Journal of Chemical Neuroanatomy, Vol. 5:503 509 (1992)

Production and Immunohistochemical Application of Monoclonal Antibodies against Delta Sleep-Inducing Peptide Y. Charnay, J. Golaz, P. G. Vallet and C. Bouras Service de la Recherche Biotogique, Institutions Universitaires de Psychiatric de Gendve. Switzerhmd

ABSTRACT Monoclonal antibodies were produced following immunization el" rats with delta sleep-including peptide (DSIP). The spleen cells of the rats were fused with the myeloma cell line SP2:0. The supernatants o1" hybridomas were screened on a solid-phase immunoassay using dot-immunobinding of DS1P and some DS1P fragments. The supernatants of six stable producer clones were Found to react with DSIP. Vrom this procedure it was also deduced that all these monoclonal antibodies recognized epitope(s) of the penta carboxy-terminal region of DSIP (DSI P5 9). Application of these monoclonal antibodies to rat median eminence sections gave a strong immunolabelling of a large population of fibres and terminallike structures, mainly localized through the lateral areas. Elution-restainmg experiments using a monochmal antibody to DSIP and a polyclonal antiserum to luteinizing hormone-releasing hormone (LH R H) showed that the patterns of immunoreactivity respectively visualized overlap almost completely. Although numerous LH RH-immunoreactive neuronal elements were also easily demonstrated in the median eminence of the mouse, the hamster and the gerbil species, incubation of sections with monoclonal antibodies to DSIP failed to give any immunoreaction. Taken together these data argue lk)r the independence of the DSIP/LHRH irnmunolabelling systems. Furthermore, it x~,as demonstrated that DSIP5 9-related epitopes detected in the rat median eminence have no counterpart in the three other rodent species investigated. These species differences may reflect the fact that the carboxyterminal sequence of the nonapeptide DSIP originally discovered in the rabbit is not conserved in all rodent species. Kl','

WORDS: l)ot-immunobinding

Cellulose polyacetate membrane LHRH

INTRODUCTION The nonapeptide delta sleep-inducing peptide (DSIP) was first isolated from cerebral blood dialysate of the rabbit (Schoenenberger el al., 1978). Besides having a controversial possible regulatory role in sleep (see Borbdly and ToNer, 1989) it was reported that DSIP may interact with several other functions including thermoregulation (Yehuda et a/., 1988) and immunity (Yehuda et al., 1987). There are several data suggesting that DSIP may modulate hormonal secretion at various levels of the hypothalamo-pituitary-adrenal axis ( G r a f el al., 1985: Iver and McCann, 1987: Iyer et al., 1988: Bjartell el al., 1989). Thus, a stimulating effect ot" DSIP on the release of luteinizing hormone (LH) from anterior pituitary cells, possibly via a hypothalamic site of action, has been reported (lyer and McCann, 1987). Although different patterns o1" distribution of DSIP-like immunoreactive To x,,hom correspondence should be addressed: Dr Yves Charnay. Division of Morphological Psychopathology, I UPG. 101L A,,. de Bcl-Air. CH-1225 Ch~}ne-Bourg, Switzerland. Fax no 41 22 :;O'S53 98.

0891 0618:92:060503 07508.50 a> 1992 by John Wiley and Sons Ltd

Muro?d median eminence

neurons through the rat brain have been described (Constantinidis el al., 1983: Feldman and Kastin, 1984; Vallet el al., 1991 : Skagerberg el a/., 1991 ), a high density ofimmunoreactive fibres and terminallike structures was generally visualized in the mediobasal hypothalamic area. Using rat polyclonal antisera against DSIP we have shown that in the rabbit (Charnay el ell., 1989) and subsequently in the human (Vallet el ell., 1990) and the cat (Charnay et al.. 1990) brain DSIP-likc immunoreactivity was mainly present in LH-releasing hormone ( L H R H ) containing neurons. Furthermore, electron microscopic studies performed in rat (Vallet cl a/., 1991) and guinea pig (Pu et al., 1991) median eminence demonstrated that DSIP-like immunoreactivity and L H R H were costored within single secretory granules. We report here the production of monoclonal antibodies against synthetic DSIP (DSIP mAbs) and the determination of their region spccificily using a dot-immunobinding procedure. Using these DSIP mAbs, a comparative immunohistochemical study was undertaken to evaluate the extent ofDSIP, L H R H colocalization in the median eminence of four rodent species.

51)4 Y. Charnay et al. Table 1.Aminoacidsequenceof delta sleep-inducingpeptide(DSIP)(Schoenenberg.~ 1978), DSIPfragments,luteinizinghormone-releasinghormone(LHRH)(Adelmane/ al.. 1986)and rat 13-endorphin(Drouinand Goodman, 1980) Peptide

Aminoacid sequence

DSIP DSIPI 4 DSIP3-6 DSIP5 9 DSIP-Tyr10 LHRH [3-Endorphin

Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu Trp-Ala-Gly-Gly Gly-Gly-Asp-Ala Asp-Ala-Ser-Gly-Glu Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu-Tyr pGlu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-Gly-NH2 Tyr-Gly-Gly-Phe-Met-Thr-Ser-Glu-Lys-Ser-Gln-Thr-Pro-LeuVal-Thr-Leu-Phe-Lys-Asn-Ala-Ile-Ile-kys-Asn-Val-His-Lyskys-Gly-GIn

MATERIALS AND METHODS

Animals Animals used for immunohistochemistry were adult male Wistar rats, golden hamsters (SAVO, Kissleg, Germany), mice of various strains (BA LB/ c; OF1, NMRI; C3H and C57/BL6; IFFACREDO, Les Oncins, France) and gerbils (CERJ, Le Genest-St-Isle, France). Peptides and proteins Synthetic DSIP (Ro 13-8274), DSIP1-4 (Rol 5-8957) and DSIPS5-9 (Ro15-8955) were generously provided by Prof. W. Haefeley (La Roche-Hoffmann, Basel, Switzerland). DSIP3-6 and LHRH were purchased from Bachem (Bubendorf, Switzerland). DSIP-Tyrl0 was synthesized by Neosystem (Strasbourg, France). Rat gonadotrophin-releasing hormone-associated peptide (GAP) (Adelman et al., 1986) was purchased from Peninsula (Palo Alto, USA). Rat 13-endorphin was provided by Sigma (St Louis, USA). Porcine thyroglobulin (THG) was purchased from Serva (Heidelberg, Germany). Production of monocionai antibodies Wistar male rats received two subcutaneous injections of approximately 50 ng of DSIP/THG (molar ratio 51:1) conjugated via glutaraldehyde (24) in 0.3 ml saline emulsified with the same volume of complete Freund's adjuvant (FA) at days 0 and 10. Two booster injections of 20 ng of immunogen in 0.3 ml saline emulsified with the same volume of incomplete FA were given intraperitoneally at days 20 and 30. Finally, each animal received an additional intraperitoneal injection of 20ng of immunogen in 0.3 ml saline without FA at day 40. Three days later 107 spleen cells of immunized rat were fused with 10v Sp/2/o mouse myeloma cells using the method reported by Lovenborg (1982). The fused cells were suspended in RPMI 1640 containing 15% fetal calf serum, 10-4 u-hypoxanthine, 1.6 × 10 5 M-thymidine and 10 - 6 M-aminopterine (HAT medium). The cell suspension was plated in

five 96-well plates with macrophage feeder layers. Seven to 10 days later supernatant screening was performed by the immunoblotting test (see below) and by indirect immunofluorescence on cryostatfixed sections through rat or rabbit hypothalamic region as previously reported (Charnay et at., 1989). Clones which gave a positive reaction in the immunoblotting test and/or in tissue sections were selected and immediately subcloned at least twice by limiting dilution. Supernatants of stable hybridoma expanded in culture flasks were collected, routinely checked as above and stored under sterile conditions at 4°C or frozen at - 80°C. Isotyping of the monoclonal antibodies was determined by ELISA using subclass-specific anti-rat IgG (IgG1, 2a, 2b and 3c) conjugated with horseradish peroxidase (see Leger et al., 1990).

Dot-immunobinding A dot-immunobinding procedure according to the cytochemical model first developed by Larsson (1981) was used with some minor modifications. DSIP, DSIP fragments and other peptides (see Table 1) (2 ng in 2 ~tl of water) were dotted in triplicate on a cellulose polyacetate membrane (Sepraphore 111, Gelman Sciences Inc., Ann Harbor, MI) and fixed for 90 min at 80°C by formaldehyde vapours. The membrane was then rinsed in 0.1 M-phosphate buffer saline, pH 7.4 (PBS) and incubated with a hybridoma supernatant ( 1:10) or a rat polyclonal antisera to DSIP (R7) and to LHRH (R1) (see Charnay et al., 1989) diluted 1:500 in PBS containing 0.1% Triton X for 1 h at 20°C. After several washes with PBS the sites of immunoreaction were revealed by incubating the membrane sequentially in anti-rat IgG (1:50; 1 h at 207C) and rat PAP (1:3000; 1 h at 20°C) (ICN, Costa Mesa, CA). 0.5 mg/ml 3-3' diaminobenzidine in the presence of 0.01% H202 was used as chromogen (Sternberger et al., 1970). Preparation of tissues and immunohistochemistry The animals, deeply anaesthetized with pentobarbital (Nembutal, 50 mg/kg), were perfused with

Monoclonal antibc)dics against I)SIP

505

DSIP i~ 2000 ng

DSIP 1-4

i~=~ 200

DSIP 3-6

20

DSIP 5-9

2

DSIP-Tyrl0

0.2

LHRH

a

aA , 8-Endorphln

b

cd

e

Fig. l, D,~t-immunobindingexperiments on cellulose polyacetatc membranes. In (a) various amounts (20 2000 ngl of I)SI P wcrc spoltcd Note lhal a spot of 20 ng of peptide can be detected with a DSIP monoclonal antibody (working dilution l:101.2 fig of DSI P. I)SI P-Txr, DgIP I'ragments. LHRH, GAP and [}-endorphin were spotted in (b e) and strips v,ere immunoreactcd with DSIP m.,\bs RoSllb) and RoS2{c), a polyclonal antiserum to DSIP (d) and to LHRH (e). All DS1P monoclonal antibodies obtained gave thc same pallcrn oI labelling Note that DS1P antibodies brad the fragment DSIP5 9 as well as the nonapeptidc. 30 200ml o f T y r o d e ' s solution followed by 100 800 ml o f chilled 4% p a r a f o r m a l d e h y d e in PBS as fixative. The brain was carefully removed and a tissue block including the entire hypothalamic region was postfixed by immersion in the same fixative l\)r 20 h at 4 C. Thc tissue block was then kept in 10% sucrose PBS for 24 48 h at 4 C. Cryostat frontal sections (12 lain thick) at various levels through the median eminence were collected on poly-L-lysincd slides and then immediately processed for i m m u n o fluorescence as previously reported (Charnay et al., 1989). Briefly, sections were first incubated alternately by a polyclonal antiserum to L H R H , to D S I P (working dilutions 1:500 1:1000) or undilutcd h y b r i d o m a supernatant for I h at 2 0 ( ' . After several rinses with PBS. sections were incubated ,aith a goat anti-rat lgG conjugated to fluoresceine isothyocyanate ( F I T C - G A R IgG) (Miles Lid Co)tit it working dilution of 1:200, l\~r 1 h at 20 C. After several rinses with PBS sections were examined and p h o t o g r a p h e d under a Zeiss A x i o p h o t microscope. Some sections immunoreacted for D S I P were subjected to tin elution-restaining procedure for LH RH (and rice versa) exactly as described elsewhere ( ( ' h a r n a y et al., 1989; Vallet el al., 1990. A complete clution of the first imrnunolabelling was obtained b} successive immersion o f the sections in 0.125%, K M I I ( ) 4 in PBS for 10rain and then in a sohltion containing I % oxalic acid and 1% potassium distilphite in PBS for I m i n . The efficiency o f the elution was rotltinely controlled tinder the microscope after incubation of the sections with the secondary antiserum ( F I T C - G A R IgG). Eluted sections were rinsed with PBS and then re-incubated successively' with a primary antibody and with a F I T C conjugated secondary antibody as reported above

and p h o t o g r a p h e d again. Other contl-~H~, included preincubation of the primary, antibodics with their respective h o m o l o g o u s (1 10t.txll and hctcrologus (1001.tM) antigens including rat (i.,\P alld [{endorphin for 1 h lit 37 ( tilld Fcpittcenlcn{ Of the m A b s a g a i n s t D S I P by the superlla.lt:.lnl, of 11011secreting hybridomas.

RESULTS

Production and specificity of mAbs against DSI P A m o n g 960 wells checked. 21 sfiowed Ii\bridoma secreting IgG inmlunodeicctablc in our screening procedures. Several o f them lost their production of positive lgG during the succcssi\e cloning by limiting dilution and linally six hybridoma remained stable after expansion in culture flasks. The antibodies respectivel\ produced b~ these hx.bridoma were referenced RoS1-S6. The isoiype of all these m A b s wlts l\)und to be IgG 2a. As illustratcd in Fig. l(a c) d o t - i m m u n o b i n d i n g experiments sho\~ed that these m A b s bound the fragment DSIP5-9 its well as the native DSIP. A weak reaction toward D S I P bearing an additional tyrosine at lhe carboxvterminal position ( D S I P - T v r l 0 t was occasionally obser,,cd. Spots of D S l P l - 4 and I)SI P3-()tragmenls were ncvcr rccognized. Finally', I . H R H . G A P and rat ~-endorphin were never recogni/cd x~hcn dotted in the same conditions (Fig. Ib aim c ).

Specificity of the polyclonal antisera Tile d o t - i m m u n o b i n d i n g o f the polyclonal antiserum to DSIP (RT) indicated that thc antiscrmn

506

Y. Charnay et al.

Fig. 2. Immunofluorescentmicrographs of a frontal section through the rat median eminence showing the distribution of DSIP (a) and LHRH (c). The section was incubated with a monoclonal antibody to DSIP (a), eluted and then restained for LHRH (c). The patterns of immunoreactivityoverlap almost completely.Compare the inserts (b) and (d). Bars indicate respectively30 Jamin (a) and (c), 15 ~Nnin (b) and (d). predominantly recognizes the carboxy-terminal portion of DSIP (DSIP5-9). An immunoreaction toward the spot D S I P - T y r I 0 was also detected. All other spots of peptides or DSIP fragments were negative (see Fig. ld). In the same conditions the polyclonal antiserum to L H R H ( R I ) recognizes L H R H except the other peptides dotted including DSIP and DSIP fragments. (Fig. le).

lmmunohistochemistry The immunohistochemical application of DSIP mAbs to the rat median eminence resulted in the visualization of a dense network of fibres and terminal-like structures mainly concentrated in the lateral area (Figs 2a and b; 3a). Medially the immunoreactive fibres were few and predominantly oriented radially (Fig. 2a and b). All the six mAbs to DSIP selected gave a distribution pattern of immunoreactivity also observable with the polyclonal antiserum to D S I P (R7). The immunoreaction was totally abolished when the mAbs to DSIP were preabsorbed with DSIP or DSIP5-9 (1 laM) or replaced by the supernatant of a non-producing hybridoma cell line. The preincubation of the mAbs to DSIP with DSIP1-4, DSIP3-6, L H R H and rat [3-endorphin (101~M) did not influence the immunoreaction (data not shown). Elution-restaining experiments using a m A b to DSIP and a polyclonal antiserum to L H R H released an overlapping distribution pattern of DSIP- and

LHRH-immunoreactive neuronal elements. In some instances, it was possible to define DSIP/ L H R H in the same fibres (Fig. 2, compare b and d). Under the same procedures using the mAbs or the polyconal antiserum to DSIP, we were unable to demonstrate any DSIP-immunoreactivity in the median eminence of the mouse, the gerbil and the hamster (Fig. 3c, e and f). In contrast, a large population of LHRH-immunoreactive fibres and terminal-like structures was observed in the median eminence of all the species investigated (Fig. 3b, d, f and h). DISCUSSION The dot-immunobinding experiments presented above clearly indicate that all the mAbs obtained recognize epitope(s) borne by the region DSIP 5-9. It was further shown that DSIP elongated by a tyrosine residue at the carboxy-terminal position is poorly recognized by these mAbs, thus suggesting a possible contribution of the DSIP carboxy-terminal residue for the antigenic specificity. The lack of apparent crossreactivity of the DStP mAbs with rat [3-endorphin, a peptide sharing a sequence homology of two residues at the C-terminal end, indicates that some other adjacent residues included in the DSIP5-7 region may also be of importance for the epitope recognition. Formaldehyde fixation of tissues, traditionally used for immunohistochemistry, cross-links proteins and peptides via

Monoclonal antibodies against DSIP

507

d

>. -i!

"~

A

f m

AN g

h

1 ig. 3. h n m u n o f l u o r c s c e n c e m i c r o g r a p h s of the m e d i a n emincnce of ra! (a, b), m o u s e (c, d), golden h a m s l e r (c, t'l and gerbil (g, I1) after i n c u b a t i o n with a m o n o c l o n a l a n t i b o d y to D S I P ( a , c , e, gl and a polyclonal a n t i s e r u m to [ . H R H (b, d, f, h). Notc that in the rat medi~ln eminence an elution of DSI P-like i m m u n o l a b e l l i n g (a) and restaining with LH R H a n t i s e r u m (b) on the same section gave the same pattern o l ' d i s l r i b u l i o n . N o t e the high density o f i m m u n o r e a c t i v e fibres and terminal-like struclurcs, cspccially in the lalcral portion of the median eminence {arrows). A l t h o u g h n u n l c r o u s L H R H - i m m u n o r e a c t i v e fibres and terminal-like struclures ~ e r e also obser'~ed t h r o u g h the medinn eminence o f l h e lnousc (d). lhe h a m s t e r ( f ) and the gerbil (11), D S I P - l i k e i m m u n o r e a c t i v i t y was hexer obser\ ed m these species (c, e, g ) AN: ai-cua.ltc nucleus. B:.u-s: 30 p.m.

508

Y. Charnay et al.

their amino groups (Puchtler and Meloan, 1985) and it is known that antibodies directed against epitopes including amino groups often do not contribute to the immunostaining on aldehyde-fixed tissues (Sternberger, 1979). Therefore it is conceivable that the screening o f our mAbs by using dots of DSIP immobilized on membranes by formaldehyde preferentially revealed mAbs directed to the carboxy-terminal region of DSIP. The same may be true for the polyclonal antiserum (R7) which virtually appears directed against the C-terminal domain of the nonapeptide. The immunohistochemical application of these mAbs on tissue sections through the rat median eminence gives distribution patterns of immunoreactivity similar to that observed elsewhere (Vallet et al., 1991; Skagerberg et al., 1991) and as reported here using a polyclonal antiserum to DSIP. Considering the absorption controls presented above (see results) the immunoreaction revealed by the mAbs to DSIP can be regarded as specific. Together, these immunohistochemical findings are compatible with previous radioimmunoassay/ chromatographic analyses, suggesting the presence of the nonapeptide DSIP in the rat hypothalamus (see G r a f a n d Kastin, 1987). The sequential doubleimmunostaining experiments using DSIP mAbs and a polyclonal antiserum to L H R H indicate that the distributions of the immunoreactive neuronal elements respectively visualized coincide. This finding agrees with our previous electron microscopic studies which have shown that in the rat and the guinea pig median eminence DSIP- and L H R H immunoreactivity were found costored within the same secretory granules (Vallet et al.. 1991; Pu et al., 1991). Despite the numerous LHRH-immunoreactive neuronal elements visualized in the median eminence of the hamster, the gerbil and various strains of mice, we were unable to detect any DSIP-like immunoreactivity as demonstrated in the rat. In addition to the controls performed at the different steps of the immunohistochemical procedures, the species differences observed argue in favour of the independence of the D S I P / L H R H immunolabelling systems reported above. Furthermore, it can be deduced that the epitope(s) immunologically related to DSIP revealed through the rat median eminence has no counterpart in the three other rodent species investigated. Although the sequences of neuropeptides are rather well conserved in mammals, e.g. L H R H (King and Millar, 1990), subtle interspecies differences concerning a few residues are known, e.g.: the respective sequences of vasoactive intestinal peptide, peptide histidine isoleucinamide and secretin in rat differ from those established in the guinea pig species (Buscail et al., 1990). Thus, some variation of amino acid residues in the epitope region which our mAbs map to (i.e. DSIP5-9) might reasonably explain the species difference observed in the present study. It was recently reported that mouse pituitary

cells in culture synthesize a small DSIP-like pepttdc chromatographically distinct from the authentic nonapeptide (Bjartell et al., 1990). However, there is as yet no biochemical evidence concerning the presence of DSlP in the mouse brain, nor the hamster or the gerbil. Due to our inability to obtain mAbs revealing epitopes borne by the N-terminal part of DSIP, the possibility that the brain of these species is devoid of DSIP cannot be totally excluded at the present time. In conclusion, this immunohistochemical study using mAbs demonstrates that one or more epitope(s) related to DSIP5-9 easily detected in the rat median eminence have no counterpart in three other rodent species, i.e.: the mouse, the hamster and the gerbil. One advantage of mAbs is that large amounts of specific lgG can be obtained. Further biochemical studies including immunoaffinity purification are necessary to determine more precisely the nature of the molecule recognized in the rat species. ACKNOWLEDGEMENTS We would like to thank Prof. V. Kovalzon and his group for helpful discussions, B. Greggio, R. Guntern, I. Mikolajewski and M. Surini for their expert technical assistance and P. Y. Vallon for the photographic work. We are also grateful to Dr N. Schaad for reading the manuscript. REFERENCES Adelman, J. P., Mason, A. J., Hayflick, J. S. and Seeburg, P. H. (1986). Isolation of the gene and hypothalamic cDNA for common precursor of gonadotropinreleasing hormone and prolactin release-inhibiting factor in human and rat. Proc. Natl. Acad. Sci. U.S.A. 83, 179-183. Bjartell, A., Ekman, R. and Peng, Lob, Y. (1990). Biosynthesis and processing of delta sleep-inducing peptide-like precursors in primary cultures of mouse anterior pituitary cells. Eur. J. Biochem. 190, 13l-137. Borb61y, A. A. and Tobler, I. (1989). Endogenous sleeppromoting substances and sleep regulation. Physiol. Rev. 69, 605-670. Buscail, L., Cauvin, A., Gourlet, P., Gossen, D., De Neef, P. Rath6, J., Robberecht, P., Vandermeers-Piret; M.-C., Vandermeers, A. and Christophe, J. (1990). Purification and amino acid sequence of vasoactive intestinal peptide, peptide histidine isoleucinamide (1-27) and secretin from small intestine of guinea pig. Biochem. Biophys. Acta 1038, 355 359. Charnay, Y., Bouras, C., Vallet, P. G., Golaz, J., Guntern, R. and Constantinidis, J. (1989). Immunohistochemical colocalization of delta sleep-inducing peptide and luteinizing hormone-releasing hormone in the rabbit brain neurones. Neuroscience 31+ 495 - 505. Charnay, Y., Leger, L., Golaz, J., Sallanon, M.. Vallet, P+ G., Guntern, R., Bouras, C., Constantinidis, J., Jouvet, M. and Tissot, R. (1990). lmmunohistochemical mapping of delta sleep-inducing peptide in the cat brain and hypophysis. Relationships with the LHRH system and corticotropes. J. Chem. Neuroanat. 3, 397 412.

M o n o c l o n a l antibodies against DSIP Constantinidis, J., Bouras, C., Guntern, R., Tabam C. H. and Tissot, R. (1983). Delta sleep-inducing peptide in Ihc rat brain: An immunohistochemical microscopic study. ,\europsychohiol. 10, 94 10. Drouin. J. and Goodman, H. M. (1980). Most of the coding region of rat ACTH, 13LPH precursor gene lacks intervening sequences. Nature 288, 610 613. Ekman, R.. Bjartell, A., Ekblad, E. and Sundler. F. 11987L hnmunoreactivc delta sleep-inducing peptide m pituitary adrenocorticotropin/alpha-melanolropin cells and adrenal medullary cells of the pig. .Vuuroemlocrimdo