- Email: [email protected]
Development and characterization of a specific and sensitive radioimmunoassay for rat galanin: Measurement in brain tissue, hypophyseal portal and peripheral serum
Brain Research Bulletin,
Vol. 24, pp. 395-399. PergamonPress pk. 1990.Printedin the U.S.A.
0361-9230190 $3.00 + .Gu
Development and Characterization of a Specific and Sensitive Radioimmunoassay for Rat Galanin: Measurement in Brain Tissue, Hypophyseal Portal and Peripheral Serum FXANCISCO JO& L6PEZ, EDWIN H. MEADE, JR. AND AND&S NEGRO-Vet* Reproductive Neuroendocrinology Section, Laboratory of Molecular and Integrative Neuroscience National Institute of Environmental Health Sciences, Research Triangle Park, NC 27709
Received 7 November 1989 L6PEZ, F. J., E. H. MEADE, JR. AND A. NEGRO-VILAR. Development and characterization of a specific and sensitive radioimmunoassczyfor rat galanin: Measurement in brain tissue, hypophyseal portal and peripheral serum. BRAIN RES BULL 24(3) 395-399, 1990. -Gala& (GAL), a 29 amino acid peptide, is extensively distributed in both brain and intestine having been described as a putative neuroendocrine modulator. Up to this date, available radioi~unoas~ys for measuring GAL have the problem of nonp~alIe1 responses when rat biological samples are used. Due to the recent availabili~ of synthetic rat GAL (rGAL), we pursued the devefopment of a specific RIA for IGAL. Our RIA system presents a high specificity since porcine GAL (pGAL), which differs from ICAL in three carboxy-terminus substitutions, displays negligible crossreactivity. In addition, in the RIA a sensitivity of at least 1 pg/tube can be obtained. All biological samples tested in this study (serum samples from both peripheral and portal blood and arcuate-nucleus median eminence extracts) displayed perfect parallelism to synthetic rGAL standard. The RIA is suitable for measuring minute amounts of rGAL directly in serum samplesfrom both portal and peripheral circulation. Interestingly, the IC, for portal serum was approximately IO-fold lower than that for peripheral serum. This indicates that rGAL may be present in higher concentrations in portal than in peripheral blood. In addition, our data suggest that the p~mi~ gland contributes appro~ma~ly 30% to the total levels of rGAL measured in peripheral serum. Together, this data reinforces the concept that GAL is involved in mechanisms controlling and/or regulating neuroendocrine functions. Rat galanin RIA
Arcuate-ME
LHRH
in the arcuate nucleus and therefore impairs many endocrine functions (3,l l), induces a drastic reduction in GAL content in the median eminence (4). This data reinforces the concept that GAL may play an important role as a neuroendocrine regulator. In the studies dealing with GAL concentrations in the hypothalamus (4, 17, 18), a drastic displacement of parallelism to the standard curve was obtained when an antisera raised against pGAL was used (418). Porcine and rat GAL have a high degree of homology, however, they differ in three amino acids located in the carboxy-terminus portion of the molecule (6). This slight difference may explain the absence of parallelism to the standard curve observed in those studies. In this study we report the development and characterization of a sensitive and specific radioi~unoassay for rat GAL that allows the direct measurement of GAL in serum samples. In addition, our
GALANIN (GAL) is a 29 amino acid peptide of the family of brain gastrointestinal peptides. Originally, it was isolated from porcine intestine (19). Later, it was shown that GAL is widely distributed in rat brain in high concentrations, especially in the hypothalamus ( 14,17). Its location in the hypothalamus and other reports indicating that GAL may stimulate PRL, LH and GH secretion (7-9, 14, 16, 17) point out that GAL may be an irn~~~t mediator in neu~nd~~ne-related phenomena. Using a heterologous radioimmunoassay, it has been shown that in the rat, the highest concentrations of GAL immunoreactivity within the hypothalamus can be found in the median eminence (18). In general, GAL concentrations in the hypothalamus are higher than those observed in telencephalon, mesencephalon and rhombencephalon ( 18). In addition, neonatal treatment with monosodium glutamate, which destroys most of the neuronal networks
‘Requests for reprints should be addressed to Dr. Andrt% Negro-V&u, Reproductive Neuroendocrinology Section, Laboratory of Molecular and Integrative Neuroscience, N.I.E.H.S., Building 101, MD C4-09, Research Triangle Park, NC 27709.
395
396
L6PEZ,
data reveal that in male rats, the pituitary gland contributes approximately 30% to the total amount of rGAL in peripheral serum. Moreover, the study also indicates that rGAL might be present in higher concentrations in portal than in peripheral serum.
1:400000
MEADE AND NEGRO-VILAR
[ t
a
/-‘,
0
\,
i
i’
A
I
,’
Peptides Synthetic rGAL, pGAL, LHRH, oxytocin, rat CRF, TRH, dermorphin, AVP, rat VIP, SRIF 28, human ACTH, frog sauvagine, alpha-helical-CRF, beta-endorphin and the fragments of the galanin message-associated peptide (GMAP) 1-41, 16-41, 25-41 and 44-59 were obtained from Peninsula Laboratories Inc. (Belmont, CA). Peptides were routinely dissolved in 0.1 N acetic acid (1 mglml), divided in aliquots and kept frozen at - 80°C until used.
,,
Radioimmunoassay Synthetic rGAL was iodinated using the lactoperoxidase method (I) by Hazleton Labs. After an initial p~i~cation by HPLC, the tracer was repurified, prior to use in the radioimmunoassay, utilizing Sephadex G-50 fine. The buffer used for the RIA consisted of 0.025 M phosphate buffer containing 1.25% BSA, 0.5% Triton X-100 and 0.1% sodium azide. RIA was performed under disequilibrium conditions to obtain full sensitivity. One hundred )-LIof either standard or samples were incubated for 24 hr with 50 ~1 of ~tise~rn at room temperature. After this preliminary incubation, 50 ~1 of freshly repurified tracer (7000 cpm) were added and the incubation continued for an additional 24-hr period at 4°C. The antiserum (FJL No. 7-3) was utilized at a final dilution of 1:120,000. Bound GAL was separated by adding 50 ~1 of 4% NRS followed by the addition of 50 ~1 of 4:25 goat anti-rabbit gamma globulins (Anti~ies Inc., Davis, CA) and incubating the mixture 4 hr at room temperature. After the incubation was finished, 0.5 ml assay
i
&’
/
,i’
,
/ 1:400
1:40
,--
-*
/
“ FJL, FJL n FJL =---A FJL
O-- --0
o--O
1
.-. 3
I-
1
~~__i_.._i 3 ,1;
#7 #8 #9 #lO
5
BLEEDING +
Generation of Anti-rGAL Serum A conjugate of rGAL/bovine thyroglobulin (bTG, Sigma Chemical Co., St. Louis, MO) was used as immunogen. rGAL was coupled to bTG using I-ethyl-3-(3-dime~y~minop~pyl) carbodiimide (EDAC; Sigma). TGAL (1.4 mg) was diluted in 0.5 ml 0.05 M phosphate buffer (pH =7.5) and mixed with 0.75 mg bTG, dissolved in 0.5 ml 0.05 M phosphate buffer, giving a rGAL: bTG ratio of 2OO:l on a molar basis. Coupling was performed by adding, dropwise, 20 mg EDAC diluted in 0.5 ml distilled water for a 30-min period with continuous stirring. The reaction was allowed to proceed at room temperature for approximately 24 hours. After the coupling reaction, the mixture was dialyzed at 4°C against 4 liters 0.9% sodium chloride for 4-5 days with two changes of the solution per day. Thereafter, the mixture was divided into aliquots and kept frozen (- 80°C) until used. Four male New Zealand White rabbits (FJL No. 7-10) were used. For the primary insulation, ap~oximately 0.1 mg of the immunogen per rabbit was emulsified with Ribi’s adjuvant [( 15); Immunochem Research, Inc., Hamilton, MT]. The emulsion was administered as suggested by the supplier: subcutaneously (1 site, 0.1 ml), intradermally (6 sites, 0.05 ml) intramuscularly (2 sites, 0.2 ml) and intraperitoneally (0.2 ml). Booster inoculations (approximately 0.05 mg immunogen) were performed monthly following the same protocol used in the primary inoculation. Rabbits were bled from the central ear artery 15 days after each booster to check for anti-rGAL titers.
s
”
-“,*,,,.” l.’
A.
METHOD
‘0
AA
I
FIG. I. Evolution of anti&AL titers in rabbit sera. Titers are defined as the dilution of the sera that bound 25% [as estimated from the “Allfit” program; (2)j of the radioactivity added.
buffer were added to the tubes. Samples were centrifuged for 30 min at 4’C (2500 rpm), the supematant was decanted and the radioactivity present in the pellet counted using an automated g~macounter. Parallelism to the standard curve was tested by using a pool of portal and peripheral serum as well as a pool of extracts from arcuate nucleus-median eminence fragments. Dissection and Extraction Procedures Arcuate nucleus-medic eminence (AN-ME) fragments were dissected under a stereoscopic microscope using a fine pair of scissors as previously described (10). AN-ME fragments were homogenized in 100 ~1 ice-cold 0.1 N acetic acid. Homogenates were heated to 90°C for 10 min and then centrifuged at 12000 x g for 5 min. The supematant was collected and aliquots equivalent to half of the fragment were frozen at - 80°C. Data Analysis and Statistics RIA data were analyzed using the 4 parameter logistic program of Rodbard and Hutt (12). In order to evaluate parallelism to the standard curve as well as to estimate crossreactivity, the algorithm “Allfit” (2) was used. Crossreactivity was expressed as percentage by considering rat GAL IC,, as 100% [Relative inhibition factor, RIF; (13)]. Where possible, the IC, of the peptide checked was used, otherwise when the IC,, was not reached, the highest concentration of the peptide added to the assay (1 p,g, expressed in fmol) was used and the final value expressed as smaller than the estimated RIF for that concentration. Data are expressed as the mean? SE unless otherwise indicated. Statistical analysis was performed by using Mann-~imey U-test (20). A pcO.05 was considered as the minimum criterion to detect significant differences. RESULTS
Figure 1 depicts the evolution of the titers for the four sera.
RAT GALANIN RIA
397
150 [
TABLE 1 . o A A
I-GAL Peri her-al serum pool AN- RrE extract Portal serum pool
SLOPES AND IC,, FOR SYNTHETIC rGAL, RAT SERUM POOLS FROM BOTH PERIPHERAL AND PORTAL BLOOD AND FOR RAT AN-ME EXTRACTS
100
slope*
ffiAL Peripheral
1 50
Portal AN-ME
0
’
-‘,.“”
“,.,,.’
0.01 CONCENTRATION
““’
0.1 (pg).
c 1
““‘*
“.,.,,,’
10
VOLUME (pl),
“““,
100 AN-ME
I%*
1.26 -c 0.03 1.23 k 0.04 1.19 -+ 0.04 1.24 2 0.07
14.93 21.97 1.96 1.58
*Slopes and IC,, were estimated using the algorithm trGAL pgkube. SSerum volume in ~1. JAN-ME equivalents x 103.
+ 0.37t f 0.58$ z!z 0.06$ + 0.085
Allfit (20).
(eq x 10’)
FIG. 2. Displacement curves for synthetic rGAL, peripheral and portal serum and a pool of extracts from arcuate-nucleus median eminence
fragments. Each point represents the mean 2 SE of triplicate observations. Since the SE was smaller than the symbol size, it is included in the symbol and therefore is not shown.
Antisera titers (defined as the serum dilution that binds 25% of the total activity) increased progressively to reach maximum levels after the third booster in two out of the four rabbits (FJL No. 7, No. 9). Using these sera, 25% binding could be obtained at final dilutions of 1:160,000 and 1:280,000 with the third bleeding, respectively. Titers decreased later, being 1:52,ooO and 1:60,000 in the fifth bleeding. The pattern in the other rabbits (FJL No. 8, No. 10) showed a progressive increase along the bleedings. Maximum titers were obtained in the fifth bleeding ( 1:20,000 and 156,000, respectively). Under our assay conditions, nonspecific binding did not exceed 2% of the total counts. In addition, after the third bleeding 80-85% of the radioactivity added was bound when antisera were used at a final dilution of 1400. Due to the high titer and sensitivity obtained with the third bleeding of rabbit FJL No. 7, this serum was used to develop and characterize the RIA. At a final dilution of 1: 120,000 a sensitivity [as defined by Rodbard and Hutt, (12)] of 1 pg or less with an ICso of approximately 15 pg could be obtained depending on tracer quality. A typical standard curve is depicted in Fig. 2. When serial dilutions of biological samples were included in the assay, displacement curves showed perfect parallelism to the standard curve (Fig. 2). In fact, slopes ranging from 1.19 to 1.24 were obtained for AN-ME extracts, portal and peripheral serum, all being similar to that of the synthetic ffiAL standard (Table 1). Thus, no apparent interferences were detected by the inclusion of plasma samples in the assay. It is interesting to note that the IC,, for peripheral plasma was approximately lo-fold higher than that for portal plasma (Table l), indicating higher rGAL levels in the hypophyseal portal circulation. Since samples consisted of a pool, no precise measurements could be accomplished from this observation; however, this strongly supports that GAL is involved in neuroendocrine functions. The antiserum displayed a very high specificity. Table 2 depicts the relative inhibition factors (RIF) for several peptides. All peptides, including pGAL, exhibited a RIF <0.004% indicating that the recognition site of the antiserum resides in the carboxy-terminus portion of the rGAL molecule, and that the antisera does not recognize fragments of the associated GAL
precursor molecule peptides tested.
or any of the hypothalamic
and pituitary
rGAL. Levels in Peripheral Plasm after HYPOX In order to evaluate pituitary contribution to GAL levels in peripheral blood, rGAL concentrations were measured in intact and hypophysectomized male rats. Two weeks after hypophysectomy (HYF’OX), rGAL levels in serum were reduced when compared with levels in intact animals (Table 3). This finding indicates that at least 30% of circulating ffiAL comes from the pituitary gland. DISCUSSION
This study describes the characterization
and development
of a
TABLE 2 RELATIVE INHIBITION FACTORS FOR SEVERAL PEPTIDES USING FJL NO. 7-3 ANTI&AL SERUM Peptide
IuF (So)
ffi AL pGAL GMAP 141 GMAP 1Wl GMAP 2541 GMAP 44-59 LHRH Oxytocin ICRF TRH Dermorphin AVP rat VIP SFUF 28 hACTH Frog Sauvagine Alpha-helical-CRF Beta-endorphin
100.0000 <0.0026 <0.0038 0.0032 <0.0015 <0.0012
RIF was calculated from the IC,,, (fmolkube) obtained from the algorithm “Alltit” (2). In those cases in which IC,, was not reached, the maximum amount of peptide added (1 pg expressed in fmol) was used. Therefore, values are expressed as smaller than that value.
398
L6PEZ,
TABLE 3 GALANIN
LEVELS
IN PERIPHERAL
SERUM
WEEK-HYPOPHYSECTOMIZED
INTACT
MALE
RATS
AND
TWO
GAL levels (pgiml)
Group
Intact
703.53 489.78
Hypox *p
FROM
? 52.24 (12) + 42.47 (8)*
U-test.
specific RIA procedure for measuring rGAL. The procedure provides the possibility of detecting small amounts (pg range) of peptide directly in serum samples, without apparent interferences due to other serum constituents. Equally, the RIA is validated for measuring rGAL concentrations in tissue extracts. Regarding the specificity of the anti&AL serum, it appears that the antigenic determinant resides in the carboxy-terminus portion of the &AL molecule. pGAL, which shares an almost 90% homology with rGAL (6), did not significantly crossreact with the antiserum. Differences between porcine and rat GAL consist of three carboxy-terminus substitutions (6; Ala/Thr, position 28; Tyr/His, position 25; His/Ser, position 22), indicating that these amino acids are essential for full recognition of the molecule by our antisera. Another finding that indirectly indicates the specificity of this antiserum is the fact that biological samples (serum samples and tissue extracts) showed displacement curves paralleling the synthetic rGAL standard curve. Earlier studies, using antisera raised against pGAL, failed to obtain parallel displacement curves to the standard when measuring GAL immunoreactivity in rat biological samples (4, 17, 18). This lack of parallelism lead the authors to postulate several hypothesis to explain this observation. It has been suggested that this nonparallelism may be due to crossreactivity with either fragments of the GAL molecule after processing (14), larger molecular weight forms of the molecule, namely precursor (18); or, simply, to some antibody subpopulations recognizing the carboxy-terminus portion of pGAL (4). Our study supports the latter since: First, a specific antiserum for rGAL
MEADE AND NEGRO-VILAR
provides displacement curves for rat biological samples perfectly paralleling synthetic rGAL curves. Second, our data indicates that our antiserum does not significantly recognize any fragment of the GAL message-associated peptide (GMAP). Due to the unavailability of rat GMAP fragments, we utilized porcine GMAP fragments in our characterization. pGMAP presents a 78% homology to its rat equivalent (6) indicating that most likely precursor molecules are not recognized by the antiserum. This study provides, in addition, some intriguing observations regarding the possible role of GAL as a neuroendocrine modulator. Recently, the presence of mRNA encoding for ffiAL in the anterior pituitary gland has been reported (5). This finding indicates that GAL is synthesized in, and probably released from the pituitary gland. Our study supports this concept, although the contribution of the pituitary gland to circulating ffiAL is limited (70% of rGAL can be measured in serum from two week-HYPOX male rats). rGAL levels in serum from male rats were approximately 700 pg/ml. Other studies have reported undetectable levels of GALlike immunoreactivity in plasma from male rats (5). However, this data should be considered with caution since the antiserum used in that study was raised against pGAL, has much lower sensitivity (25 pg/ml) than ours and was reported to produce nonparallel displacement curves with brain tissue extracts (4). Another interesting finding is the lower IC,, of a portal plasma pool (approximately 2 ~1) as compared to a peripheral plasma pool (approximately 20 ~1). Although this data is inferred from a serum pool, and therefore should be carefully considered, it indicates that rGAL levels in portal circulation should be higher (roughly IO-fold as implied by our present data) than those in the peripheral circulation. If this is correct, GAL would meet one of the criteria to be considered as a hypothalamic hormone. However, further studies are needed for the evaluation of this intriguing observation.
ACKNOWLEDGEMENTS
The authors wish to thank Dr. Melvin Ching (Department of Veterinary Anatomy, Ohio State University, Columbus, OH) for providing us with a portal serum pool. The authors are also indebted to Mr. James Clark and Ms. Page Gardinier for their help in conducting the immunizations.
REFERENCES I.
2.
3.
4.
5.
6.
7.
David, G. S. Solid state lactoperoxidase: a highly stable enzyme for simple, gentle iodination of proteins. Biochem. Biophys. Res. Commun. 48:46&l71; 1972. De Lean, A.; Munson, P. J.; Rodbard, D. Simultaneous analysis of families of sigmoidal curves: application to bioassay, radioligand assay, and physiological dose-response curves. Am. J. Physiol. 235:E97-E102; 1978. De Paolo, L. V.; Negro-Vilar, A. Neonatal monosodium glutamate treatment alters the response of median eminence luteinizing hormone releasing hormone nerve terminals to potassium and prostaglandin E,. Endocrinology 110:835-841; 1982. Gabriel, S. M.; MacGarvey, U. M.; Koenig, J. I.; Swartz, K. J.; Martin, J. B.; Beal, M. F. Characterization of galanin-like immunoreactivity in the rat brain: effects of neonatal glutamate treatment. Neurosci. Lett. 87: 114-120; 1988. Kaplan, L. M.; Gabriel, S. M.; Koenig, J. I.; Sunday, M. E.; Spindel, E. R.; Martin, J. B.; Chin, W. W. Galanin is an estrogen-inducible, secretory product of the rat anterior pituitary. Proc. Natl. Acad. Sci. USA 85:7408-7412; 1988. Kaplan, L. M.; Spindel, E. R.; Isselbacher, K. J.; Chin, W. W. Tissue-specific expression of rat galanin gene. Proc. Natl. Acad. Sci. USA 85:1065-1069; 1988. Koshiyama, H.; Kate, Y.; Inoue, T.; Murakami, Y.; Ishikawa, Y.; Yanaihara, N.; Imura, H. Central galanin stimulates pituitary prolactin secretion in rats: possible involvement of vasoactive intestinal
polypeptide. Neurosci. Lett. 75:49-54; 1987. 8. Melander, T.; Fuxe, K.; Hlrfstrand, A.; Eneroth, P.; HSkfelt, T. Effects of intraventricular injections of galanin on neuroendocrine functions in the male rat: Possible involvement of hypothalamic catecholamine neuronal systems. Acta Physiol. Stand. 131:25-32; 1987. 9. Murakami, Y.; Kato, Y.; Shimatsu, A.; Koshiyama, H.; Hattori, N.; Yanaihara, N.; Imura, H. Possible mechanisms involved in growth hormone secretion induced by galanin in the rat. Endocrinology 124:1224-1229; 1989. 10. Negro-Vilar, A.; Ojeda, S. R.; McCann, S. M. Catecholaminergic modulation of LHRH release by median eminence terminals in vitro. Endocrinology 104:1749-1757; 1979. 11. Olney, J. W. Brain lesions, obesity, and other disturbances in mice treated with monosodium glutamate. Science 164:71C-721; 1969. 12. Rodbard, D.; Hutt, D. M. Statistical analysis of radioimmunoassays and immunoradiometric (labelled antibody) assays: a generalized, weighted, iterative least squares method for logistic curve fitting. In: Radioimmunoassay and related procedures in medicine. vol. 1. International Atomic Energy Agency, Vienna; 1974: 165-192. 13. Rodbard, D.; Lewald, J. E. Computer analysis of radioligand assay and radioimmunoassay data. Acta Endocrinol. (Copenh.) 64(Suppl. 147):79-103; 1970. 14. RBkaeus, A.; Melander, T.; Hiikfelt, T.; Lundberg, J. M.; Tatemoto, K.; Carlquist, M.; Mutt, V. A galanin-like peptide in the central
RAT GALANIN FU4
nervous system and intestine of the rat. Neurosci. Lett. 47:161-166; 1984. 15. Rudbach, J. A.; Cantrell, J. L.; Ulrich, J. T. Molecularly engineered microbial immunostimulators. In: Technological advances in vaccine development. New York: Alan R. Liss, Inc.; 1988:443-459. 16. Sabu, A.; Crowiey, W. R.; Tatemoto, K.; Balasubramaniam, A.; Kalra, S. P. Effects of neuropeptide Y, NPY analog (Norleucine4NPY), galanin and neuropeptide K on LH release in ovariectomized (OVX) and OVX estrogen, p~geste~ne-bated rats. Peptides 8: 921-926; 1987. mapping of 17. Skofitsch, G.; Jacobowitz, D. M. I~unohist~he~c~
galanin-like neurons in the rat central nervous system. Peptides 6509-546; 1985. 18. Skofltsch, G.; Jacobowitz, D. M. Quantitative distribution of galaninlike immunoreactivity in the rat central nervous system. Peptides 7609-613; 1986. 19. Tatemoto, K.; Riikaeus, A.; J&wall, H.; McDonald, T. J.; Mutt, V. Galanin- a novel biologically active peptide from porcine intestine. PEBS Len. 164~124-128; 1985. 20. Wallenstein, S.; Zucker, C. L.; Pleiss, J. L. Some statistical methods useful in circulation research. Circ. Res. 47:1-9; 1980.
Vol. 24, pp. 395-399. PergamonPress pk. 1990.Printedin the U.S.A.
0361-9230190 $3.00 + .Gu
Development and Characterization of a Specific and Sensitive Radioimmunoassay for Rat Galanin: Measurement in Brain Tissue, Hypophyseal Portal and Peripheral Serum FXANCISCO JO& L6PEZ, EDWIN H. MEADE, JR. AND AND&S NEGRO-Vet* Reproductive Neuroendocrinology Section, Laboratory of Molecular and Integrative Neuroscience National Institute of Environmental Health Sciences, Research Triangle Park, NC 27709
Received 7 November 1989 L6PEZ, F. J., E. H. MEADE, JR. AND A. NEGRO-VILAR. Development and characterization of a specific and sensitive radioimmunoassczyfor rat galanin: Measurement in brain tissue, hypophyseal portal and peripheral serum. BRAIN RES BULL 24(3) 395-399, 1990. -Gala& (GAL), a 29 amino acid peptide, is extensively distributed in both brain and intestine having been described as a putative neuroendocrine modulator. Up to this date, available radioi~unoas~ys for measuring GAL have the problem of nonp~alIe1 responses when rat biological samples are used. Due to the recent availabili~ of synthetic rat GAL (rGAL), we pursued the devefopment of a specific RIA for IGAL. Our RIA system presents a high specificity since porcine GAL (pGAL), which differs from ICAL in three carboxy-terminus substitutions, displays negligible crossreactivity. In addition, in the RIA a sensitivity of at least 1 pg/tube can be obtained. All biological samples tested in this study (serum samples from both peripheral and portal blood and arcuate-nucleus median eminence extracts) displayed perfect parallelism to synthetic rGAL standard. The RIA is suitable for measuring minute amounts of rGAL directly in serum samplesfrom both portal and peripheral circulation. Interestingly, the IC, for portal serum was approximately IO-fold lower than that for peripheral serum. This indicates that rGAL may be present in higher concentrations in portal than in peripheral blood. In addition, our data suggest that the p~mi~ gland contributes appro~ma~ly 30% to the total levels of rGAL measured in peripheral serum. Together, this data reinforces the concept that GAL is involved in mechanisms controlling and/or regulating neuroendocrine functions. Rat galanin RIA
Arcuate-ME
LHRH
in the arcuate nucleus and therefore impairs many endocrine functions (3,l l), induces a drastic reduction in GAL content in the median eminence (4). This data reinforces the concept that GAL may play an important role as a neuroendocrine regulator. In the studies dealing with GAL concentrations in the hypothalamus (4, 17, 18), a drastic displacement of parallelism to the standard curve was obtained when an antisera raised against pGAL was used (418). Porcine and rat GAL have a high degree of homology, however, they differ in three amino acids located in the carboxy-terminus portion of the molecule (6). This slight difference may explain the absence of parallelism to the standard curve observed in those studies. In this study we report the development and characterization of a sensitive and specific radioi~unoassay for rat GAL that allows the direct measurement of GAL in serum samples. In addition, our
GALANIN (GAL) is a 29 amino acid peptide of the family of brain gastrointestinal peptides. Originally, it was isolated from porcine intestine (19). Later, it was shown that GAL is widely distributed in rat brain in high concentrations, especially in the hypothalamus ( 14,17). Its location in the hypothalamus and other reports indicating that GAL may stimulate PRL, LH and GH secretion (7-9, 14, 16, 17) point out that GAL may be an irn~~~t mediator in neu~nd~~ne-related phenomena. Using a heterologous radioimmunoassay, it has been shown that in the rat, the highest concentrations of GAL immunoreactivity within the hypothalamus can be found in the median eminence (18). In general, GAL concentrations in the hypothalamus are higher than those observed in telencephalon, mesencephalon and rhombencephalon ( 18). In addition, neonatal treatment with monosodium glutamate, which destroys most of the neuronal networks
‘Requests for reprints should be addressed to Dr. Andrt% Negro-V&u, Reproductive Neuroendocrinology Section, Laboratory of Molecular and Integrative Neuroscience, N.I.E.H.S., Building 101, MD C4-09, Research Triangle Park, NC 27709.
395
396
L6PEZ,
data reveal that in male rats, the pituitary gland contributes approximately 30% to the total amount of rGAL in peripheral serum. Moreover, the study also indicates that rGAL might be present in higher concentrations in portal than in peripheral serum.
1:400000
MEADE AND NEGRO-VILAR
[ t
a
/-‘,
0
\,
i
i’
A
I
,’
Peptides Synthetic rGAL, pGAL, LHRH, oxytocin, rat CRF, TRH, dermorphin, AVP, rat VIP, SRIF 28, human ACTH, frog sauvagine, alpha-helical-CRF, beta-endorphin and the fragments of the galanin message-associated peptide (GMAP) 1-41, 16-41, 25-41 and 44-59 were obtained from Peninsula Laboratories Inc. (Belmont, CA). Peptides were routinely dissolved in 0.1 N acetic acid (1 mglml), divided in aliquots and kept frozen at - 80°C until used.
,,
Radioimmunoassay Synthetic rGAL was iodinated using the lactoperoxidase method (I) by Hazleton Labs. After an initial p~i~cation by HPLC, the tracer was repurified, prior to use in the radioimmunoassay, utilizing Sephadex G-50 fine. The buffer used for the RIA consisted of 0.025 M phosphate buffer containing 1.25% BSA, 0.5% Triton X-100 and 0.1% sodium azide. RIA was performed under disequilibrium conditions to obtain full sensitivity. One hundred )-LIof either standard or samples were incubated for 24 hr with 50 ~1 of ~tise~rn at room temperature. After this preliminary incubation, 50 ~1 of freshly repurified tracer (7000 cpm) were added and the incubation continued for an additional 24-hr period at 4°C. The antiserum (FJL No. 7-3) was utilized at a final dilution of 1:120,000. Bound GAL was separated by adding 50 ~1 of 4% NRS followed by the addition of 50 ~1 of 4:25 goat anti-rabbit gamma globulins (Anti~ies Inc., Davis, CA) and incubating the mixture 4 hr at room temperature. After the incubation was finished, 0.5 ml assay
i
&’
/
,i’
,
/ 1:400
1:40
,--
-*
/
“ FJL, FJL n FJL =---A FJL
O-- --0
o--O
1
.-. 3
I-
1
~~__i_.._i 3 ,1;
#7 #8 #9 #lO
5
BLEEDING +
Generation of Anti-rGAL Serum A conjugate of rGAL/bovine thyroglobulin (bTG, Sigma Chemical Co., St. Louis, MO) was used as immunogen. rGAL was coupled to bTG using I-ethyl-3-(3-dime~y~minop~pyl) carbodiimide (EDAC; Sigma). TGAL (1.4 mg) was diluted in 0.5 ml 0.05 M phosphate buffer (pH =7.5) and mixed with 0.75 mg bTG, dissolved in 0.5 ml 0.05 M phosphate buffer, giving a rGAL: bTG ratio of 2OO:l on a molar basis. Coupling was performed by adding, dropwise, 20 mg EDAC diluted in 0.5 ml distilled water for a 30-min period with continuous stirring. The reaction was allowed to proceed at room temperature for approximately 24 hours. After the coupling reaction, the mixture was dialyzed at 4°C against 4 liters 0.9% sodium chloride for 4-5 days with two changes of the solution per day. Thereafter, the mixture was divided into aliquots and kept frozen (- 80°C) until used. Four male New Zealand White rabbits (FJL No. 7-10) were used. For the primary insulation, ap~oximately 0.1 mg of the immunogen per rabbit was emulsified with Ribi’s adjuvant [( 15); Immunochem Research, Inc., Hamilton, MT]. The emulsion was administered as suggested by the supplier: subcutaneously (1 site, 0.1 ml), intradermally (6 sites, 0.05 ml) intramuscularly (2 sites, 0.2 ml) and intraperitoneally (0.2 ml). Booster inoculations (approximately 0.05 mg immunogen) were performed monthly following the same protocol used in the primary inoculation. Rabbits were bled from the central ear artery 15 days after each booster to check for anti-rGAL titers.
s
”
-“,*,,,.” l.’
A.
METHOD
‘0
AA
I
FIG. I. Evolution of anti&AL titers in rabbit sera. Titers are defined as the dilution of the sera that bound 25% [as estimated from the “Allfit” program; (2)j of the radioactivity added.
buffer were added to the tubes. Samples were centrifuged for 30 min at 4’C (2500 rpm), the supematant was decanted and the radioactivity present in the pellet counted using an automated g~macounter. Parallelism to the standard curve was tested by using a pool of portal and peripheral serum as well as a pool of extracts from arcuate nucleus-median eminence fragments. Dissection and Extraction Procedures Arcuate nucleus-medic eminence (AN-ME) fragments were dissected under a stereoscopic microscope using a fine pair of scissors as previously described (10). AN-ME fragments were homogenized in 100 ~1 ice-cold 0.1 N acetic acid. Homogenates were heated to 90°C for 10 min and then centrifuged at 12000 x g for 5 min. The supematant was collected and aliquots equivalent to half of the fragment were frozen at - 80°C. Data Analysis and Statistics RIA data were analyzed using the 4 parameter logistic program of Rodbard and Hutt (12). In order to evaluate parallelism to the standard curve as well as to estimate crossreactivity, the algorithm “Allfit” (2) was used. Crossreactivity was expressed as percentage by considering rat GAL IC,, as 100% [Relative inhibition factor, RIF; (13)]. Where possible, the IC, of the peptide checked was used, otherwise when the IC,, was not reached, the highest concentration of the peptide added to the assay (1 p,g, expressed in fmol) was used and the final value expressed as smaller than the estimated RIF for that concentration. Data are expressed as the mean? SE unless otherwise indicated. Statistical analysis was performed by using Mann-~imey U-test (20). A pcO.05 was considered as the minimum criterion to detect significant differences. RESULTS
Figure 1 depicts the evolution of the titers for the four sera.
RAT GALANIN RIA
397
150 [
TABLE 1 . o A A
I-GAL Peri her-al serum pool AN- RrE extract Portal serum pool
SLOPES AND IC,, FOR SYNTHETIC rGAL, RAT SERUM POOLS FROM BOTH PERIPHERAL AND PORTAL BLOOD AND FOR RAT AN-ME EXTRACTS
100
slope*
ffiAL Peripheral
1 50
Portal AN-ME
0
’
-‘,.“”
“,.,,.’
0.01 CONCENTRATION
““’
0.1 (pg).
c 1
““‘*
“.,.,,,’
10
VOLUME (pl),
“““,
100 AN-ME
I%*
1.26 -c 0.03 1.23 k 0.04 1.19 -+ 0.04 1.24 2 0.07
14.93 21.97 1.96 1.58
*Slopes and IC,, were estimated using the algorithm trGAL pgkube. SSerum volume in ~1. JAN-ME equivalents x 103.
+ 0.37t f 0.58$ z!z 0.06$ + 0.085
Allfit (20).
(eq x 10’)
FIG. 2. Displacement curves for synthetic rGAL, peripheral and portal serum and a pool of extracts from arcuate-nucleus median eminence
fragments. Each point represents the mean 2 SE of triplicate observations. Since the SE was smaller than the symbol size, it is included in the symbol and therefore is not shown.
Antisera titers (defined as the serum dilution that binds 25% of the total activity) increased progressively to reach maximum levels after the third booster in two out of the four rabbits (FJL No. 7, No. 9). Using these sera, 25% binding could be obtained at final dilutions of 1:160,000 and 1:280,000 with the third bleeding, respectively. Titers decreased later, being 1:52,ooO and 1:60,000 in the fifth bleeding. The pattern in the other rabbits (FJL No. 8, No. 10) showed a progressive increase along the bleedings. Maximum titers were obtained in the fifth bleeding ( 1:20,000 and 156,000, respectively). Under our assay conditions, nonspecific binding did not exceed 2% of the total counts. In addition, after the third bleeding 80-85% of the radioactivity added was bound when antisera were used at a final dilution of 1400. Due to the high titer and sensitivity obtained with the third bleeding of rabbit FJL No. 7, this serum was used to develop and characterize the RIA. At a final dilution of 1: 120,000 a sensitivity [as defined by Rodbard and Hutt, (12)] of 1 pg or less with an ICso of approximately 15 pg could be obtained depending on tracer quality. A typical standard curve is depicted in Fig. 2. When serial dilutions of biological samples were included in the assay, displacement curves showed perfect parallelism to the standard curve (Fig. 2). In fact, slopes ranging from 1.19 to 1.24 were obtained for AN-ME extracts, portal and peripheral serum, all being similar to that of the synthetic ffiAL standard (Table 1). Thus, no apparent interferences were detected by the inclusion of plasma samples in the assay. It is interesting to note that the IC,, for peripheral plasma was approximately lo-fold higher than that for portal plasma (Table l), indicating higher rGAL levels in the hypophyseal portal circulation. Since samples consisted of a pool, no precise measurements could be accomplished from this observation; however, this strongly supports that GAL is involved in neuroendocrine functions. The antiserum displayed a very high specificity. Table 2 depicts the relative inhibition factors (RIF) for several peptides. All peptides, including pGAL, exhibited a RIF <0.004% indicating that the recognition site of the antiserum resides in the carboxy-terminus portion of the rGAL molecule, and that the antisera does not recognize fragments of the associated GAL
precursor molecule peptides tested.
or any of the hypothalamic
and pituitary
rGAL. Levels in Peripheral Plasm after HYPOX In order to evaluate pituitary contribution to GAL levels in peripheral blood, rGAL concentrations were measured in intact and hypophysectomized male rats. Two weeks after hypophysectomy (HYF’OX), rGAL levels in serum were reduced when compared with levels in intact animals (Table 3). This finding indicates that at least 30% of circulating ffiAL comes from the pituitary gland. DISCUSSION
This study describes the characterization
and development
of a
TABLE 2 RELATIVE INHIBITION FACTORS FOR SEVERAL PEPTIDES USING FJL NO. 7-3 ANTI&AL SERUM Peptide
IuF (So)
ffi AL pGAL GMAP 141 GMAP 1Wl GMAP 2541 GMAP 44-59 LHRH Oxytocin ICRF TRH Dermorphin AVP rat VIP SFUF 28 hACTH Frog Sauvagine Alpha-helical-CRF Beta-endorphin
100.0000 <0.0026 <0.0038 0.0032 <0.0015 <0.0012
RIF was calculated from the IC,,, (fmolkube) obtained from the algorithm “Alltit” (2). In those cases in which IC,, was not reached, the maximum amount of peptide added (1 pg expressed in fmol) was used. Therefore, values are expressed as smaller than that value.
398
L6PEZ,
TABLE 3 GALANIN
LEVELS
IN PERIPHERAL
SERUM
WEEK-HYPOPHYSECTOMIZED
INTACT
MALE
RATS
AND
TWO
GAL levels (pgiml)
Group
Intact
703.53 489.78
Hypox *p
FROM
? 52.24 (12) + 42.47 (8)*
U-test.
specific RIA procedure for measuring rGAL. The procedure provides the possibility of detecting small amounts (pg range) of peptide directly in serum samples, without apparent interferences due to other serum constituents. Equally, the RIA is validated for measuring rGAL concentrations in tissue extracts. Regarding the specificity of the anti&AL serum, it appears that the antigenic determinant resides in the carboxy-terminus portion of the &AL molecule. pGAL, which shares an almost 90% homology with rGAL (6), did not significantly crossreact with the antiserum. Differences between porcine and rat GAL consist of three carboxy-terminus substitutions (6; Ala/Thr, position 28; Tyr/His, position 25; His/Ser, position 22), indicating that these amino acids are essential for full recognition of the molecule by our antisera. Another finding that indirectly indicates the specificity of this antiserum is the fact that biological samples (serum samples and tissue extracts) showed displacement curves paralleling the synthetic rGAL standard curve. Earlier studies, using antisera raised against pGAL, failed to obtain parallel displacement curves to the standard when measuring GAL immunoreactivity in rat biological samples (4, 17, 18). This lack of parallelism lead the authors to postulate several hypothesis to explain this observation. It has been suggested that this nonparallelism may be due to crossreactivity with either fragments of the GAL molecule after processing (14), larger molecular weight forms of the molecule, namely precursor (18); or, simply, to some antibody subpopulations recognizing the carboxy-terminus portion of pGAL (4). Our study supports the latter since: First, a specific antiserum for rGAL
MEADE AND NEGRO-VILAR
provides displacement curves for rat biological samples perfectly paralleling synthetic rGAL curves. Second, our data indicates that our antiserum does not significantly recognize any fragment of the GAL message-associated peptide (GMAP). Due to the unavailability of rat GMAP fragments, we utilized porcine GMAP fragments in our characterization. pGMAP presents a 78% homology to its rat equivalent (6) indicating that most likely precursor molecules are not recognized by the antiserum. This study provides, in addition, some intriguing observations regarding the possible role of GAL as a neuroendocrine modulator. Recently, the presence of mRNA encoding for ffiAL in the anterior pituitary gland has been reported (5). This finding indicates that GAL is synthesized in, and probably released from the pituitary gland. Our study supports this concept, although the contribution of the pituitary gland to circulating ffiAL is limited (70% of rGAL can be measured in serum from two week-HYPOX male rats). rGAL levels in serum from male rats were approximately 700 pg/ml. Other studies have reported undetectable levels of GALlike immunoreactivity in plasma from male rats (5). However, this data should be considered with caution since the antiserum used in that study was raised against pGAL, has much lower sensitivity (25 pg/ml) than ours and was reported to produce nonparallel displacement curves with brain tissue extracts (4). Another interesting finding is the lower IC,, of a portal plasma pool (approximately 2 ~1) as compared to a peripheral plasma pool (approximately 20 ~1). Although this data is inferred from a serum pool, and therefore should be carefully considered, it indicates that rGAL levels in portal circulation should be higher (roughly IO-fold as implied by our present data) than those in the peripheral circulation. If this is correct, GAL would meet one of the criteria to be considered as a hypothalamic hormone. However, further studies are needed for the evaluation of this intriguing observation.
ACKNOWLEDGEMENTS
The authors wish to thank Dr. Melvin Ching (Department of Veterinary Anatomy, Ohio State University, Columbus, OH) for providing us with a portal serum pool. The authors are also indebted to Mr. James Clark and Ms. Page Gardinier for their help in conducting the immunizations.
REFERENCES I.
2.
3.
4.
5.
6.
7.
David, G. S. Solid state lactoperoxidase: a highly stable enzyme for simple, gentle iodination of proteins. Biochem. Biophys. Res. Commun. 48:46&l71; 1972. De Lean, A.; Munson, P. J.; Rodbard, D. Simultaneous analysis of families of sigmoidal curves: application to bioassay, radioligand assay, and physiological dose-response curves. Am. J. Physiol. 235:E97-E102; 1978. De Paolo, L. V.; Negro-Vilar, A. Neonatal monosodium glutamate treatment alters the response of median eminence luteinizing hormone releasing hormone nerve terminals to potassium and prostaglandin E,. Endocrinology 110:835-841; 1982. Gabriel, S. M.; MacGarvey, U. M.; Koenig, J. I.; Swartz, K. J.; Martin, J. B.; Beal, M. F. Characterization of galanin-like immunoreactivity in the rat brain: effects of neonatal glutamate treatment. Neurosci. Lett. 87: 114-120; 1988. Kaplan, L. M.; Gabriel, S. M.; Koenig, J. I.; Sunday, M. E.; Spindel, E. R.; Martin, J. B.; Chin, W. W. Galanin is an estrogen-inducible, secretory product of the rat anterior pituitary. Proc. Natl. Acad. Sci. USA 85:7408-7412; 1988. Kaplan, L. M.; Spindel, E. R.; Isselbacher, K. J.; Chin, W. W. Tissue-specific expression of rat galanin gene. Proc. Natl. Acad. Sci. USA 85:1065-1069; 1988. Koshiyama, H.; Kate, Y.; Inoue, T.; Murakami, Y.; Ishikawa, Y.; Yanaihara, N.; Imura, H. Central galanin stimulates pituitary prolactin secretion in rats: possible involvement of vasoactive intestinal
polypeptide. Neurosci. Lett. 75:49-54; 1987. 8. Melander, T.; Fuxe, K.; Hlrfstrand, A.; Eneroth, P.; HSkfelt, T. Effects of intraventricular injections of galanin on neuroendocrine functions in the male rat: Possible involvement of hypothalamic catecholamine neuronal systems. Acta Physiol. Stand. 131:25-32; 1987. 9. Murakami, Y.; Kato, Y.; Shimatsu, A.; Koshiyama, H.; Hattori, N.; Yanaihara, N.; Imura, H. Possible mechanisms involved in growth hormone secretion induced by galanin in the rat. Endocrinology 124:1224-1229; 1989. 10. Negro-Vilar, A.; Ojeda, S. R.; McCann, S. M. Catecholaminergic modulation of LHRH release by median eminence terminals in vitro. Endocrinology 104:1749-1757; 1979. 11. Olney, J. W. Brain lesions, obesity, and other disturbances in mice treated with monosodium glutamate. Science 164:71C-721; 1969. 12. Rodbard, D.; Hutt, D. M. Statistical analysis of radioimmunoassays and immunoradiometric (labelled antibody) assays: a generalized, weighted, iterative least squares method for logistic curve fitting. In: Radioimmunoassay and related procedures in medicine. vol. 1. International Atomic Energy Agency, Vienna; 1974: 165-192. 13. Rodbard, D.; Lewald, J. E. Computer analysis of radioligand assay and radioimmunoassay data. Acta Endocrinol. (Copenh.) 64(Suppl. 147):79-103; 1970. 14. RBkaeus, A.; Melander, T.; Hiikfelt, T.; Lundberg, J. M.; Tatemoto, K.; Carlquist, M.; Mutt, V. A galanin-like peptide in the central
RAT GALANIN FU4
nervous system and intestine of the rat. Neurosci. Lett. 47:161-166; 1984. 15. Rudbach, J. A.; Cantrell, J. L.; Ulrich, J. T. Molecularly engineered microbial immunostimulators. In: Technological advances in vaccine development. New York: Alan R. Liss, Inc.; 1988:443-459. 16. Sabu, A.; Crowiey, W. R.; Tatemoto, K.; Balasubramaniam, A.; Kalra, S. P. Effects of neuropeptide Y, NPY analog (Norleucine4NPY), galanin and neuropeptide K on LH release in ovariectomized (OVX) and OVX estrogen, p~geste~ne-bated rats. Peptides 8: 921-926; 1987. mapping of 17. Skofitsch, G.; Jacobowitz, D. M. I~unohist~he~c~
galanin-like neurons in the rat central nervous system. Peptides 6509-546; 1985. 18. Skofltsch, G.; Jacobowitz, D. M. Quantitative distribution of galaninlike immunoreactivity in the rat central nervous system. Peptides 7609-613; 1986. 19. Tatemoto, K.; Riikaeus, A.; J&wall, H.; McDonald, T. J.; Mutt, V. Galanin- a novel biologically active peptide from porcine intestine. PEBS Len. 164~124-128; 1985. 20. Wallenstein, S.; Zucker, C. L.; Pleiss, J. L. Some statistical methods useful in circulation research. Circ. Res. 47:1-9; 1980.