Brain Research, 166 (1979) 321-329 © Elsevier/North-HollandBiomedicalPress
321
SUBCELLULAR DISTRIBUTION OF BRAIN PEPTIDASES DEGRADING LUTEINIZING HORMONE RELEASING HORMONE (LHRH) AND THYROTROPIN RELEASING HORMONE (TRH)
P. JOSEPH-BRAVO*,C. LOUDES,J. L. CHARLIand C. KORDON Unitd 159 de Neuroendocrinologie, Centre Paul Broca de I'INSERM, 75014 Paris (France)
(Accepted August 17th, 1978)
SUMMARY The luteinizing hormone and thyrotropin-releasing hormones have been shown to be mostly concentrated in the nerve endings of the median eminence. In contrast, the peptidases reponsible for their degradation present an ubiquitous localization and most of the reports have dealt with the total soluble activity. A detailed study of the subcellular distribution of these enzymes was thus performed in cerebral cortical and hypothalamic preparations of rat brain. The activity of a soluble marker, lactic dehydrogenase, was also measured to control for possible contaminants. The results showed that only 10~ of peptidase activity was present in the nerve ending preparation. Evidence is provided for a non-negligible membrane-bound enzymatic component responsible for TRH degradation at the synaptosomal level of both cortex and hypothalamus. INTRODUCTION All structures of the central nervous system tested so far have been shown to contain neuropeptide-degrading enzymes9,1°,1~,16. The possibility that these enzymes may participate in the physiological regulation of brain neuropeptides has been proposed by some authors 5,11,15,16,19 but this still remains controversial7,17. In order to provide further information relevant to this question, we undertook a detailed study of the subcellular distribution of thyrotropin releasing hormone (TRH) and luteinizing hormone-releasing hormone (LHRH) degradation activity in rat hypothalamus and cerebral cortex. This seemed particularly appropriate since these peptides are predominantly localized in nerve endings13-15,20,22 and have been hypothesized to act as neuromodulators or neurotransmittersS; i.e. to exert synaptic actions. Enzyme * Research Fellow,UniversidadNacionalAut6nomade M6xico.
322 activity was evaluated under controlled kinetic conditions, as previously described 17, by measuring the disappearance of immunoreactive TRH and L H R H as an index of the reaction rate. The study was carried out on cerebral cortex and hypothalamus since L H R H is virtually restricted to the hypothalamus, while TRH is located in appreciable amounts in both brain regions. Our results indicate that around 10 ~ of the peptidase activity is contained in the nerve ending preparation; within this fraction, a significant proportion of enzyme activity, especially that active against TRH, appeared to be associated in some way with synaptosomal membranes. MATERIALS AND METHODS Male Wistar rats (200-250 g) kept under normal lighting conditions and fed ad libitum were used in all experiments. The animals were decapitated and the tissues (cerebral cortex and hypothalamus) excised and homogenized in cold 0.32 M sucrose. Synaptosomes were prepared following the method previously described 6. After centrifugation of the discontinuous sucrose gradient, band A myelin was recentrifuged at lC0,000 × g and the resulting pellet taken up in 0.05 M Tris, pH 7.2, 1 g/ml of original tissue wet weight, sonicated and diluted with the same buffer to 0.75, 0.5 and 0.2 g/ml. Band B (synaptosomes; 1.0 M sucrose interface) was adjusted to 0.45 M sucrose by addition of cold water and centrifuged at 35,000 x g in order to obtain the synaptosomes in a pellet form which was then treated in the same way as the pellet of band A. The mitochondrial pellet C was treated similarly. Soluble peptidase activity was measured in the $2 fraction at 0.005, 0.01, 0.015 and 0.02 g/ml wet weight tissue. The activity of lactic acid dehydrogenase (LDH) was concomitantly measured 13 in each fraction in the absence or presence of Triton (10 ~ final concentration, v/v). In subsequent experiments, the synaptosomal fraction was further separated into its membrane and soluble components. For this, band B was adjusted to 0.45 M sucrose and the suspension was separated into two aliquots and centrifuged at 35,000 x g. The resulting pellets served for the determination of soluble (a) and membrane (b) peptidase activity as follows: pellet (a) was resuspended in 0.05 M Tris, sonicated twice for 10 sec and centrifuged at 75,000 x g for 30 rain. The peptidase activity was measured in supernatants diluted with 0.05 M Tris, pH 7.2, to give 0.75, 0.50 and 0.20 g/ml; pellet (b) served as membrane-enzyme source; it was resuspended in 30 ml 0.01 M Tris, pH 7.2, sonicated for 10 sec, and centrifuged at 100,000 x g for 30 min. The activity of LDH was determined and the procedure of washing repeated twice. Negligible LDH activity was present in this fraction after this treatment. Only high concentrations of Triton were able to reveal any activity of this cytosolic marker in this fraction, which may correspond to cytosolic enzymes trapped within the membrane; but, since this activity was negligible without the detergent treatment and was not modified after the second osmotic shock and sonication, the preparation was used as such for measuring peptidase activity. For the determination of enzymatic activity, 50 #1 of enzyme source were preincubated for 20 min with the concentrations of L H R H and T R H equivalent to 3 x Kin: 9 × 10-5 M for L H R H and 5 x 10-~ for TRH. Ten #1 of the mixture were
323 taken out at the beginning and the end of the incubation times (20 min) for every s a m p l e , and added to 500 #1 o f 0.1 N HC1. L H R H and T R H concentrations were determined by radioimmunoassays ( R I A ) in which the antibody was proved not to crossreact with any of the proposed metabolites14, z0. T R H antibody was raised 14 in New Zealand rabbits against T R H - B D B - t h y r o g l o b u l i n conjugate and used at titre o f 1/25000. RIA-buffer consisted o f 0.05 M phosphate, 0.25 ~ BSA. Peptidase activity was calculated by the disappearance rate of i m m u n o r e a c t i v e - L H R H and i m m u n o reactive-TRH, hereafter designated as I R A - T R H and I R A - L H R H . RESULTS AND DISCUSSION Since the present study concerned the detection of peptidases acting on L H R H and T R H in synaptosomes and a correlation of these activities with the intracellular localization of the peptide h o r m o n e substrates, it was important to demonstrate the integrity of our synaptosomal preparations. The recovery of I R A - L H R H and I R A T R H , as well as that of the cytosol marker lactic dehydrogenase (LDH), c o m p a r e d well with that reported by other authors1-3,20,22; the average yield was 8 2 ~ for the low speed (900 × g) supernatant Sl, from which 96 ~ was recovered after further fractionation into a microsomal supernatant and a crude mitochondrial fraction. Separation on a discontinuous gradient showed that 90 ~ of I R A - L H R H and 67 ~ of I R A - T R H migrated along with the band B which contains the synaptosomal material (Fig. 1). In contrast to the neuropeptides, the highest proportion of degradative
100
5O
ii i
i
i
II
.
.
-
P1
S1
$2
P2
A
B
C
Fig. 1. Subcellular distribution of LHRH, TRH, LDH and proteins in rat hypothalamus. After the sucrose gradient an aliquot of each band was acidified to a concentration of 0.1 N with HCI and the concentration of releasing factors determined with radioimmunoassays for LHRH 2° and TRH 13 in triplicate. The values are expressed as per cent of the concentration of the 10~o homogenate. One hypothalamus: 4500 :~ 200 pg LHRH ; 5220 ± 350 pg TRH ; LDH 21: 62/tmol NA DH2 oxidized/h/rag protein;4.4 mg protein, mlLHRH; [ ] TRH; I~ LDH; I~ protein.
ii i
~
i i
~i
~ii
!iI
324
1
LDH
@l
10
TRH
5
l
NI i0
LHRH
5
A
B
C
Fig. 2. Enzymatic activity in subcellular fractions of rat cortex (full bars) and hypothalamus (empty bars). TRH and LHRH peptidase and LDH activities were calculated as per cent of the activity present in the Sz; for LHRH degradation: 261 ± 15 ng/min/mg wet weight (cortex) or 299 :~ 27 ng/min/mg (hypothalamus); for TRH degradation: 15 ± 0.5 ng/min/mg in both structures. Enzymatic activity was determined at 4 different enzyme concentrations (each concentration consisted of 5 experimental tubes assayed in duplicate by RIA). LDH activity was measured in triplicate, in presence of Triton (10 ~): 1.24/~mol/h/mg w.w. (cortex); 1.0 b~mol/h/mgw.w. (hypothalamus). activity seemed to be cytoplasmic and was recovered from the $2 supernatant with only 10 ~ found in the B fraction (Fig. 2). On the discontinuous gradient, most of the remaining L H R H and T R H degrading enzymes are recovered from band B. This distribution correlates well with that of L D H , which could suggest that the activity present in P2 is in fact contained within synaptosomes. There is no difference in the subcellular distribution of peptidases sampled from hypothalamic or cortical tissue (Fig. 2, Table I). In both brain regions, the peptidase activity found in the synaptosome preparation represents about 7-12 ~ of that measured in S~. In order to determine whether the T R H and L H R H degrading activities were soluble or membrane-bound, the synaptosomes were subsequently disrupted and separated into a membrane and a soluble fraction. As mentioned in Materials and Methods, the L D H activity recovered from the membrane fraction was very low, showing that it is relatively devoid of synaptoplasmic contaminants under the conditions where peptidase activity was measured. After a Triton solubilization of the membrane an increased activity of L D H was detected but only in fractions A and B
325 TABLE I
Subcellular distribution of LHRH and TRH peptidases in cortex and hypothalamus For L D H determination Triton was added to make a final concentration of 10 %. Peptidase activity was determined in a 4-point assay ($2:0.0054).02 g/ml wet weight tissue). L H R H and T R H (3 Kin) incubated with 50 #1 of enzyme source and the concentration determined by radioimmunoassay. Values are the mean 4- standard error.
LDH (#mol NADH2/h/mg prot.) Cortex Hypothalamus +Triton--Triton +Triton--Triton S~ A B C
35.5 12.0 24.5 11.8
35.5 2.42 12.1 10.2
26.7 10.7 20.4 7.8
27.2 3.5 10.5 7.6
TRH (spec. act. ng/ min/mg prot.)
LHRH* (spec. act. ng/min/mg prot.) Cortex
6411 d:417 749-4-153 11564-65 10104-121
Hypothalamus 64764-178 7754-124 14004-87 1453 4- 199
Cortex
Hypothalamus
383 4-50 3 6 6 4 - 4 0 63 -t-14 654-4 1184- 11 121 4-23 864-10 61 4-23
* Degrading activity; A (myelin); B (synaptosomes) and C (mitochondria) measured at 0.24).75 g/ml wet weight.
TABLE II
Distribution of LHRH and TRH peptidases in the membrane and soluble components of the nerve ending preparation Enzymatic activity was expressed per mg of protein. The synaptosomal preparation (B band) was separated in 2 aliquots and centrifuged at 35,000 × g. The pellet served for the determination of soluble (a) and membrane (b) peptidase activity. Pellets (a) were resuspended in Tris 0.05 M sonicated twice for 10 sec and centrifuged at 75,000 x g. The peptidase activity (ng degraded product/min/mg prot.) was measured in supernatants diluted to give 0.75, 0.5, 0.2 g/ml. Pellets (b) served as membrane-enzyme source: they were resuspended in 30 ml Tris 0.01 M, sonicated for 10 sec and centrifuged at 100,000 x g. The Sz (15,000 x g supernatant) was diluted with 0.05 M Tris (pH 7.2) to 0.0054).02 g/ml. L D H activity, ~ m o l N A D H / h / m g prot.) measured with Triton (10 %).
Sz (15,000 × g supernatant spec~[ic activity)
Protein content LDH --Triton 27 +Triton 33 L H R H peptidase activity 6322 :[: 284 T R H peptidase activity 350 4- 10
Synaptosomal fractions Membrane
Soluble
% activity* found Specific in intact B band activity (per mg wet weight tissue)
% B* Specific (mg w.w.) activity
65 4- 5
19 4- 4
1.6 10 ± 3 28 4- 3 60 4- 9
0.27 7.6 411 4- 40 73 i 12
73 :L 4 75 i 5 64 -4- 10 27:1:4
30 33 5952 4- 159 343:1:100
* The results in per cent are calculated with regard to the activity in mg wet weight measured in the non-separated fraction B.
326 which are rich in vesicle-like entities. Band C, which corresponds to the mitochondrial pellet, did not show any increased L D H activity by Triton, ruling out a direct effect of the detergent on these organelles at the synaptosomal level, nor did it affect the soluble activity of $2. It thus seemed likely that this increase in activity was due to a rupture of the vesicles that had resisted the ultrasonic and hypoosmotic treatment. Since the activity of this soluble enzyme was undetected before treatment with the detergent it was assumed that the soluble component of the peptidases was trapped in a similar way. No Triton was added for the determination of peptidase activity due to interference with RIA and to avoid solubilization of this possible membrane component. As shown in Table II, the L H R H and T R H specific activities are increased in the soluble portion of the band B relative to the unseparated material (Table I). Although the L H R H specific activity is reduced 3-fold in the particulate fraction (indicative that this is a soluble enzyme) there is only a small decrease in the specific activity of the T R H degrading enzyme. Furthermore, 6 0 ~ of the B fraction L H R H degrading activity is in the soluble component while only 30 ~ of the T R H activity is present (Fig. 3, Table I1). These results suggest that L H R H degrading activity is totally soluble while there is a significant portion of the T R H degrading activity present in the membranes. The existence of at least 3 enzymes responsible for T R H degradation has been reported4,19, While the enzymology of T R H inactivation is not yet well worked out, at the present time there is no evidence which indicates that any of these activities are non-specific peptidases. In fact, a recent study (Hersh and McKelvy, 1978" submitted for publication) indicates that T R H deamidation in brain is catalyzed by a post-proline cleaving enzyme similar to that described in kidney by Walter 21 and which inactivates vasopressin. This enzyme is distinct from previously known peptidases and exhibits stringent substrate structural requirements for its action. While a given peptidase may act on more than one hormone, this may reflect a 'specific' mechanism in that these enzymes act only on hormonal peptides at residues which are critical for peptide receptor interactions. Moreover, the existence of multiple peptidases acting on a peptide such as T R H may be a reflection of more than one metabolic role for the peptide, e.g. T R H actions on both TSH and prolactin release, and as a neuromodulator. Thus, in the present study, it was of importance to initially follow total T R H degradation, which was most convenient by RIA, since crossreaction of T R H metabolites with the antibody is virtually non-existent. It is possible that one or more of the above mentioned T R H degrading activities, i.e., the pGlul-His 2, the His2-Pro 3 or the Pro3-NH2 peptidases, constitute the enzyme activity which we find associated with the synaptosomal membrane fraction. Further studies, in which direct assays for these enzymes can be carried out on membrane preparations, before and after Triton or other solubilizing agents, and on membrane preparations obtained from nerve endings, are needed to resolve this issue and the question of whether such an activity can influence T R H release. The lack of correlation between the distribution of soluble hypothalamic peptidases and the releasing hormones raises the question as to whether or not these
327
PROT.
10
D
LDH
10
TRH
10
LHRH
m M
A
B
S
I
C
Fig. 3. Distribution of protein, LDH activity, L H R H and TRH peptidase activity in myelin (A), synaptosomes (B), membrane (M) and soluble (S) components of synaptosomes, and mitochondria (C). Results were expressed as per cent of the activity found for Se. L H R H peptidase activity: 299 -4- 43 ng/min/mg wet weight tissue; T R H : 13.9 + 0.1 ng/min/mg wet weight tissue, LDH activity: 1.47 /~mol/h/mg wet weight tissue; protein: 43.5/~g prot./mg wet weight tissue.
peptidases are intimately involved in a synaptic action of releasing hormones. The equal distribution in cortex and hypothalamic nerve ending preparations suggests that these peptidases, although not regulating hormone levels at the hypothetical site of release, could serve to regulate the overall tissue levels of these peptide hormones. Furthermore, the similarity between specific activity values of bands B and C (Table I) could indicate the presence of these enzymes in lysosomes. Careful further study, using techniques oriented to obtain intact lysosomes, could give some insight into the intracellular localization of these enzymes, keeping in mind that homogenization techniques destroy the subcellular localization to a considerable extent. ACKNOWLEDGEMENTS
The authors wish to acknowledge Drs. J. McKelvy and L. Hersh for their
328 advice in correcting the m a n u s c r i p t a n d M. C. S i m o n a n d B. Lewis for skillful secretarial assistance. P. Joseph Bravo acknowledges U N A M a n d Banco de Mexico for the scholarship awarded.
REFERENCES 1 Barnea, A., Ben-Jonathan, N., Colston, C., Johnston, J. M. and Porter, J. C., Differential subcellular compartmentalization of thyrotropin releasing hormone (TRH) and gonadotropin releasing hormone (LRH) in hypothalamic tissue, Proc. nat. Acad. Sci. (Wash.), 72 (1975) 3153-3157. 2 Barnea, A., Ben-Jonathan, N. and Porter, J. C., Characterization of hypothalamic subcellular particles containing luteinizing hormone-releasing hormone and thyrotropin releasing hormone, J. Neurochem., 27 (1976) 477~,84. 3 Barnea, A., Neaves, W. B. and Porter, J. C., Ontogeny of the subcellular compartmentalization of TRH and LHRH in the rat hypothalamus, Endocrinology, 100 (1977) 1068-1079. 4 Bauer, K., Degradation of TRH. Its inhibition by pyroglu-his-OCH8 and the effect of the inhibitor in attempts to study the biosynthesis of TRH. In Lipmann Symposium: Energy, Biosynthesis and Regulation in Molecular Biology, De Gruyter, Berlin, 1974, 53 pp. 5 Bauer, K., Regulation of degradation of TRH by thyroid hormones, Nature (Lond.), 259 (1976) 591-593. 6 Bradford, H. F., Cerebral cortex slices and synaptosomes. In vitro approaches to brain metabolism. In R. L. Fried (Ed.), Methods of Neurochemistry, Vol. 3, Marcel Dekker, New York, 1972, p. 155. 7 Clayton, R. N., Shakespear, R. A. and Marsha, J. C., Effect oftestosterone and oestradiol on LHRH degradation by purified pituitary plasma membranes, Acta endocrinol. (Basel), 85, Suppl. 59 (1977) 85. 8 Edwardson, J. A. and Bennet, G. W., Hypothalamic hormones and mechanisms ofneuroendocrine integration. In D. A. Ames (Ed.), Biologically Active Substances: Exploration and Exploitation, 1977, p. 281. 9 Griffiths, E. C., Hooper, K. C. and Hopkinson, C. R. N., Further studies on enzymic inactivation of LHRH by peptidases in the rat hypothalamus, Acta endocrinoL (Basel), 79 (1975) 7-15. 10 Griffiths, E. C., Hooper, K. C., Jeffcoate, S. L. and Holland, D. T., Peptidases in different areas of the rat brain inactivatingluteinizinghormone releasing hormone, Brain Research, 85 (1975) 161-164. 11 Griffiths, E. C., Hooper, K. C., Jeffcoate, S. L. and Holland, D. T., The effects of gonadectomy and gonadal steroids on the activity of hypothalamic peptidases inactivating luteinizing hormone releasing hormone (LHRH), Brain Research, 88 (1975) 384--388. 12 Griftiths, E. C., Hooper, K. C., Jeffcoate, S. L. and White, N., Peptidases in the rat hypothalamus inactivating TRH, Acta endocrinol. (Basel), 79 (1975) 209-216. 13 Johnson, N. K. and Whittaker, V. D., Lactate dehydrogenase as a cytoplasmic marker in brain, Biochem. J., 88 (1963) 404. 14 Joseph-Bravo, P., Charli, J. L., Palacios, J. M. and Kordon, C., Effect ofneurotransmitters on the in vitro release of immunoreactivethyrotropin releasing hormone from rat mediobasal hypothalamus, Endocrinology, in press. 15 Kuhl, H., Rosniatowshi, C. and Taubert, H. D., The regulatory function of a pituitary LHRH degrading enzyme system in the feedback control of gonadotropins, Acta endocrinoL (Basel), 86 (1977) 60-70. 16 Kuhl, H. and Taubert, H. D., Short-loop feedback mechanism of luteinizinghormone: LH stimulates hypothalamic L-cysteine arylamidase to inactive LHRH in the rat hypothalamus, Acta endocrinol. (Basel), 78 (1975) 649-663. 17 Loudes, C., Joseph-Bravo, P., Leblanc, P. and Kordon, C., Specific activity of LHRH and TRH degradingenzymes in various tissues of normal castrated male rats, Biochem. Biophys. res. Commun., in press. 18 Marks, N., Biodegradation of hormonally active peptides in the central nervous system. In N. Naftolin, K. J. Ryand and J. Davies (Eds.), Subcellular mechanisms in reproductive neuroendocrinology, 1976, p. 129-147. 19 Prasad, C. and Peterkovsky, A., Demonstration of pyroglytamyl peptidase and amidase activities toward TRH in hamster hypothalamus extracts, J. biol. Chem., 251 (1976) 3229-3234.
329 20 Ramirez, V. D., Gautron, J. P., Epelbaum, J., Pattou, E., Zamora, A. and Kordon, C., Distribution of LHRH in subcellular fractions of the basomedial hypothalamus, Molec. Cell Endocr., 3 (1975) 339-350. 21 Walter, R., Partial purification and characterization of post-proline cleaving enzyme: enzymatic inactivation of neurohypophyseal hormones by kidney preparations of various species, Biochim. biophys. Acta ( Amst.) , 402 (1976) 138-158. 22 Winokur, A., Davis, R. and Utiger, R. D., Subcellular distribution of TRH in rat brain and hypothalamus, Brain Research, 120 (1977) 423-434.
321
SUBCELLULAR DISTRIBUTION OF BRAIN PEPTIDASES DEGRADING LUTEINIZING HORMONE RELEASING HORMONE (LHRH) AND THYROTROPIN RELEASING HORMONE (TRH)
P. JOSEPH-BRAVO*,C. LOUDES,J. L. CHARLIand C. KORDON Unitd 159 de Neuroendocrinologie, Centre Paul Broca de I'INSERM, 75014 Paris (France)
(Accepted August 17th, 1978)
SUMMARY The luteinizing hormone and thyrotropin-releasing hormones have been shown to be mostly concentrated in the nerve endings of the median eminence. In contrast, the peptidases reponsible for their degradation present an ubiquitous localization and most of the reports have dealt with the total soluble activity. A detailed study of the subcellular distribution of these enzymes was thus performed in cerebral cortical and hypothalamic preparations of rat brain. The activity of a soluble marker, lactic dehydrogenase, was also measured to control for possible contaminants. The results showed that only 10~ of peptidase activity was present in the nerve ending preparation. Evidence is provided for a non-negligible membrane-bound enzymatic component responsible for TRH degradation at the synaptosomal level of both cortex and hypothalamus. INTRODUCTION All structures of the central nervous system tested so far have been shown to contain neuropeptide-degrading enzymes9,1°,1~,16. The possibility that these enzymes may participate in the physiological regulation of brain neuropeptides has been proposed by some authors 5,11,15,16,19 but this still remains controversial7,17. In order to provide further information relevant to this question, we undertook a detailed study of the subcellular distribution of thyrotropin releasing hormone (TRH) and luteinizing hormone-releasing hormone (LHRH) degradation activity in rat hypothalamus and cerebral cortex. This seemed particularly appropriate since these peptides are predominantly localized in nerve endings13-15,20,22 and have been hypothesized to act as neuromodulators or neurotransmittersS; i.e. to exert synaptic actions. Enzyme * Research Fellow,UniversidadNacionalAut6nomade M6xico.
322 activity was evaluated under controlled kinetic conditions, as previously described 17, by measuring the disappearance of immunoreactive TRH and L H R H as an index of the reaction rate. The study was carried out on cerebral cortex and hypothalamus since L H R H is virtually restricted to the hypothalamus, while TRH is located in appreciable amounts in both brain regions. Our results indicate that around 10 ~ of the peptidase activity is contained in the nerve ending preparation; within this fraction, a significant proportion of enzyme activity, especially that active against TRH, appeared to be associated in some way with synaptosomal membranes. MATERIALS AND METHODS Male Wistar rats (200-250 g) kept under normal lighting conditions and fed ad libitum were used in all experiments. The animals were decapitated and the tissues (cerebral cortex and hypothalamus) excised and homogenized in cold 0.32 M sucrose. Synaptosomes were prepared following the method previously described 6. After centrifugation of the discontinuous sucrose gradient, band A myelin was recentrifuged at lC0,000 × g and the resulting pellet taken up in 0.05 M Tris, pH 7.2, 1 g/ml of original tissue wet weight, sonicated and diluted with the same buffer to 0.75, 0.5 and 0.2 g/ml. Band B (synaptosomes; 1.0 M sucrose interface) was adjusted to 0.45 M sucrose by addition of cold water and centrifuged at 35,000 x g in order to obtain the synaptosomes in a pellet form which was then treated in the same way as the pellet of band A. The mitochondrial pellet C was treated similarly. Soluble peptidase activity was measured in the $2 fraction at 0.005, 0.01, 0.015 and 0.02 g/ml wet weight tissue. The activity of lactic acid dehydrogenase (LDH) was concomitantly measured 13 in each fraction in the absence or presence of Triton (10 ~ final concentration, v/v). In subsequent experiments, the synaptosomal fraction was further separated into its membrane and soluble components. For this, band B was adjusted to 0.45 M sucrose and the suspension was separated into two aliquots and centrifuged at 35,000 x g. The resulting pellets served for the determination of soluble (a) and membrane (b) peptidase activity as follows: pellet (a) was resuspended in 0.05 M Tris, sonicated twice for 10 sec and centrifuged at 75,000 x g for 30 rain. The peptidase activity was measured in supernatants diluted with 0.05 M Tris, pH 7.2, to give 0.75, 0.50 and 0.20 g/ml; pellet (b) served as membrane-enzyme source; it was resuspended in 30 ml 0.01 M Tris, pH 7.2, sonicated for 10 sec, and centrifuged at 100,000 x g for 30 min. The activity of LDH was determined and the procedure of washing repeated twice. Negligible LDH activity was present in this fraction after this treatment. Only high concentrations of Triton were able to reveal any activity of this cytosolic marker in this fraction, which may correspond to cytosolic enzymes trapped within the membrane; but, since this activity was negligible without the detergent treatment and was not modified after the second osmotic shock and sonication, the preparation was used as such for measuring peptidase activity. For the determination of enzymatic activity, 50 #1 of enzyme source were preincubated for 20 min with the concentrations of L H R H and T R H equivalent to 3 x Kin: 9 × 10-5 M for L H R H and 5 x 10-~ for TRH. Ten #1 of the mixture were
323 taken out at the beginning and the end of the incubation times (20 min) for every s a m p l e , and added to 500 #1 o f 0.1 N HC1. L H R H and T R H concentrations were determined by radioimmunoassays ( R I A ) in which the antibody was proved not to crossreact with any of the proposed metabolites14, z0. T R H antibody was raised 14 in New Zealand rabbits against T R H - B D B - t h y r o g l o b u l i n conjugate and used at titre o f 1/25000. RIA-buffer consisted o f 0.05 M phosphate, 0.25 ~ BSA. Peptidase activity was calculated by the disappearance rate of i m m u n o r e a c t i v e - L H R H and i m m u n o reactive-TRH, hereafter designated as I R A - T R H and I R A - L H R H . RESULTS AND DISCUSSION Since the present study concerned the detection of peptidases acting on L H R H and T R H in synaptosomes and a correlation of these activities with the intracellular localization of the peptide h o r m o n e substrates, it was important to demonstrate the integrity of our synaptosomal preparations. The recovery of I R A - L H R H and I R A T R H , as well as that of the cytosol marker lactic dehydrogenase (LDH), c o m p a r e d well with that reported by other authors1-3,20,22; the average yield was 8 2 ~ for the low speed (900 × g) supernatant Sl, from which 96 ~ was recovered after further fractionation into a microsomal supernatant and a crude mitochondrial fraction. Separation on a discontinuous gradient showed that 90 ~ of I R A - L H R H and 67 ~ of I R A - T R H migrated along with the band B which contains the synaptosomal material (Fig. 1). In contrast to the neuropeptides, the highest proportion of degradative
100
5O
ii i
i
i
II
.
.
-
P1
S1
$2
P2
A
B
C
Fig. 1. Subcellular distribution of LHRH, TRH, LDH and proteins in rat hypothalamus. After the sucrose gradient an aliquot of each band was acidified to a concentration of 0.1 N with HCI and the concentration of releasing factors determined with radioimmunoassays for LHRH 2° and TRH 13 in triplicate. The values are expressed as per cent of the concentration of the 10~o homogenate. One hypothalamus: 4500 :~ 200 pg LHRH ; 5220 ± 350 pg TRH ; LDH 21: 62/tmol NA DH2 oxidized/h/rag protein;4.4 mg protein, mlLHRH; [ ] TRH; I~ LDH; I~ protein.
ii i
~
i i
~i
~ii
!iI
324
1
LDH
@l
10
TRH
5
l
NI i0
LHRH
5
A
B
C
Fig. 2. Enzymatic activity in subcellular fractions of rat cortex (full bars) and hypothalamus (empty bars). TRH and LHRH peptidase and LDH activities were calculated as per cent of the activity present in the Sz; for LHRH degradation: 261 ± 15 ng/min/mg wet weight (cortex) or 299 :~ 27 ng/min/mg (hypothalamus); for TRH degradation: 15 ± 0.5 ng/min/mg in both structures. Enzymatic activity was determined at 4 different enzyme concentrations (each concentration consisted of 5 experimental tubes assayed in duplicate by RIA). LDH activity was measured in triplicate, in presence of Triton (10 ~): 1.24/~mol/h/mg w.w. (cortex); 1.0 b~mol/h/mgw.w. (hypothalamus). activity seemed to be cytoplasmic and was recovered from the $2 supernatant with only 10 ~ found in the B fraction (Fig. 2). On the discontinuous gradient, most of the remaining L H R H and T R H degrading enzymes are recovered from band B. This distribution correlates well with that of L D H , which could suggest that the activity present in P2 is in fact contained within synaptosomes. There is no difference in the subcellular distribution of peptidases sampled from hypothalamic or cortical tissue (Fig. 2, Table I). In both brain regions, the peptidase activity found in the synaptosome preparation represents about 7-12 ~ of that measured in S~. In order to determine whether the T R H and L H R H degrading activities were soluble or membrane-bound, the synaptosomes were subsequently disrupted and separated into a membrane and a soluble fraction. As mentioned in Materials and Methods, the L D H activity recovered from the membrane fraction was very low, showing that it is relatively devoid of synaptoplasmic contaminants under the conditions where peptidase activity was measured. After a Triton solubilization of the membrane an increased activity of L D H was detected but only in fractions A and B
325 TABLE I
Subcellular distribution of LHRH and TRH peptidases in cortex and hypothalamus For L D H determination Triton was added to make a final concentration of 10 %. Peptidase activity was determined in a 4-point assay ($2:0.0054).02 g/ml wet weight tissue). L H R H and T R H (3 Kin) incubated with 50 #1 of enzyme source and the concentration determined by radioimmunoassay. Values are the mean 4- standard error.
LDH (#mol NADH2/h/mg prot.) Cortex Hypothalamus +Triton--Triton +Triton--Triton S~ A B C
35.5 12.0 24.5 11.8
35.5 2.42 12.1 10.2
26.7 10.7 20.4 7.8
27.2 3.5 10.5 7.6
TRH (spec. act. ng/ min/mg prot.)
LHRH* (spec. act. ng/min/mg prot.) Cortex
6411 d:417 749-4-153 11564-65 10104-121
Hypothalamus 64764-178 7754-124 14004-87 1453 4- 199
Cortex
Hypothalamus
383 4-50 3 6 6 4 - 4 0 63 -t-14 654-4 1184- 11 121 4-23 864-10 61 4-23
* Degrading activity; A (myelin); B (synaptosomes) and C (mitochondria) measured at 0.24).75 g/ml wet weight.
TABLE II
Distribution of LHRH and TRH peptidases in the membrane and soluble components of the nerve ending preparation Enzymatic activity was expressed per mg of protein. The synaptosomal preparation (B band) was separated in 2 aliquots and centrifuged at 35,000 × g. The pellet served for the determination of soluble (a) and membrane (b) peptidase activity. Pellets (a) were resuspended in Tris 0.05 M sonicated twice for 10 sec and centrifuged at 75,000 x g. The peptidase activity (ng degraded product/min/mg prot.) was measured in supernatants diluted to give 0.75, 0.5, 0.2 g/ml. Pellets (b) served as membrane-enzyme source: they were resuspended in 30 ml Tris 0.01 M, sonicated for 10 sec and centrifuged at 100,000 x g. The Sz (15,000 x g supernatant) was diluted with 0.05 M Tris (pH 7.2) to 0.0054).02 g/ml. L D H activity, ~ m o l N A D H / h / m g prot.) measured with Triton (10 %).
Sz (15,000 × g supernatant spec~[ic activity)
Protein content LDH --Triton 27 +Triton 33 L H R H peptidase activity 6322 :[: 284 T R H peptidase activity 350 4- 10
Synaptosomal fractions Membrane
Soluble
% activity* found Specific in intact B band activity (per mg wet weight tissue)
% B* Specific (mg w.w.) activity
65 4- 5
19 4- 4
1.6 10 ± 3 28 4- 3 60 4- 9
0.27 7.6 411 4- 40 73 i 12
73 :L 4 75 i 5 64 -4- 10 27:1:4
30 33 5952 4- 159 343:1:100
* The results in per cent are calculated with regard to the activity in mg wet weight measured in the non-separated fraction B.
326 which are rich in vesicle-like entities. Band C, which corresponds to the mitochondrial pellet, did not show any increased L D H activity by Triton, ruling out a direct effect of the detergent on these organelles at the synaptosomal level, nor did it affect the soluble activity of $2. It thus seemed likely that this increase in activity was due to a rupture of the vesicles that had resisted the ultrasonic and hypoosmotic treatment. Since the activity of this soluble enzyme was undetected before treatment with the detergent it was assumed that the soluble component of the peptidases was trapped in a similar way. No Triton was added for the determination of peptidase activity due to interference with RIA and to avoid solubilization of this possible membrane component. As shown in Table II, the L H R H and T R H specific activities are increased in the soluble portion of the band B relative to the unseparated material (Table I). Although the L H R H specific activity is reduced 3-fold in the particulate fraction (indicative that this is a soluble enzyme) there is only a small decrease in the specific activity of the T R H degrading enzyme. Furthermore, 6 0 ~ of the B fraction L H R H degrading activity is in the soluble component while only 30 ~ of the T R H activity is present (Fig. 3, Table I1). These results suggest that L H R H degrading activity is totally soluble while there is a significant portion of the T R H degrading activity present in the membranes. The existence of at least 3 enzymes responsible for T R H degradation has been reported4,19, While the enzymology of T R H inactivation is not yet well worked out, at the present time there is no evidence which indicates that any of these activities are non-specific peptidases. In fact, a recent study (Hersh and McKelvy, 1978" submitted for publication) indicates that T R H deamidation in brain is catalyzed by a post-proline cleaving enzyme similar to that described in kidney by Walter 21 and which inactivates vasopressin. This enzyme is distinct from previously known peptidases and exhibits stringent substrate structural requirements for its action. While a given peptidase may act on more than one hormone, this may reflect a 'specific' mechanism in that these enzymes act only on hormonal peptides at residues which are critical for peptide receptor interactions. Moreover, the existence of multiple peptidases acting on a peptide such as T R H may be a reflection of more than one metabolic role for the peptide, e.g. T R H actions on both TSH and prolactin release, and as a neuromodulator. Thus, in the present study, it was of importance to initially follow total T R H degradation, which was most convenient by RIA, since crossreaction of T R H metabolites with the antibody is virtually non-existent. It is possible that one or more of the above mentioned T R H degrading activities, i.e., the pGlul-His 2, the His2-Pro 3 or the Pro3-NH2 peptidases, constitute the enzyme activity which we find associated with the synaptosomal membrane fraction. Further studies, in which direct assays for these enzymes can be carried out on membrane preparations, before and after Triton or other solubilizing agents, and on membrane preparations obtained from nerve endings, are needed to resolve this issue and the question of whether such an activity can influence T R H release. The lack of correlation between the distribution of soluble hypothalamic peptidases and the releasing hormones raises the question as to whether or not these
327
PROT.
10
D
LDH
10
TRH
10
LHRH
m M
A
B
S
I
C
Fig. 3. Distribution of protein, LDH activity, L H R H and TRH peptidase activity in myelin (A), synaptosomes (B), membrane (M) and soluble (S) components of synaptosomes, and mitochondria (C). Results were expressed as per cent of the activity found for Se. L H R H peptidase activity: 299 -4- 43 ng/min/mg wet weight tissue; T R H : 13.9 + 0.1 ng/min/mg wet weight tissue, LDH activity: 1.47 /~mol/h/mg wet weight tissue; protein: 43.5/~g prot./mg wet weight tissue.
peptidases are intimately involved in a synaptic action of releasing hormones. The equal distribution in cortex and hypothalamic nerve ending preparations suggests that these peptidases, although not regulating hormone levels at the hypothetical site of release, could serve to regulate the overall tissue levels of these peptide hormones. Furthermore, the similarity between specific activity values of bands B and C (Table I) could indicate the presence of these enzymes in lysosomes. Careful further study, using techniques oriented to obtain intact lysosomes, could give some insight into the intracellular localization of these enzymes, keeping in mind that homogenization techniques destroy the subcellular localization to a considerable extent. ACKNOWLEDGEMENTS
The authors wish to acknowledge Drs. J. McKelvy and L. Hersh for their
328 advice in correcting the m a n u s c r i p t a n d M. C. S i m o n a n d B. Lewis for skillful secretarial assistance. P. Joseph Bravo acknowledges U N A M a n d Banco de Mexico for the scholarship awarded.
REFERENCES 1 Barnea, A., Ben-Jonathan, N., Colston, C., Johnston, J. M. and Porter, J. C., Differential subcellular compartmentalization of thyrotropin releasing hormone (TRH) and gonadotropin releasing hormone (LRH) in hypothalamic tissue, Proc. nat. Acad. Sci. (Wash.), 72 (1975) 3153-3157. 2 Barnea, A., Ben-Jonathan, N. and Porter, J. C., Characterization of hypothalamic subcellular particles containing luteinizing hormone-releasing hormone and thyrotropin releasing hormone, J. Neurochem., 27 (1976) 477~,84. 3 Barnea, A., Neaves, W. B. and Porter, J. C., Ontogeny of the subcellular compartmentalization of TRH and LHRH in the rat hypothalamus, Endocrinology, 100 (1977) 1068-1079. 4 Bauer, K., Degradation of TRH. Its inhibition by pyroglu-his-OCH8 and the effect of the inhibitor in attempts to study the biosynthesis of TRH. In Lipmann Symposium: Energy, Biosynthesis and Regulation in Molecular Biology, De Gruyter, Berlin, 1974, 53 pp. 5 Bauer, K., Regulation of degradation of TRH by thyroid hormones, Nature (Lond.), 259 (1976) 591-593. 6 Bradford, H. F., Cerebral cortex slices and synaptosomes. In vitro approaches to brain metabolism. In R. L. Fried (Ed.), Methods of Neurochemistry, Vol. 3, Marcel Dekker, New York, 1972, p. 155. 7 Clayton, R. N., Shakespear, R. A. and Marsha, J. C., Effect oftestosterone and oestradiol on LHRH degradation by purified pituitary plasma membranes, Acta endocrinol. (Basel), 85, Suppl. 59 (1977) 85. 8 Edwardson, J. A. and Bennet, G. W., Hypothalamic hormones and mechanisms ofneuroendocrine integration. In D. A. Ames (Ed.), Biologically Active Substances: Exploration and Exploitation, 1977, p. 281. 9 Griffiths, E. C., Hooper, K. C. and Hopkinson, C. R. N., Further studies on enzymic inactivation of LHRH by peptidases in the rat hypothalamus, Acta endocrinoL (Basel), 79 (1975) 7-15. 10 Griffiths, E. C., Hooper, K. C., Jeffcoate, S. L. and Holland, D. T., Peptidases in different areas of the rat brain inactivatingluteinizinghormone releasing hormone, Brain Research, 85 (1975) 161-164. 11 Griffiths, E. C., Hooper, K. C., Jeffcoate, S. L. and Holland, D. T., The effects of gonadectomy and gonadal steroids on the activity of hypothalamic peptidases inactivating luteinizing hormone releasing hormone (LHRH), Brain Research, 88 (1975) 384--388. 12 Griftiths, E. C., Hooper, K. C., Jeffcoate, S. L. and White, N., Peptidases in the rat hypothalamus inactivating TRH, Acta endocrinol. (Basel), 79 (1975) 209-216. 13 Johnson, N. K. and Whittaker, V. D., Lactate dehydrogenase as a cytoplasmic marker in brain, Biochem. J., 88 (1963) 404. 14 Joseph-Bravo, P., Charli, J. L., Palacios, J. M. and Kordon, C., Effect ofneurotransmitters on the in vitro release of immunoreactivethyrotropin releasing hormone from rat mediobasal hypothalamus, Endocrinology, in press. 15 Kuhl, H., Rosniatowshi, C. and Taubert, H. D., The regulatory function of a pituitary LHRH degrading enzyme system in the feedback control of gonadotropins, Acta endocrinoL (Basel), 86 (1977) 60-70. 16 Kuhl, H. and Taubert, H. D., Short-loop feedback mechanism of luteinizinghormone: LH stimulates hypothalamic L-cysteine arylamidase to inactive LHRH in the rat hypothalamus, Acta endocrinol. (Basel), 78 (1975) 649-663. 17 Loudes, C., Joseph-Bravo, P., Leblanc, P. and Kordon, C., Specific activity of LHRH and TRH degradingenzymes in various tissues of normal castrated male rats, Biochem. Biophys. res. Commun., in press. 18 Marks, N., Biodegradation of hormonally active peptides in the central nervous system. In N. Naftolin, K. J. Ryand and J. Davies (Eds.), Subcellular mechanisms in reproductive neuroendocrinology, 1976, p. 129-147. 19 Prasad, C. and Peterkovsky, A., Demonstration of pyroglytamyl peptidase and amidase activities toward TRH in hamster hypothalamus extracts, J. biol. Chem., 251 (1976) 3229-3234.
329 20 Ramirez, V. D., Gautron, J. P., Epelbaum, J., Pattou, E., Zamora, A. and Kordon, C., Distribution of LHRH in subcellular fractions of the basomedial hypothalamus, Molec. Cell Endocr., 3 (1975) 339-350. 21 Walter, R., Partial purification and characterization of post-proline cleaving enzyme: enzymatic inactivation of neurohypophyseal hormones by kidney preparations of various species, Biochim. biophys. Acta ( Amst.) , 402 (1976) 138-158. 22 Winokur, A., Davis, R. and Utiger, R. D., Subcellular distribution of TRH in rat brain and hypothalamus, Brain Research, 120 (1977) 423-434.
