Vol. 262, No. 23, Issue of August 15,pp. 11149-11155,198’7 Printed in U.S.A.
THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1987 hy The American Society for Biochemistry and Molecular Biology, Inc.
Gonadotropin-releasing Hormone Action uponLuteinizing Hormone Bioactivity in Pituitary Gland: Roleof Sulfation* (Received for publication, April 21, 1987)
Maria Luz Sardaiions, Angela Rosaria Solano,and Ernest0 Jorge Podestas From the Centro de Investigaciones Endocrinoldgicas, Hospital de N i h s , “Dr. R. Gutiirrez,” Buenos Aires, Argentina
Luteinizing hormone (LH)’ is a glycoprotein hormone synTheregulation of rat luteinizinghormone(rLH) thesized in the anterior pituitary gland under the control of bioactivity was studied in an in vitro system using GnRH and sex steroids. It is known that there are several isolated pituitaries frommale rats. Stored and released rLH was evaluated in terms of mass (I-LH), bioactivity forms of LH in the pituitary gland (1-5) and thatthese forms (B-LH), mobility in nonequilibrium pH gradient elec- differ in their biological properties (1, 3, 6). These different forms of LH cannot be distinguished on the basis of their trophoresis, and mannose and sulfate incorporation either in the presence or absence of gonadotropin- molecular weights, but they differ in their charge as observed releasing hormone (GnRH). GnRH increased the bio- in isoelectric focusing studies (1-5). It has also been established that steroid hormones can modulate the appearance of logical potency of stored and releasedrLH. The pituitary content revealed seven I-LH species these different forms of LH (4, 7). The question then arises, what is responsible for the differ(pH 7.2, 7.8,8.5,9.0,9.1,9.3, and 9.7) and fiveB-LH species (pH 8.5, 9.0, 9.2, 9.4, and 9.7). The major I- ences in charge and in biological properties of LH in the LH and B-LH peaks were at pH 9.0 and 9.2, respec- various pools found in the pituitary gland? Regarding the tively. I-LH peaks at pH 7.2 and 7.8 are devoid of biological properties of LH, it is well established that LH acts bioactivity; at these pH values, free rLH subunits are by binding specifically to receptors in its respective target tissues. It is also known that binding alone is not sufficient detectable. GnRH increases the amountof both I-LH and B-LH to elicit a biological response (8). Several studies have been material secreted into the medium, and themajor com- conducted in vitro to determine which portion of the LH ponent migratesat pH 8.5 and is probably the afi dimer. molecule is responsible for the biological activity. These studies have shown that the carbohydrate moieties are necessary [3H]Mannose and [3sS]sulfate can be incorporated into stored and released rLH (pH 7.2,7.8,9.0,9.1, and for this activity (9, 10). It is reasonable to suspect that the 9.3 and 7.2, 7.8, 8.5, and 9.0, respectively). GnRH different forms with different biological activity and di.fferent charges may differ in their carbohydrate content. It is, howdecreases [2-3H]maanose incorporation into secreted rLH. [3SS]Sulfatewas incorporated into I-LH released ever, difficult to combine these concepts since most of the spontaneously into themedium; the format pH 7.2 has carbohydrates are neutral. It has been recently demonstrated that ovine (11),bovine no biological activity andis probably the free a subunit. GnRH decreases the [3sS]sulfate-labeled rLH content and human ( E ) , and rat (13, 14) LH contain sulfated oligoof the pituitaryconcomitantly with a 500%increase in saccharides. These findings are highly interesting in view that [36S]sulfate-labeledreleased rLH, suggesting that, soon a difference in the sulfated oligosaccharide content may be after [3”Slsulfate is incorporated, sulfated rLH is re- the basis for the heterogeneity in charge among the forms of leased. Sulfatase action on released rLH reveals that LH with different biological properties found in the pituitary sulfation may be related to release of rLH but that gland. However, evidence has been presented that thecharge sulfate residues are not involved in the expression of heterogeneity of rat LH isrelated to terminal sialic acid residues (15). Therefore, it was of interest to study the possible rLH bioactivity. In conclusion, GnRH stimulates carbohydrateincor- association among sulfated oligosaccharide content, isoelecporation andprocessing of the oligosaccharide residues tric point,and biological activity (e.g.,testosterone production giving thehighest biological potent rLH molecule and by isolated Leydig cells) among the various forms of LH. The also increases sulfation; this step is closely related to studies were performed in pituitary-stored LH as well as in the step limiting the appearanceof LH in the medium released LH under the control of GnRH. in theabsence of GnRH. EXPERIMENTAL PROCEDURES
* This work was supported by Grant 3082400/85 from the Consejo Nacional de Investigaciones Cientificas y Ticnicas (CONICET) and in part by a grant-in-aid from “Fundaci6n Albert0 J . Roemmers,” Buenos Aires, Argentina. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. To whom correspondence should be addressed Centro de Investigaciones Endocrinol6gicas, Hospital de Niiios “Dr. R. Gutikrrez,” Gallo 1330, 1425 Buenos Aires, Argentina.
Materials-Collagenase type I was obtained from Worthington. GnRH (Luteoliberina) was purchased from Laboratorios Elea (Buenos Aires, Argentina). ~-[2-~H]Mannose (27.2 Ci/mmol), %S as sodium sulfate (10-1000 mCi/mmol), and [35S]methionine(1.125Ci/ mmol) were obtained from Du Pont-New England Nuclear. AcrylThe abbreviations used are: LH, luteinizing hormone (prefix r indicates rat); GnRH, gonodotropin-releasing hormone; hCG, human chorionic gonadotropin; NaDodSO,, sodium dodecyl sulfate; NPGE, nonequilibrium pH gradient electrophoresis; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonicacid; TEMED, N,N,N’,N’-tetramethylethylenediamine; I-LH, rat LH measured by radioimmunoassay; B-LH, rat LH measured by bioassay; LH-1-5, rLH standard.
Rat LH Bioactivity and Sulfation
amide, N,N'-methylenebisacrylamide,and TEMED were purchased from Bio-Rad, and ampholytes obtained from LKB-Produkter AB (Bromma, Sweden). Bovine serum albumin (radioimmunoassaygrade) and sulfatase (arylsulfatase, aryl-sulfate sulfohydrolase from limpets (Patella uulgata), type V, 18 units/mg) were purchased from Sigma Anzmals-Adult male rats (60 days old) were used in these experiments. The animals were killed by decapitation. The bloodwas collected from the trunk, and the pituitary gland was removed immediately, as previously described (7). The anterior pituitary, separated from the posterior lobe, was washed with phosphate-buffered saline and used in the incubation studies. Pituitary Incubations-Anterior lobes in halves were first incubated in medium 199 containing 0.1% bovine serum albumin, 0.07% HEPES, and [2-3H]mannose (20 pCi/gland, one gland in 250 pl) for 2 h at 37 "C. The medium was then removed, the glands were washed, and a fresh medium with or without GnRH at a final concentration of 50 ng/ml was added. [35S]Sulfate(500 pCi/gland, three glands in a final volume of 250 pl) and [35S]methionine(50 pCi/gland, one gland in a finalvolume of 200 pl) incorporation into rLH was performed in Krebs-Ringer bicarbonate buffer, pH 7.4, containing 0.18% glucose and 0.5% bovine serum albumin for 5 h. The medium was then removed, and the glands were washed, followed by a SO-min incubation in the presence or absence of GnRH at a final concentration of 50 ng/ml. In the experiments of sulfatase action, the enzyme (1.8 units) was added to the incubation medium during the last 60 min of GnRH treatment. In the experiments with endoglycosidase F treatment, aliquots of pituitary homogenates labeled with [Y3]sulfate (436 pgof protein) and [%S]methionine (308 pg of protein) and aliquots (250 pl) of the incubation media were incubated with 0.15 unit of endoglycosidase F (Diplococcuspneumoniae) (a generous gift of Prof. L. F. Leloir and V. Idoyaga Vargas, Instituto de Investigaciones Bioquimicas, Fundacion Campomar) in 100 mM phosphate buffer, pH 6.1,50 mM EDTA, 0.5% Nonidet P-40 in a final volume of 200 pl for 42 h at 37 'C and subjected to NaDodS04-polyacrylamide gel electrophoresis. All incubations were performed under an atmosphere of 95% 02, 5% CO,. At the end of the incubation, the medium was removed and immediately processed or frozen a t -70 "C for later analysis. The pituitary tissue was washed with medium 199 or Krebs-Ringer bicarbonate buffer and homogenized in water as previously described (7); homogenization with standard buffers gave 50% less recovery of the rLH extracted from the gland. After centrifugation for 10 min at 200 X g, the supernatant was immediately processed or frozen a t -70 "C for later determinations. Studies were performed on the incubation medium and tissue homogenate before and after isolation of the different forms of LH by isoelectric focusing. Aliquots were takento determine (i)LH concentration using the LHradioimmunoassay, (ii) LH concentration using a rat testis interstitial cells bioassay, and (iii) carbohydrateand sulfate contents after specific LH immunoprecipitation. Rat LH Radioimmunoassay (I-LH)-Hormones and antisera used in the radioimmunoassay were kindly provided by the National Institute of Arthritis, Metabolism, and Digestive Diseases Rat Pituitary Hormone Distribution Program. Iodination of rat LH-1-5 was performed by the lactoperoxidase method (16) and purification as previously described (17). The general procedure for radioimmunoassay followed kit instructions with minor modifications (7). Results are expressed in terms of LH-1-5. This standard was shown to have a potency of 1 ag of LH-I-5/ng of LHRP-2 standard, and we used the former to express our results since, in our hands, theLHRP-2 standard is not bioactive (at least up to a concentration of 100 ng/ ml). The minimal detectable dose of LH-1-5 is 0.07 ng/tube. Rat LH Bioassay (B-LH)-Measurements of LH bioactivity in pituitary homogenates and incubation media were performed using the ratinterstitial cell testosterone production as previously described (17, 18). Decapsulated testes were incubated with collagenase a t 0.25 mg/ml/testis for 10 min at 34 "C with continuous shaking at 100 cycles/min. The final suspension contained lo6 cells/ml of medium 199 containing 0.1% bovine serum albumin. Results are expressed in terms of LH-1-5,which showed a minimal detectable dose of 0.18 ng/ incubation vial. Immunoprecipitation-In vitro mannose- or sulfate-labeled rat LH was isolated by immunoprecipitation (19) using anti-rat LHNational Institute of Arthritis, Metabolism, and Digestive Diseases antiserum. The rLH content inthe samples was evaluated by radioimmunoassay, and we also evaluated the radioactivity incorporated into the total protein; this enabled us to calculate the aliquot to be immunoprecip-
itated with 85-90% of recovery. The values are the specific counts immunoprecipitated. The nonspecific counts were evaluated omitting the first antibody or by addition of an excess of unlabeled rat LH; in both cases, the nonspecific counts were in theorder of 9% of the total counts. The efficiency of the second antibody was studied by the use of protein A. The supernatant from the second antibody precipitation was incubated with 100 units/ml protein A in 0.02 M Tris-HC1, pH 7.6,O.Ol M EDTA, 0.15 M NaCl, 0.1% (v/v) Triton X-100. In addition, the protein A procedure was performed before the second antibody. In both cases, specific counts were not detected on the second precipitation. The nonspecific precipitation with protein A in the absence of anti-rLH antibody was 2.5%.Even if the nonspecific precipitation using the second antibody technique was higher than thenonspecific precipitation obtained with the protein Amethod, the values are still low; therefore, the second antibody technique was used throughout the experiment. Antiserum was used in a concentration 10 times sample which higher than theamount needed to precipitate LH in the was calculated by radioimmunoassay. Nonequilibrium pH Gradient Electrophoresis (NPGE)-NPGE studies were performed in polyacrylamide gels (18) and according to Chrambach et al. (20) with a nonequilibrium pH gradient (21, 22). and 0.02 M NaOH, The upper and lower buffers were 0.01 M respectively. The final gel preparation contained 4.36% acrylamide, 0.14% N,N'-methylenebisacrylamide, 10% glycerol, 1.2% pH 7-10 ampholytes, 3.5% pH 3-10 ampholytes, 0.2% TEMED, and 0.4% ammonium persulfate. 2.5 ml of the final preparation were loaded in 6-mm diameter gel tubes and polymerized in the dark. The samples were prepared in the presence of 10% sucrose and 5% pH 3-10 ampholytes in afinal volume of 200pl. Electrophoresis was performed a t 4 "C for 10 min a t 200 V and for 2 h at 500 V. The isoelectric mobility of the LH subunits in our system was assessed by running a pituitary sample in the presence of 4 M urea. After electrophoresis, the gel wassliced, and theeluted fractions were assayed by rLH radioimmunoassay. Fig. 1 is the pattern obtained under these conditions, showing the subunits appearing a t pH 7.2 and 7.8. Similar patterns were obtained when "'1-rLH-1-5 was treated under same conditions as thepituitary samples (data not shown). NaDodS04-Polyacrylamide Gel Electrophoresis-NaDodS0,-polyacrylamide gel electrophoresis was performed as previously described (21, 22), with minor modifications. The separating gel was a polyacrylamide slab gel (1 mm thick) containing acrylamide and N,N'methylenebisacrylamide (150.8); the samples were prepared in 0.2 ml of electrophoresis sample buffer (3.3% NaDodS04, 13.8% (v/v) 0.5 M Tris-HC1, pH 6.8, 5.5% (v/v) mercaptoethanol, 0.01% bromphenol blue, 11.1% (v/v) glycerol), transferred to a boiling water bath for 2 min, and kept frozen until processed. 50 pl(80-120 pg of protein) were subjected to electrophoresis; rat LH standardwas treated in the same manner and used as marker for the (Y and (3 subunits. The running buffer was Tris/glycine, pH 8.3, 0.1% NaDodSO4. The gels were prepared the day before use, and electrophoresis was carried out
FIG. 1. NPGE of rat LH subunits. Rat LH from pituitary homogenates with immuno- and bioactivity migrating at pH 9.1 in NPGE was treated with 4 M urea and analyzed by NPGE in the presence of 4 M urea. The gel was sliced, and the eluted fractions were assayed by rat LH radioimmunoassay. . . . ., pH gradient. The values are expressed in nanograms of rLH-I-5/slice.
Rat LH Bioactivity and Sulfation a t constant current (10 mA) and at an initial 150 V for 6 h. At the end of electrophoresis, the gels were immersed in 50% (w/v) trichloroacetic acid/water for 30 min, stained with 1%Coomassie Blue in 50% trichloroacetic acid (30 min),and destained in acetic acid/ methanol/water (1:1089, v/v) overnight. They were then dried and subjected to autoradiography. RESULTS
I n Vitro GnRH Action on Pituitary LH Bioactiuity-The in vitro effect of GnRH was investigated in isolated pituitaries from male rats. rLH released into the medium and the pituitary content of rLH were evaluated by radioimmunoassay (ILH) and bioassay (B-LH). The results are shown in Table I. GnRH produced a 25% decrease in thepituitary content of ILH; however, the pituitary content of B-LH proved to be very similar in the absence or presence of GnRH. GnRH increased the release of I-LH 3.9-fold, but the GnRH stimulation was 5.7-fold when the incubation medium was evaluated in terms of B-LH. These results suggest that the evaluation of the rLH level by I-LH is not a directindication of the LHbioactivity present inthe sample. In fact,there is immunoreactive material, TABLEI In vitro effect of GnRH on immunological and biological active rLH in the pituitary gland andreleased into the incubation medium Values are mean f S.D. ( n = 3) expressed in terms of rat LH-1-5. rLH content in:
@&?I&& Pituitary No additions GnRH
14.47 f 0.72 14.96 f 1.01
13.13 f 0.71 10.82 2 0.52
&I Medium No additions GnRH
0.35 0.03 1.35 k 0.09
FIG.2. rLH immunoactivity (- - -) and bioactivity (-) from pituitaries incubated in the presence or absence of GnRH analyzed after NPGE. A , pituitary rLH in the absence of GnRH; B , in the presence of GnRH; C, rLH released into the medium in the absence of GnRH; D,in the presence of GnRH. After the pituitaryincubation, the homogenate of the gland andthe incubation mediumwere analyzed by NPGE. The gelswere sliced, andthe elutedfractions were analyzed by radioimmunoassay (- - -) and bioassay (-). . . . ., pH gradient. Values are expressed in terms of nanograms of rLH-15/vial.
0.45 f 0.04 2.58 f 0.08
released mostly in the absence of GnRH, that is not fully bioactive. These observations could be related to theheterogeneity of rat LH. We further studied this point by analysis of stored and released rLH in the pituitary by isoelectric focusing polyacrylamide gel electrophoresis. NPGE of Rat LH in Polyacrylamide Gels-After NPGE, the gels were sliced, and the eluted fractions were evaluated for their content of B-LH andI-LH. Results are shown in Fig. 2. The pituitary content revealed several immunoreactive rLH species appearing at pH 7.2,7.8,8.5,9.0,9.1, and 9.3 (Fig. 2 A ) . The major immunoreactive species appears at pH 9.0. When the same material is evaluated in terms of its bioactivity, only five species can be defined which migrate at pH 8.5, 9.0,9.2,9.4, and 9.7. The major bioactive species migrates at pH 9.2. Immunoreactive material migrating at pH7.2 and 7.8 is absolutely devoid of bioactivity. At these pH values, free rLHsubunitsare detected (see Fig. 1 and "Experimental Procedures"). The immunoreactive material at pH 7.2 and 7.8 contributes 16 and 12% of the total material loaded onto the gel, respectively. The results obtained in rat pituitary gland afterstimulation with GnRHare shown in Fig. 2B. The general pattern is maintained, when compared with the pituitary content in the absence of GnRH. However, the relative amount of some species is different. The major immunoreactive material is now obtained at pH 9.3. This increase in immunoactivity is associated with an increase in bioactivity in the same species at pH 9.3. The species migrating at pH 8.5 doubled its bioactivity in comparison to the same specie stored in the pituitary when notstimulated with GnRH. Species appearing at pH 7.2 and 7.8 are again absolutely devoid of bioactive material. TheNPGEpatterns of the products secreted intothe medium as a resultof incubation of pituitaries in the absence of GnRH are shown in Fig. 2C. There are threeimmunoreac-
Rat LH Bioactivity and Sulfatwn
tive products migrating at pH 7.2, 8.5, and 9.0. The major immunoreactive product migrates at pH 7.2. Bioactivity is only detected in products migrating at pH 8.5 and 9.0. Thus, the major product secreted into the medium may correspond to one of the subunits of the LH a:/3 dimer. The patterns obtained by NPGE analysis of the incubation medium from pituitaries in presence of GnRH are shown in Fig. 2 0 . The absolute amount of immuno- and bioactive material was much greater than that in the absence of GnRH. The major immuno- and bioactive species both migrate at the same pH (8.5). Immunoreactive species at pH 7.2 and 9.0-9.2 are still present in the medium. The species migrating at pH 7.2 is again devoid of bioactivity. [2-3H]Mannose and f‘S1Sulfate Incorporation in rLH in Isolated Rat Pituitaries-Sulfation is one of the final events in the maturation of the subunits andoccurs after theirjoining (14). There is a covalent linkage of sulfate to both subunits of bovine and rat LH, and the sulfate is linked to the sugar of LH (12, 13). Labeling with sulfate should provide a useful new approach to examine the synthesis and maturation of pituitary glycoprotein hormones. In addition, it is also known that the subunits are translated from separate mRNA (2325), and high mannose oligosaccharide units are transferred to asparagine residues in thenascent chains as theycross the endoplasmic reticulum (26). Therefore, in the following experiments, we studied the [2-3H]mannose and [35S]sulfate incorporation into the pituitary and secreted rLH under the influence of GnRH and the association among sulfate content, isoelectric point, and biological activity in the various isohormones. Pituitaries were incubated with labeled mannose or [35S]sulfateas indicated under “Experimental Procedures,” and labeled rLH was isolated by immunoprecipitation from either the gland or the incubation medium. Under our experimental conditions, it is possible to [3H]mannose label rLH stored in the pituitary and also LH released (Table 11).In the presence of GnRH, thereis an increase in rLH-specific activity in the pituitary; however, GnRH produces a decrease in the [3H]mannose incorporation in secreted rLH when compared with spontaneously releasable or pituitary rLH (Table 11). [35S]Sulfateis also incorporated into both pituitary and releasable rLH (Table 11). Moreover, these incorporations are under the control of GnRH. In fact, the specific activity in the pituitary decreases 37% when pituitaries are incubated in the presence of GnRH. The mirror image is found in secreted LH under the action of GnRH; the specific activity of [35S] sulfate-labeled rLH increases 43% with respect to spontaneously secreted LH. These results imply that the sulfate is
incorporated into a pool of immunoreactive LH ready to be released under the influence of GnRH and/or that GnRH produces the increase in sulfate incorporation. This step is closely related to the impaired step limiting the appearance of LH in the medium in the absence of GnRH. In order to provide some evidence that part of the sulfate label is linked to thesugar of LH, [35S]sulfate-labeledrat LH was treated with endoglycosidase F whichrecognizes both high mannose and complex-type oligosaccharides (27), presumably the structure where the SO, is inserted (12); as a control, [35S]methionine-labeled LH was also subjected to endoglycosidase F treatment. As shown in Fig. 3, [35S]methionine-labeled LH a and @ subunits from pituitary or released rat LH showed no difference with or without gtycosidase treatment. On the other hand, endoglycosidase F treatment partially reduced the [35S]sulfatelabel into the LH (Y and /3 subunits. Thereis a decrease in some [35S]methionine-labeled protein bands in the autoradiograms from endoglycosidase F treatment, suggesting the presence of protease activity during the 42-h incubation period; however, this protease activity does not seem to affect the LH a: and /3 subunits as seen on the gels from incubations with [35S]methionine(Fig. 3, lanes 1-4).
When the incorporation of [2-3H]mannose is analyzed after the isolation of rat LH by NPGE, the label co-migrates with several pituitary isohormones with pH 7.2, 7.8, 9.0, 9.1, and 9.3(Fig. 4A). [2-3H]Mannose incorporation intospontaneously released rLH analyzed by NPGE is shown in Fig. 4C. The label is associated with isohormones at pH 7.2, 7.8, 8.5, and 9.0. GnRH produced a marked increase in the label within the pituitary isohormones migrating at pH 9.1 and 9.3 (Fig. 4B).The effect of GnRH on the species released into the medium is shown in Fig. 40. There was an increase in the label associated with the isohormones at pH 7.2 and 8.5; however, the increase is lower than the increase in the immuno- or bioactive material migrating at the same pH (Fig. 20). Thespecies at pH 9.0 is devoid of radioactivity. NPGEstudiesin [35S]sulfate-labeledpituitaryrLHare shown in Fig. 5 ( A and B ) . Radioactivity is associated with immunoprecipitable forms at pH 7.2, 7.8, 8.5, and 9.1 (Fig. 5A). GnRH produces a marked decrease in all the sulfatelabeled isohormones (Fig. 5 B ) . Spontaneously released rLH (Fig. 5C) shows radioactivity associated with an immunoreactive material at pH7.2. GnRH produces an increase of [35S]sulfatecontent in a secreted immunoreactive material at pH 7.2 and the appearance of label in the pH 8.5 form (Fig. 50).
TABLEI1 Mannose and sulfate incorporation into immunoprecipitable rLH stored in the gland and released in the medium from isolated pituitary incubations in the presenceor absence of GnRH Values are mean f S.D. (n= 31. [“%]Sulfate
Pituitary No additions GnRH
21,780 f 1,380 30,933 ? 1,910
1,658 f 775 2,858 f 978
Medium No additions GnRH
4,825 f 318 2,840 f 149
13,402 f 980 2,103 & 108
Total 26,605 k 1,830 No additions 33,773 f 1,943 GnRH a cpm/Fg = cpm/pg of I-LH.
16,920 f 968 8.760 k 320 9,500 f 456 53,500 f 2,254 26,420 f 1,710 62,266 f 3,928
1,288 f 115 809 f 180 26,388 +- 965 39,629 f 888
Rat LH Bioactivity and Sulfation 2
FIG.3. Pituitary was incubated with [%]methionine or [s5S]su1fate as described under "ExperimentalProcedures." The homogenized tissue ( A and C)and themedium ( B and D )were incubated with (+) or without (-) endoglycosidase F.After incubation, the samples were properly treated andsubjected to NaDodS0,-polyacrylamide gel electrophoresis; under these conditions, rLH dissociates in its subunits. Lanes 1-4, [35S]methionine-labeledproteins; lanes 5-8, [35S]sulfatelabeled proteins.
FIG.4. [2-SH]Mannose incorporation into pituitary andreleased rLH analyzed by NPGE. Pituitary glands were incubated with [2-3H]mannose as described under "Experimental Procedures." After incubation, the homogenate from the gland and theincubation medium were analyzed by NPGE. The gels were sliced, the eluted fractions were analyzed by bioassay (-), and a fraction of the eluate was immunoprecipitated and counted in a 8-scintillationcounter (- - - ) . A and B, profiles of pituitary homogenate after incubation in the absence or presence of GnRH, respectively; C and D,profiles of medium from incubations in the absence or presence of GnRH, respectively. .. . .,pH gradient.
FIG.5. [a6S]Sulfate incorporation into pituitary and
leased rLH analyzed by NPGE. Pituitary glands were incubated with [35S]sulfateas described under "ExperimentalProcedures." After incubation, the homogenate from the gland and the incubation medium were analyzed by NPGE. The gels were sliced, the eluted fractions were analyzed by bioassay (-), and a fraction of the eluate was immunoprecipitated and counted in a @-scintillationcounter (- - - ) . A and B, profiles of pituitary homogenate after incubation in the absence or presence of GnRH, respectively; C and D,profiles of medium from incubations in the absence or presence of GnRH, respectively. . . ..,pH gradient.
not affected by the presence of sulfatase (I-LH = 1.71 & 0.17 versus 1.60 & 0.14 pg/vial and B-LH= 5.31 f 0.60 versus 5.07 +. 0.6 pglvial in the presence versus absence of sulfatase, respectively); B-LH activity after sulfatase treatment was always higher than without treatment, but the differences were statistically nonsignificant. However, there was a marked change in the isoelectric focusing mobility, as shown in Fig. 6. In fact, 65% of B-LH migrates at pH 8.5 and 35% at pH 9.15 in medium not exposed to sulfatase activity; whereas after sulfatase action, 70% of B-LH was found at pH 9.05 and 30% a t pH 9.7. Similar switch of the pH values was found when analysis after isoelectric focusing was made by radioimmunoassay (I-LH); thisshows that the sulfatase acts also upon the free subunit, changing the pH from 7.5 in the nontreated incubation medium to 8.2 in the sulfatase-exposed medium. I-LH also reveals that thebroad species at pH9-9.3 in the nontreated medium is split in two peaks after sulfatase action appearing at pH 9.2 and 9.7. DISCUSSION
The regulation of rat LH bioactivity was studied in an in vitro system using isolated pituitaries from male rats. We analyzed stored and released rLH in terms of mass (I-LH), bioactivity (B-LH), mobility in NPGEand mannose and sulfate incorporation either in the presence or absence of GnRH. GnRH not only promotes the release of LH into theincubation medium, but also the conversion of pituitary pools of Biological activity, [2-3H]mannose, [35S]sulfate,and im- low bioactive LH into apool with higher bioactivity, probably munoreactive material are associated with the isohormones the releasable pool. Then, having no increase of I-LH immuat pH 8.5 (Figs. 20, 40, and 5 0 ) . As shownfor [2-3H] noactivity, the increase in bioactivity could be explained by mannose, [35S]sulfateimmunoreactivematerial at pH7.2 does an effect of GnRH in the LH molecule at the level of the not exhibit any biological activity. defining steps, after which the hormone becomes releasable. The role of sulfate residues was assessed by exposure of This suggestion came from the fact that total immunoactive released rLH to sulfatase action. This sulfate-free rLH was LH (pituitary + released) reflecting the amount of the hor-
Rat LH Bioactivity and Sulfation
which one is the target for GnRH tochange its bioactivity. It is important to mention that other authors (1-6) have suggested an effect of GnRH upon the LH molecule during the stimulation of its release from the pituitary, but this concept 100 ’15! A was not further investigated. Mukhopadhyay et u1. (28) have also found in an in vitro pituitary study an increase in LH bioactivity released into theincubation medium under GnRH stimulation, but they did not study the pituitary content in its absence. In agreement with the effect of GnRH under our experimental conditions, the released hormone alwaysshowed higher bioactivity than stored rLH, not only in this in vitro study and others (29, 30), but also in an in vivo analysis (7). Theseresults couldbe explained by the presence, in the pituitary, of at least two different pools of LH with different biopotencies. Isoelectric moieties of rLH isohormones from rat pituitary have been described by several authors (1-5). Immunoquantification of rLH subunits secreted by rat isolated anterior pituitary cells and the effect of GnRH have also been described (30). However, the effect of GnRH on the isoelectrical behavior of immunoreactive and biological activities of rLH in the pituitary and rLH secreted into the medium has not been previously studied. Our results on NPGE of pituitary rLH show seven species of immunoreactive material. These results arein agreement with previous works (1,2,4),showing also several species of rLH which migrate at similar pH. Moreover, the species at -pH 9.0, whichhave the major immunoreactive material, has also been described (1, 2). The major bioactive form migrating at pH9.2 has also been shown previously (1). There are two immunoreactive species at pH s11ce n u m h r 7.2 and 7.8 which are devoid of any biological activity, and FIG. 6. NPGE analysis of incubation medium from pituitarthey migrate at thesame pH as theisolated LH subunits (see ies incubated with GnRH. During last 30 min of incubation, 1.8 units of sulfatase were added ( B ) or not ( A ) ,and the medium was “Experimental Procedures” and Fig. 1). The contribution of immunoreactive subjected to NPGE analysis. After NPGE, the gel was sliced, andthe both species is 16 and 12% tothetotal elutedfractions were assayedfor I-LH (-- -) and B-LH (-). material, respectively. These forms were detected with an . . . . , pH gradient. anti-rat LH antiserum, which cross-reacted 33% with the (Y subunitand 100% with the /3 subunit of theLH dimer. mone is approximately the same in thepresence (13.48 pg) or Therefore, the contribution of the (Y subunit is 48% to the absence (12.17 pg) of GnRH (0.02 < p < 0.05), whereas the total immunoreactive material. These values are close to the the released pool is estimates of Grotjan et al. (30) of uncombined rLH a subunit bioactive LHcontentin the gland significantly higher in the presence (17.54 pg) than absence in pituitary cell extract after isolation by Sephadex G-100 superfine. These findings are supported by NPGE studies of of GnRH (14.92 pg) ( p < 0.01). This concept is strengthened by the analysis of the biopo- the incubation medium where the major immunoreactive matency/unit of mass obtained from stored and released LH. In terial migrates at pH 7.2 as described in bovine andrat fact, the pituitary biopotency/unit of mass is 1.38 & 0.07 in pituitary, where predominantly a single subunit, the a! subglands incubated in the ‘presence of GnRH; whereas in its unit, appeared to be secreted (13,30). These data areconsistabsence, it is 1.1 k 0.06 ( p < 0.001). These results are even ent with other observations showing a preponderance of a more clear in released rLH, which shows a biopotency/unit subunits in media containing choriocarcinoma cells (31) and of mass of 1.90 & 0.08 in LH released during GnRH action mouse pituitary cells (32). Grotjan et ul. (30) have also shown compared with 1.2 f 0.04 for rLH released spontaneously. that unstimulated cultures appear to release a large excess of However, it is worthwhile pointing out that, in view of the rLH a subunit and aminimal amount of LH /3 and a/3 dimer. findings in this and other publications related to the hetero- In addition, this immunoreactive material migrating at pH geneity of rat LH in the pituitary (1-4), as well in released 7.2 is devoid of any biological activity. The origin of the free unassociated (Y subunit in the medium rLH (this report), it is important to note the concept of the biopotency/unit of mass or bioactive:immunoactive ratio, as is not clear. Based on preliminary pulse-chase experiments, usually expressed. This ratio would only reflect the potency Hoshina and Boime (14) suggested that itis not derived from of every species if that species is isolated, i.e. not contaminated uncoupling of LH because a free /3 subunit in the medium was with other immunoactive molecules, and this is not true yet. not observed. However, it is possible that uncoupled /3 subIn fact, the released most potent biological hormone has units are preferentially degraded (32). They also suggested different mobility than the most potent biological hormone that enhanced degradation of LH may explain the small in the pituitary (Fig. 2, A and C), and one explanation for amount of LH secreted into the medium under basal condithis discrepancy could be the presence of more than one tions. They proposed that an essential component is absent immunoreactive species with similar mobility. Therefore, it is or limiting (e.g. LH-releasing hormone), and it may be that very difficult yet to determine which of the species of immu- the impaired step limiting the appearance of LHin the noreactive material is responsible for the bioactivity and also medium occurs after sulfate attachment. In agreement with
Sulfation Rat LH and Bioactivity previous works (13, 14,33), we also found [35S]sulfateincorporated into species liberated spontaneously into themedium which has no biological activity and is probably the free a subunit. In addition,ourresults also showed thatGnRH produces a dramaticchange in the amount of immunoreactive and also bioactive material secreted into the medium. The major component in terms of immuno- or bioactivity migrates at pH 8.5 and is probably the [email protected]
dimer. This species has the major [35S]sulfate-specificactivity. These findings suggested that GnRH treatment may produce the incorporation of sulfate into molecules of rLH isohormones which migrate before GnRH treatmentat pH 9.0 and after treatmentat pH 8.5 and which are readier to be released. These results are supported by the [35S]sulfate content in pituitaryand released rLH under the action of GnRH(Table 11). GnRH produces a decrease in the pituitary [35S]sulfate-labeled rLHcontent concomitantly with a 500% increase in [35S]sulfate-labeled released rLH, suggesting that, soon after [35S]sulfateis incorporated, the sulfated rLH is released. This observation is inline with the hypothesis that LH in the medium occurs after sulfate attachment (13, 14). This hypothesis opens a question: is this the physiological role of sulfateincorporation?LHand hCGhave similar physiological effects (34) and also bind to the same receptor. Because hCG lacks sulfate, this comparison eliminates the sulfate moieties as essential components in binding to receptor and receptor coupling. hCG contains sialic acid at the nonreducing termini of its oligosaccharides in a position analogous to sulfate in LH. Removal of sialic acid from hCG does not markedly affect binding, but it is rapidly cleared from circulation (35, 36). Sialic acid and sulfate possibly perform this same function. The finding that sulfation is markedly concomitant with the most potent biological active form of LH may open an alternative role of sulfate as part of the oligosaccharide necessary for LH function. However, the experiments exposing released rLH to sulfatase action changing the mobility of the rat LH species from pH 8.5 to 9.15, but preserving full biological activity, would rule out the latter hypothesis. The latter mobility is comparable to the most potent biological species found in thepituitary. It has been described that some sulfatase failed to hydrolyze sulfate from ovine pituitary LH (11); we used sulfatase type V from limpets which was able to release 50% of the sulfate label in rat LH andalso was able to change its electrophoretic mobility. This limpet sulfatase was able to release only small amounts of sulfate from ovine LH (11). It is worthwhile mentioning that the sulfatase treatment of rLH in our study was only performed in rat LHreleased into themedium from rat pituitaries under the action of GnRH. It is known that, in the synthesis of glycoproteins, there occurs a first mechanism, the so-called lipid-linked pathway involving the transfer of sugar from sugar nucleotide donors to lipid acceptor. An oligosaccharide composed of N-acetylglucosamine, mannose, and glucose is synthesized on the lipid carrier,and the entire oligosaccharide is transferred to a protein acceptor. Following transfer, the glucose residues and some of the mannose residues are removed enzymatically. This processed glycoprotein is the substrate to which are added peripheral sugars (37) and sulfate (11-14, 33, 38). We found that, in the most potent biological species released into the medium under the action of GnRH, there was an association between the low mannose-specific activity and thehigh sulfate content. These results may suggest that LHcan follow the pattern of glycoprotein synthesis described above. Nevertheless, GnRH also produces an increase in the mannose incorporation into pituitary rLH. Thus, the low mannose form
of rLH may occur before sulfation and release of rLH under the influence of GnRH, and sulfation is abiological pathway related to the release of the hormone, but is apparently not involved in the expression of the biological activity. In conclusion, these results would imply that GnRH stimulates incorporation of carbohydrates into the rLH molecule; these residues are further processed and complexed finally with sulfate, becoming an rLH ready to be released under the influence of GnRH. This stepof sulfation is closely related to the step limiting the appearance of LH in the medium in the absence of GnRH. Acknowledgments-We are very grateful to Mrs. Barcala for her expert technical assistance and to Dr. J. Lemos for his helpful comments and discussion of the manuscript. REFERENCES 1. Dufau, M. L., Nom, K., Dehejia, A., Garcia-Vela, A., Solano, A. R., Fraioli,
F., and Catt, K. J. (1982) in Pituitary Hormones and Rebted Pe tides (Motta, M., Zanisi, M., and Piva, F., eds) pp. 117-138, Academic (ress, -NPW .- .. Vnrk - -Keel, B.A., and Grotjan, H. E., Jr. (1984) Anal. Biochem. 142,267-270 Wakahayashi, K. (1977) Endocrinol. Jpn. 24,473-485 Keel, B. A,, and Grotjan, H. E., Jr. (1985) Endocrinology 117,-354-360 Robertson, D. M., Foulds, L. M., and Ellis, S. (1982) Endocrcnology 1 1 1 , "
2. 3. 4. 5.
3%5-391 ". "_
6. Hattori, M.A,, Sakomoto, K., and Wakabayashi, K. (1983) Endocrinol. Jpn. 3 0 , 289-296 7. Soland. A. R.. Garcia-Vela. A.. Catt. K. J.. and Dufau.M.L. (1980) . . Endocrinolo& 1 0 6 , 194111948 8. Podesta, E. J., Solano, A. R., Attar, R., Sanchez, M. L., and Molina y Vedia, L. (1983) Proc. Natl. Acad. Sci. U. S. A. 80,3986-3990 9. Bahl, 0.P., Man, L., and Moyle, W. (1974) in Hormone Binding and Cell Activation in the Testis (Dufau, M. L., and Means, A. R.,eds) pp. 125144 10. Kalyah, N. K., Lippes, H. A., and Bahl, 0.P. (1982) J. Biol. Chem. 2 5 7 , 12624-12631 11. Anumula, K. R., and Bahl, 0.P. (1983) Arch. Biochem. Bioph. 2 2 0 , 645651 12. Parson, T. F., and Pierce, J. G. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 7089-71x13 .- -- . - - 13. Hortin, G.,Natowicz,M., Pierce, J. G., Baezinger, J., Parson, T., and Boime, I. (1981) Proc. Natl. Acad. Sci. U. S. A. 78, 7468-7472 14. Hoshina, H., and Boime, I. (1982) Proc. Natl. Acad. Sci. U. S. A. 79,76497G.51
15. HaGiii, M.,Ozawa,K., and Wakabayashi, K. (1985) Biochem. Biophys. Res. Commun. 127,501-508 16. Thorell, J. I., and Johansson, B.G. (1971) Biochim. Biophys. Acta 2 6 1 , 363-367 17. Solano, A. R., Dufau, M.L., and Catt, K. J. (1979) Endocrinology 1 0 6 , 372-381 18. Podesta, E. J., Dufau, M. L., Solano, A. R., and Catt, K. J. (1978) J . BioL Chem. 253,8994-9001 19. Burek, C. L., and Frohman, L. A. (1970) Endocrinology 86,1361-1366 20. Chrambach, A,, Jovin, T. M., Svendsen, P. J., and Rodbard, D. (1976) in Methods of ProteinSeparation (Catsimpoolas, N., ed) pp. 144-176, Plenum Press, New York 21. O'Farrel, P. Z., Goodman, H. M., and O'Farrel, P. H. (1977) Cell 12,11331141 22. Tbtm_as,G., and Luther, H. (1981) Proc. Natl. Acad. Sci.U. S. A. 78,57123'116
23. Daniels-Mc Queen, S., McWilliams, D., Birken, S., Canfield, R., Landefeld, T., and Boime, I. (1978) J. Biol. Chem. 2 5 3 , 7109-7114 24. Chin. W. W.. Habener. J. F.. Kieffer., J. D.., and Maloof., F. (1978) . . J. Biol. C&m. 2 5 3 , 7985-7988 25. Kourides, I., and Weintrauh, B. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 298-302 26. Bielinska, M.,and Boime, I. (1978) Proc.Natl.Acad. Sci. U. S. A. 7 5 , '
27. Elder, J. H., and Alexander, S. (1983) Proc. Natl. Acad. Sci. U. S. A. 7 9 , 4540-4544 28. Mukhopadhyay, A.K., Leidenherger, F.A,, and Llchtenberg, V. (1979) Endocrinology 104,925-931 29. S h a r p , R. M., Shahmanesh, M., Ellwood, M. G., Hartog, M., and Brown, P. S. (1975) J. Endoerinol. 65,265-273 30. Grotjan, H. E., Jr., Leveque, N. W., Bertowitz, A. S., and Keel, B. A. (1984) Mol. Cell. Endoer. 3 5 , 121-129 31. Ruddon, R., Hanson, C., and Addison, N. (1979) Proc. Natl. Acad. Sci. U. S. A. 76,5143-5147 32. Chin, W.W., Maloof, F., and Habener, J. F. (1981) J. Bwl. Chem. 2 5 6 , 3059-3066 33. Corless, C. L., and Boime, I. (1985) Endocrinology 117,1699-1706 34. Birken, S., and Canfield, R. (1978) in Structure and Function of Gonudotropim (Mc Kerns, K. W., ed) pp. 47-80, Plenum Press, New York 35. Van Hell, H., Goverde, B. C.,Schuurs, A. H. W. M., Dejager,E., Matthijsen, R., and Homan, J. D. H. (1966) Nature 221,261-264 36. Van Hall, E. V., Vaitukaitis, J. L., Ross, G. T., Hickman, J. W., and Ashwell, G. (1971) Endocrinology 8 9 , 11-15 37. Bahl, 0.P., Reddy, M. S., Kalyan, N., and Henner, J. (1980) J. Sc. Ind. Res. (India) 39,734-744 38. Bahl, 0.P., Reddy, M. S., and Beddi, G. S. (1980) Biochem. Biophys. Res. Commun. 9 6 , 1192-1199