin in patients with thyroid disorders. J Clin Endocrinol Metab. 1975;40:872-85. 10. Landolt AM, Heitz PU, Zenklusen HR. Production of the alpha-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
186
11. 12. 13.
14. 15.
16. 17. 18.
HOERMANN ET AL.
subunit of glycoprotein hormones by pituitary adenomas. Path Res Pract. 1988; 183:610-2. Comi RJ, Gesundheit N, Murray P, Gorden P, Weintraub BD. Response of thyrotropin-secreting pituitary adenomas to a longacting somatostatin analogue. N Engl J Med. 1987;317:12-7. Korallus-Prinz K, Mann K, Moncayo R, Siddle K. hCG and free beta-subunit exist in the human pituitary (Abstract). Acta Endocrinol (Copenh). 1988;117(Suppl):287-90. Blithe DL, Akar AH, Wehmann R, Birken S, Nisula B. Purification of /3-core fragment from pregnancy urine and demonstration that its carbohydrate moieties differ from those of native human chorionic gonadotropin-/3. Endocrinology. 1988;122:173-80. Hoermann R, Amir SM, Nomura T, Ingbar SH. Design of a longlived thyrotropin antagonist from derivatives of human chorionic gonadotropin. Endocrinology 1989;124:223-32. Siddle K, Gard T, Thomas D, Cranage MP, Coombs RRA. Red cell-labelled monoclonal antibodies for assay of human chorionic gonadotropin and luteinising hormone by reverse passive haemagglutination. J Immunol Methods. 1984;73:169-76. Mann K, Siddle K. Evidence for free beta-subunit secretion in socalled human chorionic gonadotropin-positive seminoma. Cancer. 1988;62:2378-82. Mann K, Karl HJ. Molecular heterogeneity of human chorionic gonadotropin and its subunits in testicular cancer. Cancer 1983;52:654-60. Schleyer M, Voigt KH. Preparation of highly purified human
19. 20. 21. 22. 23.
24. 25.
JCE & M • 1990 Vol 71 • No 1
somatotropin (growth hormone). Hoppe-Seylers Z. Physiol. Chem. 1977;358:1557-64. Mann K, Schneider N, Hoermann R. Thyrotropic activity of acidic isoelectric variants of human chorionic gonadotropin from trophoblastic tumors. Endocrinology. 1986;118:1558-66. Fiddes JC, Talmadge K. Structure, expression, and evolution of the genes for the human glycoprotein hormones. Recent Prog Horm Res. 1984;40:43-74. van Hell H. Purification and characterization of urinary hCG. In: Moudgal NR, ed. Gonadotropins and Gonadal function. New York: Academic Press Inc; 1974;66-78. Merz WE, Hilgenfeldt U, Doerner M, Brossmer R. Biological, immunological and physical investigations on human chorionic gonadotropin. Hoppe-Seylers Z. Physiol Chem. l974;355:1035-45. Yazaki K, Yazaki C, Wakabayashi K, Igarashi M. Isoelectric heterogeneity of human chorionic gonadotropin: presence of choriocarcinoma specific components. Am J Obstet Gynecol. 1980;138:189-94. Dufau ML, Tsuruhara T, Catt J. Interaction of glycoprotein hormones with agarose concanavalin A. Biochem Biophys Acta. 1972;278:281-92. Alfthan H, Schroeder J, Stenman UH. Demonstration of core jShCG fragment in serum of pregnant women and cancer patients (Abstract). XVIIth Meeting of the International Society for Oncodevelopmental Biology and Medicine, Theoretical and Clinical Cancer Research, Freiburg, West Germany. 1989;58.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
Vol. 71, No. 1 Printed in U.S.A.
Evidence for the Presence of Human Chorionic Gonadotropin (hCG) and Free /?-Subunit of hCG in the Human Pituitary R. HOERMANN, G. SPOETTL, R. MONCAYO, AND K. MANN Medical Department II, Klinikum Grosshadern, University of Munich, D-8000 Munich 70 and Medical Department I, University of Ulm (R.M), D-7900 Ulm, West Germany
ABSTRACT. Previous studies have indicated that the pituitary gland may produce free a-subunit and small quantities of hCG in addition to other glycoprotein hormones. Since synthesis of holo-hCG requires the presence of both subunits, we have investigated the occurrence in human pituitary of free /3-subunit of hCG, in addition to intact holo-hCG. We processed a pituitary extract by fractionated ammonium sulfate precipitation followed by sequential chromatography on Sephadex G-100 and Ultrogel AcA 44. The fractions obtained were assessed for their reactivities with a panel of polyclonal and monoclonal antibodies specific for holo-hCG, /3-subunit of hCG, a-subunit, or hCG/LH. In addition to the expected LH and a-subunit, we detected materials which eluted from the column in positions very similar to those of cochromatographed 125I-hCG tracer and hCG-|8 (NIH CR123-/8), and which showed immunoreactivity in specific immunoradiometric assays for holo-hCG and hCG-/3, respectively. Holo-hCG and hCG-|8 material derived from the urine of a postmenopausal woman showed behaviors on the column similar to the pituitary forms. Both the pituitary holo-hCG- and free hCG-0-subunit activity could be enriched (~500 times) by affinity chromatography on an hCG antibody-coupled Sepharose column. When subjected to isoelectric focusing in granulated gel holo-hCG and hCG-/3-subunit of pituitary origin were focused in the pi-range of pregnancy hCG and pregnancy hCG-jS-subunit, respectively. Like pregnancy hCG, most (75%) of the pituitary hCG was bound to a column of Con A-Sepharose; however, the Con A-nonbinding hCG fraction (~25%) was much
higher than that found in pregnancy hCG. On the basis of immunoreactivites, the content of holo-hCG in our pituitary extract was estimated to be 60 Mg/g. arid that of free /3-subunit 45 Mg/g; for comparison, LH was approximately 20 mg/g, and free a-subunit 1.6 mg/g. In addition, we could demonstrate the presence of both holo-hCG- and free hCG-jS-subunit-like immunoreactivity in NaCl-extracts from single pituitaries of two postmenopausal women. In these studies a second hCG-/3-immunoreactive material eluting far behind the hCG-/3-position was found. Chromatography of purified LH-/3-subunit, which crossreacts 1.56% in the hCG-/3 IRMA, yielded an elution pattern clearly distinguishable from that of the hCG-jS-immunoreactive substances. Spiking of the pituitary extract with hCG-/3 and a /3-fragment of hCG similar to the so-called /3-core fragment resulted in a rise of the first and second hCG-/?-activity peak, respectively.Taken together, the immunological and the various chromatographic studies provide strong evidence for the existence of both free hCG-/?-subunit and holo-hCG in the human pituitary gland. The properties of hCG- and hCG-/3-like materials found in the human pituitary compared favorably with those of the respective purified standard preparations of pregnancy sources and of urinary material from a postmenopausal woman. Apparently, the human pituitary is able to produce both an a-subunit, which has been well recognized, and an hCG-/3subunit as described in this report, and to assemble the two subunits to produce holo-hCG. {J Clin Endocrinol Metab 7 1 : 179-186, 1990)
I
MMUNOREACTIVE hCG-like material, apart from its abundance in human placenta, has been found in relatively small concentrations in a number of other tissues including testis, liver, colon, and pituitary (1-5). Recent advances in assay techniques for the measurement of hCG, particularly the development of sensitive immunoradiometric assays (IRMAs) have allowed detection of very low concentrations of hCG-like material in the serum of healthy normal women and men (6-8). Serum hCG, like LH, is elevated in postmenopausal women, and its serum concentration rises upon stimu-
lation by GnRH in both postmenopausal women and normal men (7, 8). These observations have pointed to the pituitary gland as the source of hCG present in serum of non pregnant individuals. However, direct evidence for secretion of the intact hCG molecule by the human pituitary is lacking and no information is presently available on pituitary production of the free hCG-/?-subunit. Only the synthesis and secretion of free a-subunit, which is common to all the pituitary glycoprotein hormones, by both the normal human pituitary and pituitary adenomas is well known (9-11). In the present studies we have therefore undertaken to further elucidate the immunological and biochemical properties of hCG-like material derived from human pituitary extract and to look for evidence of the presence
Received March 3,1989. Address requests for reprints to: Dr. Rudolf Hoermann, Medical Department II, Klinikum Grosshadern, University of Munich, Marchionistrasse 15, D-8000 Munich 70, West Germany 179
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
HOERMANN ET AL.
180
of hCG-/?-subunit in the pituitary. A portion of these findings has been presented in abstract form (12). Materials and Methods Highly purified preparations of intact hCG (CR121), FSH (HSl) a-subunit of hCG (hCG-a, CR121a), and /?-subunit of hCG (hCG-j8, CR123/3) were supplied by the Hormone Distribution Program of the NIDDK. TSH, /3-subunit of TSH (TSH-/3), /?-subunit of FSH (FSHfi), and j8-subunit of LH (LH-/?) were obtained from Calbiochem (San Diego, CA); hLH was obtained from Serono (Freiburg, FRG). The fragment of the hCG-/?-subunit used in some chromatography studies had been isolated from crude hCG and resembles in its properties /3-core fragment described by others (13), as we have reported previously (14). Monoclonal antibodies 11/6 (against hCG and LH), 3/6 (hCG and hCG-/3-subunit), 12/17 (hCG and hCG-/?-subunit), and 2/6 (specific for the /?subunit of hCG) were raised and purified by Dr. K. Siddle (University of Cambridge, Cambridge, UK, as previously described (15, 16). Polyclonal anti-hCG/LH antiserum, which recognizes both hCG and LH, was obtained by immunization of a rabbit with hCG (17). The titer of this antiserum was 1:40,000, yielding a bound to total ratio (B/T) of 20%. The polyclonal antibody against a-subunit was produced in a rabbit by immunization with a-subunit. It was used in a titer of 1:1,500 (B/T 21%) (17). Immunoassays for hCG IRMA for hCG (holo-hCG). For the determination of holo-hCG we have developed a specific IRMA, which has recently been described in detail (16). This assay, which employs two monoclonal antibodies (3/6 as solid phase antibody and 11/6 as tracer) measures holo-hCG with negligible cross-reactivity with free hCG-/3-subunit ( as compared to approximately 20 mg/g of LH and 1.6 mg/g of a-subunit. Gel chromatography of NaCl extract of a single pituitary gland on a column of AcA 44 A 200-JUI aliquot of the NaCl extract obtained from a single pituitary gland of a postmenopausal woman (see Materials and Methods) was chromatographed together with 125I-hCG (20,000 cpm) on a column of Ultrogel AcA 44. The fractions (0.9 ml each) eluted with 0.01 M TRISHC1, 0.15 M NaCl, pH 7.3, were assessed by various immunoassays. Holo-hCG-like activity measured by 3/ 6-11/6-IRMA eluted in Fr. 57 to Fr. 72 with a maximum in Fr. 63, which corresponded well with the position of 125 I-hCG. For comparison, LH-activity (hCG/LH-RIA) ranged from Fr. 70 to Fr. 81 with a maximum rise in Fr. 73 (Fig. 3). Testing the same fractions in the hCG-/3 assay (2/6-12/17-IRMA) revealed two distinct peaks of the activity, one in Fr. 62 to Fr. 75 (maximum in Fr. 68) and the other in Fr. 87 to Fr. 99 (maximum in Fr. 92; Fig. 4B). Since some degree of cross-reactivity of LH-/3 had to be considered in this assay the elution pattern of purified LH-/3 (2.5 ng) on the same column was studied by measuring the eluted fractions in the hCG-/3-assay: the activity was found in Fr. 75 to 90 with a maximum in Fr. 83 and did not correspond with either of the two hCG-/3-immunoreactivity peaks of pituitary origin (Fig. 4, C and B). In another experiment, we added a defined
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
HOERMANN ET AL.
182
FlG. 1. Chromatography of urinary extract (50 mg) obtained from a postmenopausal woman on a column of Ultrogel Ac A 44. The column was eluted with 0.1 M TRIS-C1, 0.15 M NaCl, pH 8.0, and the fractions (2 ml) obtained were assessed for their content of holo-hCG (hCG), free hCG-0-subunit (hCG-/?), asubunit, and hCG/LH by the use of IRMAs and RIAs of different specificities described in Materials and Methods. The elution positions from the column of 125IhCG tracer and 12SI-a-subunit are indicated by arrows.
hCG-6 (pg/l) 0.4 T
hCG (pg/1) 41
12S
JCE & M • 1990 Vol71«Nol 12S
l-hCG
hCG-a (pg/t)
!-hCGi
I
0.3 .
hCG/LH-RIA (pg/1)
r 40
[40
30
" 30
20
• 20
10
•
-hCG/LH
0.2 hCG-or
0.1
10
100 Fraction (No.)
hCG-a (\igfl) hCG/LH-RIA (mg/l)
hCG (pg/i) hCG-6 (pg/l)
r 24
12-
10-
18 8-
12
6-
4-
40 holo-hCGactivity (\ig/l) 20 i
2-
60
B 10 -
50
60
70
80
90 Fraction (No.)
FlG. 2. Chromatography of pituitary extract (21 mg) on a column of ultrogel AcA 44 preequilibrated and eluted with 0.1 M TRIS-C1, 0.15 M NaCl, pH 8.0. The fractions (2 ml) were measured in various RIAs and IRMAs with different specificities for hCG and its subunits: hCG/LHRIA, specific holo-hCG-IRMA (hCG), specific hCG-/3-subunit-IRMA (hCG-0), and a-subunit-RIA.
amount of hCG-/3 (100 ng) to the pituitary extract before the chromatography of the material on AcA 44. This led to a rise of hCG-/?-immunoreactivity eluting in a position virtually identical to the first peak described above, while the second peak was apparently unaffected (Fig. 4D). Chromatography of the pituitary extract supplemented by jft-fragment of hCG (100 ng) resulted in an elevation of the second hCG-/?-immunoreactive peak without a change of the first peak (Fig. 4E).
40 SO LHactivity (\igfl)
60
70
80
90
100
110
120
60
70
80
90
100
110
120
20.000 1
10,000 -
40
50
Fraction No.
Affinity purification of pituitary hCG on hCG-antibodycoupled Sepharose
FIG. 3. Chromatography of a NaCl extract obtained from a single pituitary gland of a postmenopausal woman on a column of ultrogel AcA 44. The fractions (0.9 ml each) eluted with 0.1 M TRIS-C1, 0.15 M NaCl, pH 8.0, were assessed for holo-hCG- (B) and hCG/LH-immunoreactivities (C), as shown here. 125I-hCG was cochromatographed (A).
The pituitary extract (270 mg in 5 ml 0.15 M Na2HPO4, 0.2 M NaCl, pH 9.0) was affinity-purified on a column of cyanbromide-activated Sepharose coupled with the monoclonal antibody 3/6 that reacts with both holo-hCG and hCG-jff-subunit (Fig. 5). As expected, both holo-hCG-
and hCG-jS-immunoreactivity were adsorbed to the column and could subsequently be eluted from the column by 2 M KSCN, pH 5.0. In contrast, a-subunit- and LHactivity were apparently not adsorbed to the column and
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
HOLO-hCG AND hCG-/3 IN HUMAN PITUITARY
183
hCG (yg/l) hCQ-0 (yg/1)
hCG/LH-RIA (mg/1) hCG-a (mg/1)
200 -I
150 -
120
400
r 40
• 300
30
200
• 20
100
- 10
2 M KSCN, pH 5.0
hCG
100 -
50 -
0
10
20
30 Fraction (No.)
FIG. 5. Immunopurification of hCG from pituitary extract. Pituitary extract (270 mg) was applied to a column of Sepharose coupled with monoclonal hCG/hCG-/3-antibody 11/6 (see Materials and Methods). The column was preequilibrated with 0.15 M Na2HPO4, 0.2 M NaCl, pH 9.0, and eluted up to fraction 15 with the same buffer. The elution proceeded with adding 2 M KSCN, pH 5.0. Determination of the various immunoreactivities, i.e. holo-hCG (hCG), hCG-^-subunit (hCG-/3), asubunit, and hCG/LH in the fractions revealed a selective adsorption of holo-hCG- and hCG-/S subunit-activity by the column and the recovery of the adsorbed hCG-material from the column, as shown here. hCG-6 (pg/l)
r3
40 Fraction No.
FIG. 4. Chromatography of a NaCl extract obtained from a single pituitary gland of a postmenopausal woman on a column of ultrogel AcA 44. The fractions (0.9 ml each) eluted with 0.1 M TRIS-C1, 0.15 M NaCl, pH 8.0, were assessed for hCG-jS-immunoreactivity in a specific IRMA for the /3-subunit of hCG, as shown here. 125I-hCG was cochromatographed and the elution patterns were very similar in the various experiments (A). The pattern of the pituitary extract alone is shown in B. For comparison in C, the behavior of a large quantity of LH-/3 (2.5 Mg) was measured by the hCG-/3-IRMA. In other experiments, the pituitary extract was supplemented before chromatography on the same column with hCG-/3 (NIH CR123/3 100 ng, D) or a fragment of the 0subunit of hCG (100 ng, E).
eluted with the void volume (Fig. 5). This procedure resulted in an approximately 500-fold enrichment of hCG/hCG/3-like activity per gram protein.
40 Fraction FIG. 6. Con-A-Sepharose column chromatography of 20 mg pituitary extract. The column was preequilibrated and eluted with 10 mM TRISCl, pH 7.5, 0.15 M NaCl, 1 mM CaCl2, 1 mM MnCl2,1 mM MgCl2 up to fraction 10; elution then continued with the addition of 1 M a-methylglycoside. Fractions of 2 ml each were collected and tested for holohCG (hCG)- and hCG-/?-subunit (hCG-/3)-immunoreactivities.
Chromatography of pituitary extract on Con A-Sepharose When pituitary extract (20 mg) was chromatographed on a column of Con A-Sepharose as described in Materials and Methods, the fractions (2 ml) obtained before and after treating the column with 1 M a-methyl-glucoside were assayed for holo-hCG and free hCG-/3-subunit (Fig. 6). Approximately 75% of the total holo-hCG-immunoreactivity bound to the Con A-column and 25% was unbound. The ratio of the bound to unbound hCG-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
HOERMANN ET AL.
184
0-subunit immunoreactivity by the lectin was approximately 62%-38%. When a portion of the nonbinding fractions was rechromatographed in another experiment, no significant binding to the column was found. For LHactivity measured by RIA, the bound/unbound ratio was 84.6%/15.4%. Isolectric focusing of pituitary extract in granulated gel
Pituitary extract (220 mg) was subjected to preparative isoelectric focusing in granulated gel (pH 3-6). Thirty sections (0.5 cm each) were eluted with distilled water and measured for their immunoreactivities in specific assays for holo-hCG and hCG-/?-subunit. The peak of holo-hCG activity was found around pi 4.5, while the peak of hCG-/8 activity appeared slightly ahead of the holo-hCG peak at pi 3.5-5. In addition, there was a rise in hCG-/? activity close to the cathode. For comparison, in the same experiment LH-immunoreactivity focused also in a pi-range greater than 6 (Fig. 7). In other experiments, purified ^-fragment of hCG was subjected to electrofocusing and showed a pi about 9 (data not shown).
Discussion Although many studies have suggested secretion of hCG by the human pituitary (1-8), direct evidence is still hCG (pg/1) hCG-6 (M9/1) LH ^ 40-]
30"
-2
20 Fraction No.
30
FIG. 7. Isoelectric focusing of pituitary extract in granulated gel (12.5 X 26 cm). A linear pH gradient from pH 3-6 was applied. After focusing, the gel was sectioned into 30 x 0.5-cm slices eluted in distilled water. The fractions were measured in a specific holo-hCG-IRMA (hCG) and an IRMA specific for the free hCG-/3-subunit (hCG-/3). For comparison, LH-activity determined by RIA is also shown.
JCE & M • 1990 Vol 71 • No 1
lacking. Studies using serum and urine are ambiguous because of the possibility of hCG secretion by tissues other than the pituitary. Studies with pituitary extracts are subjected to doubt due to a lack of specificity of the hCG antibodies used. The expression of the a-subunit gene in the human pituitary has been well recognized (911); however, there are several genes for the hCG-j8subunit, and their regulation in the placenta is poorly understood; regulation of hCG-/3 genes in nontrophoblastic tissues is not understood at all (20). Obviously, the synthesis of intact hCG requires the presence of both an a- and a /3-subunit of hCG. In the present studies we have therefore focused our attention on demonstrating the presence of free hCG-/?-subunit in the human pituitary as well as further characterizing the immunological and physical properties of pituitary holo-hCG. To this
end, hCG- and hCG-/3-like materials in pituitary extracts were purified and characterized by means of gel filtration on Ultrogel AcA 44, affinity chromatography on Con ASepharose, affinity chromatography on hCG-antibodycoupled Sepharose, and preparative electrofocusing in granulated gel. The immunoreactivities of the fractions obtained were determined in a panel of assays for holohCG, hCG/LH, hCG/hCG-^-subunit, free hCG-,3-subunit, and a-subunit. With respect to the presence of holo-hCG, these studies provide several lines of evidence that the human pituitary contains intact holo-hCG rather than an hCGrelated peptide: first, the behavior of pituitary hCG during gel chromatography on AcA 44 is indistinguishable from that of cochromatographed 125I-hCG; second, the material exhibits immunoreactivity in a monoclonal IRMA with a high specificity for the holo-hCG molecule and negligible cross-reactivity with free hCG-/?-subunit and LH, since there was only negligible holo-hCG-immunoactivity in the fractions containing LH-peak-activity (Fig. 3); third, use of an hCG-antibody-coupled Sepharose column resulted in differential adsorption and selective enrichment of the material compared with LH; fourth, electrofocusing of the material occurred in a plrange very similar to that we and others have previously reported for pregnancy hCG (19, 21-23); and, fifth, the ability of most of the material (75%) to bind Con A Sepharose, similar to the Con A-binding properties of pregnancy hCG. In addition, there were other variant forms of pituitary holo-hCG that are apparently not bound by Con A-Sepharose. These appear to be considerably more prevalent in pituitary hCG than in pregnancy hCG, which contains only a minute amount of material not bound by Con A (17, 24). Thus, pituitary hCG exhibits microheterogeneity similar to that of hCG of tumorous origin from trophoblastic diseases or testicular cancer, which contains high proportions of Con Anonbinding forms (17). The content of hCG per gram
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
HOLO-hCG AND h.CG-,3 IN HUMAN PITUITARY pituitary extract was roughly estimated to be 61 fig, which was about 0.3% the content of LH. Our evidence for the presence of the free hCG-j8subunit in the human pituitary is the demonstration of material showing an immunoreactivity with a specific monoclonal antibody against hCG-|3-subunit. Interestingly, in the experiments with NaCl-extracts of single glands we found yet another peak of hCG-jS-immunoreactive material, eluting in a much lower molecular size range than that expected for hCG-0. The hCG-/3-IRMA used was sufficiently specific to distinguish pituitary hCG-jQ-immunoreactivity from that of the holo-hCG and other pituitary hormones and subunits (LH, FSH, TSH, FSH-/3, TSH-/?), except for LH-jS-subunit, which showed a cross-reactivity of 1.56%. Therefore, we applied LH-jSsubunit in a relatively high dosage to the same column, and measured the eluted fractions by hCG-0 assay. The single peak of immunoreactivity obtained with LH-/3 differed sharply in its position and pattern from pituitary hCG-/3-activity, its elution being clearly retarded compared to the first hCG-/3-activity peak. In another experiment, we spiked the pituitary extract with hCG-/3 (NIH CR123/3). This led to a rise of the hCG-,3-immunoreactivity that coincided with the first pituitary-hCG-/3-peak. Next, together with holo-hCG-, pituitary hCG-0-activity was differentially removed from LH/LH-/3-immunoreactivity and immunopurified by a Sepharose-coupled antibody that recognizes hCG-/?-subunit in both the holohCG and the free form. When subjected to isoelectric focusing, the pituitary hCG-/3-activity exhibited similar characteristics to those previously described for the hCGiS-subunit of pregnancy urine (21-23). Further, like holohCG of pituitary origin, pituitary free hCG-/? differed somewhat from its pregnancy counterpart in that it contained a relatively higher fraction of Con A-nonbinding variant forms (25%). The majority of the material, however, was bound by Con A. Taken together, the immunological and the various chromatographic studies provide convincing evidence for the existence of free hCG-j8-subunit, in addition to holohCG, in the human pituitary gland. The properties of hCG- and hCG-/5-like materials found in the human pituitary compared favorably with those of the respective purified standard preparations and of urinary material from a postmenopausal woman. Less clear at this point, is the nature of the second form of hCG-/5-like immunoreactivity whose molecular size on AcA 44-chromatography is much smaller than hCG-/?. Interestingly, the elution position of this material was the same as that of beta-fragment of hCG added to the pituitary NaCl-extract. A rise of hCG-/3 activity in the alkaline pi-range during electrofocusing of pituitary extract is also consistent with the assumption of the presence of a /?fragment. Further, we demonstrated a considerable
185
cross-reactivity of ^-fragment in the hCG-(8-IRMA. The preparation and properties of our /3-fragment have been described previously (14) and are similar to those reported by others for so-called j8-core fragment (13). One may speculate from these data on the presence of /3-core in the human pituitary, which would fit well with another recent report on the occurrence of /3-core fragment in serum (25). Further investigations, including measurement of pituitary extracts by specific immunoassays for 0-core, are warranted. Because of the possibility that the hCG we detected in the pooled pituitary extract was due to contamination by a specific specimen that contributed large amounts of hCG to the pool, NaCl-extracts obtained from single pituitary glands of two postmenopausal women (data of the second case similar, but not shown) were also investigated. These experiments indeed confirmed the presence of holo-hCG- and hCG-/?-subunit-like material in individual glands. Acknowledgment The authors are grateful to Dr. Robert E. Wehmann, National Institutes of Health, Bethesda, for his assistance in improving the style of the manuscript.
References 1. Braunstein GD, Kamdar V, Rasor J, Swaminathan N, Wade ME. Widespread distribution of a chorionic gonadotropin-like substance in normal human tissues. J Clin Endocrinol Metab. 1979;49:917-25.
2. Chen HC, Hodgen GD, Matsuura S, Lin LJ, Gross E, Reichert Jr LE, Birken S, Canfield RE, Ross GT. Evidence for a gonadotropin from nonpregnant subjects that has physical, immunological, and biological similarities to human chorionic gonadotropin. Proc Natl Acad Sci USA. 1976;73:2885-9. 3. Robertson DM, Suginami H, Hernandez Montes H, Puri CP, Choi SK, Diczfalusy E. Studies on a human chorionic gonadotropin-like material present in nonpregnant subjects. Acta Endocrinol (Copenh). 1978;89:492-505. 4. Yoshimoto Y., Wolfsen AR, Hirose F, Odell WD. Human chorionic gonadotropin-like material: Presence in normal human tissue. Am J Obstet Gynecol. 1979;134:729-33. 5. Stockell Hartree A, Shownkeen RC, Stevens VC, Matsuura SM, Ohashi M, Chen HC. Studies of the human chorionic gonadotropinlike substance of human pituitary glands in its significance. J Endocrinol 1983;96:115-26. 6. Odell WD, Griffin J. Pulsatile secretion of human chorionic gonadotropin in normal adults. N Eng J Med. 1987;317:1688-91. 7. Borkowski A, Puttaert V, Gyling M, Muquardt C, Body JJ. Human chorionic gonadotropin-like substance in the plasma of normal nonpregnant subjects and women with breast cancer. J Clin Endocrinol Metab. 1984;58:1171-8. 8. Stenman UH, Altthan H, Ranta T, Vartiainen E, Jalkanen J, Sepala J. Serum levels of human chorionic gonadotropin in nonpregnant women and men are modulated by gonadotropin-releasing hormone and sex steroids. J Clin Endocrinol Metab. 1987;64:7306. 9. Kourides IA, Weintraub BD, Ridgway EC, Maloof F. Pituitary secretion of free alpha and beta subunit of human thyrotropin in patients with thyroid disorders. J Clin Endocrinol Metab. 1975;40:872-85. 10. Landolt AM, Heitz PU, Zenklusen HR. Production of the alpha-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
186
11. 12. 13.
14. 15.
16. 17. 18.
HOERMANN ET AL.
subunit of glycoprotein hormones by pituitary adenomas. Path Res Pract. 1988; 183:610-2. Comi RJ, Gesundheit N, Murray P, Gorden P, Weintraub BD. Response of thyrotropin-secreting pituitary adenomas to a longacting somatostatin analogue. N Engl J Med. 1987;317:12-7. Korallus-Prinz K, Mann K, Moncayo R, Siddle K. hCG and free beta-subunit exist in the human pituitary (Abstract). Acta Endocrinol (Copenh). 1988;117(Suppl):287-90. Blithe DL, Akar AH, Wehmann R, Birken S, Nisula B. Purification of /3-core fragment from pregnancy urine and demonstration that its carbohydrate moieties differ from those of native human chorionic gonadotropin-/3. Endocrinology. 1988;122:173-80. Hoermann R, Amir SM, Nomura T, Ingbar SH. Design of a longlived thyrotropin antagonist from derivatives of human chorionic gonadotropin. Endocrinology 1989;124:223-32. Siddle K, Gard T, Thomas D, Cranage MP, Coombs RRA. Red cell-labelled monoclonal antibodies for assay of human chorionic gonadotropin and luteinising hormone by reverse passive haemagglutination. J Immunol Methods. 1984;73:169-76. Mann K, Siddle K. Evidence for free beta-subunit secretion in socalled human chorionic gonadotropin-positive seminoma. Cancer. 1988;62:2378-82. Mann K, Karl HJ. Molecular heterogeneity of human chorionic gonadotropin and its subunits in testicular cancer. Cancer 1983;52:654-60. Schleyer M, Voigt KH. Preparation of highly purified human
19. 20. 21. 22. 23.
24. 25.
JCE & M • 1990 Vol 71 • No 1
somatotropin (growth hormone). Hoppe-Seylers Z. Physiol. Chem. 1977;358:1557-64. Mann K, Schneider N, Hoermann R. Thyrotropic activity of acidic isoelectric variants of human chorionic gonadotropin from trophoblastic tumors. Endocrinology. 1986;118:1558-66. Fiddes JC, Talmadge K. Structure, expression, and evolution of the genes for the human glycoprotein hormones. Recent Prog Horm Res. 1984;40:43-74. van Hell H. Purification and characterization of urinary hCG. In: Moudgal NR, ed. Gonadotropins and Gonadal function. New York: Academic Press Inc; 1974;66-78. Merz WE, Hilgenfeldt U, Doerner M, Brossmer R. Biological, immunological and physical investigations on human chorionic gonadotropin. Hoppe-Seylers Z. Physiol Chem. l974;355:1035-45. Yazaki K, Yazaki C, Wakabayashi K, Igarashi M. Isoelectric heterogeneity of human chorionic gonadotropin: presence of choriocarcinoma specific components. Am J Obstet Gynecol. 1980;138:189-94. Dufau ML, Tsuruhara T, Catt J. Interaction of glycoprotein hormones with agarose concanavalin A. Biochem Biophys Acta. 1972;278:281-92. Alfthan H, Schroeder J, Stenman UH. Demonstration of core jShCG fragment in serum of pregnant women and cancer patients (Abstract). XVIIth Meeting of the International Society for Oncodevelopmental Biology and Medicine, Theoretical and Clinical Cancer Research, Freiburg, West Germany. 1989;58.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 September 2014. at 08:00 For personal use only. No other uses without permission. . All rights reserved.
