- Email: [email protected]
Serotonin transporter antibodies: production, characterization, and localization in the brain
Molecular Brain Research 43 Ž1996. 267–278
Research report
Serotonin transporter antibodies: production, characterization, and localization in the brain Feng C. Zhou b
a,)
, Yi Xu b, Sharon Bledsoe a , Renee Lin c , Mark R. Kelley
b
a Department of Anatomy and Medical Neurobiology Program, Indiana UniÕersity School of Medicine, Indianapolis, IN 46202, USA Department of Pediatrics Section of Pediatric Endocrinology, and Biochemistry and Molecular Biology, Herman B. Wells Center for Pediatric Research, Indiana UniÕersity School of Medicine, Indianapolis, IN 46202, USA c Department of Biochemistry and VA Hospital, Indiana UniÕersity School of Medicine, Indianapolis, IN 46202, USA
Accepted 9 July 1996
Abstract Serotonin Ž5-HT. transporter, the mechanism for 5-HT high affinity uptake, is the essential component for the termination of 5-HT transmission. In order to identify transporter sites on 5-HT neurons or on other 5-HT uptaking cells, three rabbit antisera against cocaine sensitive-serotonin transporter Ž5-HTT. were produced. Antisera 5-HTT55 Žagainst amino acid sequence 55–68 in cytoplasmic N-terminal. and 5-HTT315 Žagainst amino acid sequence 315–325, a 3rd external loop peptide. were produced against synthetic multiple-antigenic peptides ŽMAP.. Antiserum 5-HTTN was produced against a fusion protein of the first 71 amino acids of N-terminal peptide expressed in recombinant DNA transformed bacteria. SDS-PAGErWestern blots indicate that 5-HTT55 and 5-HTT315 recognized bands of 74 and 64 kDa in rat brains with densities in the order of cortexG hippocampus) cerebellum, but not in liver, or muscle. The 5-HTTN recognized the fusion protein expressed in the bacteria, and the 64 kDa band with a similar density profile in the rat brain regions, and negative staining in liver and muscle. The immunocytochemistry of all three antisera revealed 5-HTT-immunostaining Ž5-HTT-im. in a pattern similar to 5-HT fiber distribution. 5-HTT55 and 5-HTT315 stainings were punctate in appearance; while 5-HTTN outlined the fibers in the 5-HT fiber areas, and neurons in raphe but not in substantia nigra or locus ceruleus. The preimmune serum and immune serum preabsorbed with 5-HTTN showed negative or diminished staining. Specific neurotoxin, 5,7-dihydroxytryptamine lesion removed all of the 5-HTTN fibers from the injection site, indicating that 5-HTTN-im fibers are 5-HT fibers in nature. Our study indicates that the three antibodies we produced recognize various domains of the 5-HTT. Our 5-HTT antibodies could be used as new markers of 5-HT fibers, and are particularly useful for the study of the plasticity of 5-HT fibers free of the complications involved with 5-HT content. Keywords: Site specific peptide; Multiple antigenic peptide; Polymerase chain reaction; Immunocytochemistry; 5,7-Dihydroxytryptamine; Fusion protein; Dorsal raphe
1. Introduction Serotonin Ž5-hydroxytryptamine; 5-HT. transporter Ž5HTT. plays an important role in the reuptake of 5-HT and subsequently terminates 5-HT transmission w13,20x. Imbalances in 5-HT transmission have been proposed to be involve in mood and personality disorders, obsessive– compulsive disorders, anxiety, insomnia, and eating disorders Žfor review see w6x., and alterations of 5-HT transmission are speculated to underlie many drug Že.g., cocaine, )
Corresponding author at: Department of Anatomy, Indiana University School of Medicine, 635 Barnhill Drive, Indianapolis, IN 46202, USA. F a x : q 1 Ž 3 1 7 . 2 7 4 - 4 9 3 4 , o r 2 7 8 - 2 0 4 0 ; E - m a il: [email protected]
LSD. and alcohol abuses w17,31x. 5-HTT plays a pivotal role in the presynaptic regulation of 5-HT transmission. 5-HTT is itself a target for many drugs of abuse such as cocaine, 3,4-methylenedioxymethamphetamine ŽMDMA, ‘ecstasy’. and methamphetamine w18,25,30x. Blocking 5-HT uptake at one of the 5-HTT sites prolongs 5-HT transmission in the synaptic junction, which is one of the most effective ways of alleviating 5-HT deficiency disorders, such as depression and alcoholism. Several structural related chemicals labeled as Selective Serotonin Reuptake Inhibitors ŽSSRI., were developed on 5-HTT for clinical use in 5-HT related disorders. 5-HTT cDNAs have recently been cloned in rat Žfrom brain w4x, from basophilic leukemia cells w11x., human Žfrom brains w15x, from placenta w24x, from platelet w16x.,
0169-328Xr96r$15.00 Copyright q 1996 Elsevier Science B.V. All rights reserved. PII S 0 1 6 9 - 3 2 8 X Ž 9 6 . 0 0 2 0 9 - 4
268
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
and Drosophila w8,9x, and are highly conserved among species. In the rat and human species; each 5-HTT cDNA codes approximately 630 amino acids and shares more than 90% common sequence. 5-HTT proteins Žin human and rats. are commonly featured with 12 transmembrane domains, cytoplasmic N- and C-terminals, glycosylated sites in the second external loop, a potential kinase C phosphorylation site in the N- and C-terminals, and a leucine zipper motif in the second transmembrane domain. The N- and C-terminals and the third external loop are the least conserved regions and have been used for discrimination among 5-HTTs ŽFig. 1..
Three methods have been used to detect 5-HTT sites and 5-HT uptake activity: Ža. 5-HT uptake has been demonstrated by high affinity uptake of radioactive 5-HT through synaptosomes derived from tissue homogenates; Žb. distribution of 5-HTT uptake sites have been indicated by autoradiography of specific affinity ligand binding with 5-HT, citalopram, paroxetine, b-CIT, and RTI-55 w26x; Žc. the existence of 5-HTT has also been indicated by the detection of 5-HTT mRNA w4x. 5-HT uptake has been found in such selective tissue types as brain w1x, lymphocytes w10x, placenta chorionic brush border w2,7x, blood platelet w19x, adrenal medulla w12x, and small intestine
Fig. 1. The three specific antigen peptides in rat 5-HT transporter. The 5-HTT55 and 5-HTT315 antisera were made against the synthetic octomer peptide in the multiple antigenic peptide ŽMAP. forms. The 14-amino acid 5-HTT55 begins at 55th amino acid sequence located in the cytoplasmic N-tail, while the 11-amino acid 5-HTT315 begins at 315th amino acid sequence located in the 3rd external loop. The 5-HTTN antibody is made against a fusion protein with a 71-amino acid epitope specific and highly antigenic located in the first inner cytoplasmic N-tail which contains a putative phosphorylation protein kinase C binding cite ŽS or T.. All peptides are distal to the two glycosylation sites ŽN. in the second extracellular loop. The chance of a glycogen structure to hinder antibody binding to our peptide antigens is slim. These peptides all have greater than 80% homology with the human sequence. The 5-HTT55 MAP antigen is shown in the insert. The Multiple Antigenic Peptide ŽMAP. is an octomer polypeptide with a unique oligopeptide sequence which is linked by the core cysteine. The MAP provides sizeable molecular mass and does not compromise the specificity of the sequence as it increases in size. It also presents eight epitope copies.
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
w5,28x. However, the study of 5-HTT cell biology and the cellular distribution would not be achievable without direct visualization of the transporter protein w23x. In this study, three antibodies immunospecific to 5-HTT protein were produced by immunization against three sitespecific 5-HTT peptides, two against multiple antigenic peptides ŽMAP. form, one against fusion protein Žsee Section 2.. The antibodies were characterized by immunoblotting, immunocytochemistry, compared to the known distribution of 5-HT neurons and fibers, and verified by immunostaining after neurotoxic lesion. We produced and characterized these antibodies hoping to provide additional and unique markers to study the neuroanatomy of 5-HTT. This report indicates that these antisera can be used for detecting a specific 5-HTT transmembrane domain. 5HTTN antiserum can be used as a new marker for 5-HT neurons and their extensive fibers, and has the potential of detecting 5-HTT plasticity independent of 5-HT content.
2. Materials and methods 2.1. Synthetic multiple antigenic peptides 5-HTT55 and 5-HTT315 The multiple antigenic peptides ŽMAP. were made in the Molecular and Biochemical Technology Facilities at Indiana University School of Medicine. The selected peptides were named according to the number of the first amino acid in their coding sequences. 5-HTT315 has 11 residues beginning at position 315 of the 5-HTT coding sequence; it is in the third external loop. The 5-HTT315 was made as a multiple antigenic peptide ŽMAP, 8 copies of peptides branch-linked by lysine in the core.. 5-HTT55 is a 14-residue peptide; it is proximal to the first transmembrane domain near the N-terminal and is located in the cytoplasm. 5-HTT55 was made in MAP form ŽFig. 1, w22x. Žtotal Mr approximately 17 000.. MAP is multiple antigenic, and the octomer size is large enough to be immunogenic; no carrier protein coupling was used. The sequences and key characters of the two peptides are shown in Fig. 1. 2.2. 5-HTTN fusion protein 2.2.1. The construction of PGEX-5HTTN Two oligo-primers Ž5X-GATCGTAGGATCCCCTTGAATTCTCAG and 3X GCGCTCTGAATTCTATTGGCGAATCTC, containing added BamHI and EcoRI sites respectively. flanking the first 213 bp of the N-terminal coding region of the serotonin transporter were synthesized at the Wells Center for Pediatric Research Oligonucleotide Synthesis Facility. These oligonucleotides were used in a standard PCR reaction using 1 m g of the 5-HTT cDNA as a template Žfrom Dr. R.D. Blakely. and 25 cycles. The
269
PCR products were purified, digested with the restriction enzymes, BamHI and EcoRI, repurified and ligated into pGEX3X, a glutathione S-transferase ŽGST. fusion vector ŽPharmacia.. Competent HB101 bacteria cells were transformed with the ligation reactants and transformants screened for inserts using the colony PCR procedure. Potential positive clones were over-expressed by isopropyl-b-D-thiogalactopyranoside ŽIPTG. induction and proper protein expression verified. Furthermore, the clone used for antigen production, hereafter called PGEX5HTTN, was sequenced to confirm correct reading frame and lack of mutations. 2.2.2. OÕer-expression of the PGEX-5HTT construct and antigen purification The overexpression of the PGEX-5HTT was performed as previously described w29x. Briefly, bacterial cultures Ž5 ml. containing the PGEX-5HTT fusion construct were grown overnight at 378C in Luria broth ŽLB. plus 100 m grml ampicillin. Overnight cultures were diluted 1:10 in fresh, pre-warmed Ž378C. LB medium supplemented with the appropriate antibiotics and grown for 1 h at 378C. Expression of the PGEX-5HTT fusion protein was induced by the addition of IPTG to a final concentration of 0.1 mM and growing the cells for an additional 3 h at 378C. Cells were harvested by centrifugation at 1000 = g for 10 min and washed once with PBS, pH 7.4. Packed cells were resuspended in 2 ml of PBS and lysed by mild sonication Žthree 15-s bursts. on ice. The insoluble proteins were pelleted by centrifugation at 10 000 = g for 10 min at 48C; the PGEX-5HTTN fusion protein was soluble. The soluble fusion protein in the supernatant was mixed with 1% Triton X-100 and loaded onto a glutathione Sepharose 4B column pre-equilibrated with PBS plus 1% Triton X-100. Binding of the PGEX-5HTTN protein was carried out within the column overnight on a nutator at 48C. The column was subsequently washed with 20 column volumes of PBS containing 1% Triton X-100 ŽTBST.. Following the wash, the fusion protein was eluted with 50 mM Tris, pH 7.5, containing 10 mM glutathione. The sample was collected from the column and analyzed by SDS–polyacrylamide gel electrophoresis ŽSDS-PAGE. and Western blotting. This column purified protein was used for antibody production and for biochemical analysis. This procedure resulted in a single protein band upon SDS-PAGE analysis. 2.3. Antiserum production and purification For 5-HTTN antigen preparation, the fusion protein was further purified by transferring the protein onto nitrocellular membrane ŽNC., and stained with phenol red. The NC at the single band of 36 kDa Mr was cut out, sealed in dialysis tubing with TBS elution buffer, and eluted out of the polyacrylamide gel in a submarine electrophoresis
270
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
chamber for 20–30 min. The elute was then dialyzed overnight against distilled water. This final dialyzed elute was used as antigen for antibody production. The 5-HTT315, 5HTT55, and 5-HTTN antigens were used to individually immunize rabbits over three months to produce antibodies in our Laboratory Animal Research Facilities or at HTI, Ramona, CA, using standard protocol. In brief, 1 ml of antigen complex in PBS was mixed with an equal amount of complete Freund’s adjuvant to obtain a 0.5 mgrml mixture. This solution was injected subcutaneously Žtotal 300 m l. on the back of pathogen-free adult,
female white rabbits under sedation. Booster injections were given with incomplete adjuvant at 3–4 week intervals. The first bleed was obtained 9–10 weeks after the first injection, the second bleed at 11 weeks, and the final bleed at 12 weeks. The 5-HTT315 and 5-HTTN antisera were further affinity purified by incubating the antisera with 5-HTT315-MAP peptides or PGEX-5HTTN fusion protein respectively on strips of nitrocellulose membrane overnight in a 15 ml conical tube, washed three times, 10 min each, with TBST, and eluted with 250 m l antibody elution buffer for 30 min
Fig. 2. Western blotting of brain protein immunoprecipitated by 5-HTT315 Župper panel. and 5-HTT55 Žlower panel. antisera in 10% SDS-PAGE. The 5-HT315 recognizes the mobility of species in 92, 74 and 64 kDa. The 92 kDa and 74 kDa protein could be glycosylated 5-HTT, with 64 kDa being a fragment of deglycosylated 5-HTT. The 5-HT55 recognizes 74 and 64 kDa species. Both antisera quantitatively distinguish the low level of 5-HTT in the least 5-HT innervated cerebellum, to a high abundance in 5-HT dense brain regions, such as brainstem, cortex, and hippocampus. The numbers on the top of lane represent different animals. No binding was seen with liver or muscle Žnot shown.. The numbers on top of the lane represent different animals.
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
271
X Fig. 3. Ža. The construct p GE X.5-HTTN DNA Žlane 3. digested with enzyme BamHI and EcoRI yields a predicted size of 0.2 kb fragment Ž5 first 213 bp of coding region. and a large fragment which has a size equal to the vector Žlane 1. in 1.5% agarose gel. The p GE X.5-HTTN DNA was verified with correct sequence in frame. Žb. The HB101 bacteria expressed fusion protein before ŽBacterial pellet. and after purification with GST-column or GST-resin tube. All detectable eluted proteins show a single band with predicted mobility between 43 and 35 kDa in 10% SDS-PAGE.
Fig. 4. Western blot analysis using affinity purified 5-HTTN antibody. Total protein extracts Ž20 m g. from cortex ŽCX., hippocampus ŽHIPP., cerebellum ŽCB., liver, muscle Žlanes 1 to 5., and GST-5-HTTN fusion protein Ž10 ng, lane 6. were run on a 10% SDS–polyacrylamide gel, blotted to nitrocellolose and reacted with purified 5-HTTN antibody. The antibody only recognized the GST-5-HTTN fusion protein Ž; 37 kDa. and a 64 kDa single band in the brain, but not protein in liver, nor in muscle.
272
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
ŽPierce Immunopure, Rockland, IL.. The antibody was aliquoted and stored in y808C. 2.4. Western blotting Frontal cortex, hippocampus, cerebellum, and raphe were dissected from normal rat brains and homogenized by sonication in 0.1 mM PBS. Proteins were separated on
12% SDS–polyacrylamide gels and transferred to nitrocellulose membrane w29x. Gels, Whatman filter paper, and membranes were soaked in electroblotting buffer Ž25 mM Tris-HCl; 193 mM glycine; 20% methanol. for 15 min prior to transferring. Proteins separated on SDS–polyacrylamide gels, were transferred to nitrocellulose sheets by electroblotting in a Transblot BioRad transfer apparatus in 25 mM Tris, 192 mM glycine, 20% methanol at 150 mA
Fig. 5. 5-HTT315 Ža. and 5-HTTN Žb. immunocytochemistry in the dentate gyrus of hippocampus. The punctate pattern of 5-HTT315 immunostaining Ž5-HTT315-im. is characteristic of this antibody. Punctate type 5-HTT315-im is shown in infragranule layer where 5-HT fibers accumulate. The 5-HTTN characteristically shows fiber type staining and more intensively stain in infragranular layer, and hippocampal CA regions. The 5-HTTN is a very sensitive antibody for the fiber staining. All illustrations are darkfield microscopy. Scale bars: a, b s 300 m m.
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
Ž70 V.. The transfer was carried out for 1 h at 48C. Following protein transfer, the membrane was blocked with Blotto Ž1 = TBST Ž10 = TBSTs 1.5 M NaCl; 100 mM Tris-HCl, pH 8.0; 0.5% Tween-20; 2% NP-40; 0.2% SDS.; 5% nonfat dry milk; 0.02% sodium azide. for 1 h at room temperature on a rotator. The 5-HTT antibody was diluted 1:1000 in Blotto and incubated with the membrane at 48C overnight. The membrane was then washed three times in 1 = TBST for 10 min at 48C. Detection of antigen–antibody complex was accomplished using chemiluminescence method ŽBoehringer Mannheim, Indianapolis, IN.. 2.5. 5,7-Dihydroxytryptamine lesion Animals under anesthesia were pretreated with desipramine Žto protect dopamine and norepinephrine fibers, 10 mgrkg, Merrill Labs, Cincinnati, OH. 45 min prior to the lesions to protect the dopaminernorepinephrine fibers. The serotonin neurotoxin 5,7-dihydroxytryptamine Ž5,7DHT, 4 m gr0.4 m l of 0.01% ascorbic normal saline. was unilaterally delivered by two injections Žeach 0.4 m l. into the hippocampus of rats Žcoordinates to Bregma: 2.5 mm posterior, 2.8 mm lateral, 3.7 mm below. to remove serotonin innervation in the hippocampal region. Five normal rats and 4 rats 7 days after the 5,7-DHT injections were
273
sacrificed, their brains removed, sectioned and immunocytochemically stained with 5-HTTN antiserum Žwhich is more prominent for fiber staining than 5-HTT55 or 5HTT315.. Alternate sections were used for 5-HT immunocytochemical staining. 2.6. Immunocytochemistry All animals were perfused intracardially with formaldehyde ŽReagent grade, Fisher Scientific. freshly made from 4% paraformaldehyde and 0.1 M phosphate-buffered saline ŽPBS. under deep ketamine cocktail Ž120 mgrkg. anesthesia. Their brains were removed, left in the same fixative overnight, and coronally sectioned at 40 m m with a vibratome for immunocytochemical staining w30,34x. The 5-HTT315, 5-HTT55, 5-HTTN, and control preimmune sera Žsee control below. were used to stain the normal brain tissue. The 5-HTTN serum was used to stain the 5,7-DHT lesion tissue. A pilot study was used to characterize the dilution profile and Triton-X100 concentrations Ž0, 0.03%, and 0.3%.. In general, a 1:2000–3000 dilution of 5-HTT antisera, and 0.05% Triton were used in the later process. Sternberger’s peroxidase-anti-peroxidase ŽPAP. indirect-enzyme method was used for staining w27x. The PAP reaction was performed with 0.003% H 2 O 2 and 0.05% 3X ,3-diaminobenzidine. The primary, secondary and
Fig. 6. Examples of 5-HTTN immunostaining Ž5-HTTN-im. in the 5-HT terminal regions in cingulate cortex. The 5-HTTN antiserum has greater sensitivity in detecting the 5-HTT in the brain. A characteristic fibrous staining pattern is shown in the brain regions where 5-HT terminals are located. The 5-HTT-im fibers share many similarities in fiber morphology, distribution pattern, and density in regions of the brain with the 5-HT-im fibers. Characteristic dense 5-HTTN-im fibers in layer I and II near the surface of the brain, less dense in layer III–IV, and dense layer V highly resemble the distribution of 5-HT fibers in the correspondent subregions Ža.. Few, if any fibers were stained in the preimmune serum with same dilution Žb.. Few fibers were stained with antigen-preabsorbed 5-HTTN serum control Žc.. Straight 5-HTT-im fibers are seen in cingulum bundle ŽCB. dorsal to corpus callosum ŽCC. in Ža. but not in Žb. or Žc.. All illustrations are darkfield microscopy. Scale bars: a, b, c s 300 m m.
274
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
marker antibodies were diluted with PBS containing 0.05% Triton X-100 and 1% normal sheep serum. The primary antibodies were incubated overnight; the second and third, for 1 h each. Three washes with PBS each for 5 min were used between each antibody. All the sections were Nisslcounterstained with methyl green to reveal non-5-HTTN-im cells, and profile or brain structure. Control. Three sets of controls were used for immunocytochemistry, the preimmune serum of each antiserum, the 5-HTTN antiserum preabsorbed with fusion protein, and skipping of the primary antiserum. All controls were processed under the same conditions and in parallel with the 5-HTTN staining.
3. Results 3.1. 5-HTT55 and 5-HTT315 antisera In Western blots from 12% SDS–polyacrylamide gels and detection of antigen–antibody complexes using the
alkaline phosphatase method, 5-HTT55 and 5-HTT315 recognized rat frontal cortex, hippocampus, cerebellum ŽFig. 2., but not liver and muscle Žnot shown.. The apparent molecular masses were based on migration in the gel. 5-HTT315 recognized the mobility of species at 92, 74 and 64 kDa. The 5-HTT55 recognizes 74 and 64 kDa species. The possible glycosylation-rendered diversity in size is speculated in Section 4. The 5-HTTN has a single band at 64 kDa. The antisera quantitatively distinguished the low level of 5-HTT in the least 5-HT innervated cerebellum, to a high abundance in 5-HT dense brain regions, such as brainstem, cortex, and hippocampus. The apparent binding density was observed with cortex G hippocampus ) cerebellum ŽFig. 2.. This is consistent with the immunocytochemically stained 5-HT Ž5-HT-im. fiber density in these three regions. The brainstem was also observed with high density of 5-HTT55 binding as compared with the above three regions. Immunocytochemical staining using 5-HTT55 and 5HTT315 in the rat brain revealed similar punctate type staining along fibers in many forebrain regions with den-
Fig. 7. 5-HTTN-im neurons in the dorsal raphe nucleus ŽDRN. and median raphe nucleus ŽMRN.. 5-HTTN-im neurons distributed in a distinct central group Žarrows. and two lateral wings ŽL, arrowheads. of the dorsal raphe Ža, b., and in midline neuronal groups in median raphe Žcrossed arrows, c, d.. Many 5-HTTN-im punctates are seen surrounding the neurons and outline the somatodendritic profiles. The immunocytochemical DAB signals are brown in color; the background cells are green in color revealed by Nissl-counterstaining with methyl green. mlf, medial longitudinal fascicularis. All illustrations are brightfield microscopy. Scale bars: Ža,c. s 250 m m; Žb,d. s 20 m m.
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
sity and distribution patterns similar to 5-HT-im fibers. 5-HTT55 with 0.3% Triton-X was the optimal staining condition, while the 5-HTT315 was not affected by the Triton-X condition. Immunostaining using 5-HTT315 is shown in the hippocampus ŽFig. 5a.; the 5-HTT315-im is distributed predominately in the infragranular regions and the hippocampal fissure. Perisomatic staining and staining along membrane of primary dendrites was seen in the septal nucleus Žnot shown.. This type of staining is seen in 5-HT-im fibers in the septal nucleus of rats. The above observations indicate that antisera, 5-HTT55 and 5-HTT315 Ža. recognize a protein in the brain with highrlow density in brain regions of 5-HT, but not in liver
275
or muscle, and Žb. immunocytochemically stained a diffuse pattern Žin punctate form. along 5-HT-like fibers. 3.2. 5-HTTN antiserum PCR amplification of the 5-HTT cDNA with primers of a 210 bp regions at the amino end resulted in a single band which was digested with BamHI–EcoRI and cloned into pGEX.3X Ž4.9 kb. giving a correct 5.1 kb construct ŽFig. 3a.. The correct sequence of the construct and confirmation of the correct reading frame for the coded protein was confirmed, and hereafter called pGEX-5-HTTN. Three transformants with enhanced expression of fusion
Fig. 8. The effect of 5,7-DHT on the 5-HTTN-im fibers in the brain. Under desipramine pretreatment, the 5,7-DHT specifically damage the 5-HT fibers, but not dopamine or norepinephrine. The 5,7-DHT injection in the dorsal hippocampus removed 5-HTTN-im fibers in the entire ipsilateral side Žb. and majority of the contralateral side. A contrast was seen as compared with the normal hippocampus of non-lesioned animals Ža.. Degeneration extended rostrally to the anterior septum Žc. where fibers were also removed ipsilateral to the lesion Žstars. as compared with that of contralateral side. Adjacent to the degenerated anterior septal area where fibers survived, degeneration profile Žarrows. is seen with large degeneration stumps Žd, arrows.. a, b, c: darkfield microscopy; d: brightfield microscopy. The immunocytochemical DAB signals are brown in color; the background cells are green in color revealed by Nissl-counterstaining with methyl green. Scale bars: a, b s 500 m m; c s 500 m m; d s 20 m m.
276
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
protein with IPTG were obtained. They were selected for large scale production, and the purified protein revealed a distinct 37 kDa band as shown in Fig. 3b. Western blotting of 5-HTTN showed the fusion protein expressed in bacteria as a single 37 kDa band ŽFig. 3b and Fig. 4.. Western blotting analysis of rat frontal cortex, hippocampus, and cerebellum showed a single 64 kDa band. No staining was seen in liver or muscle protein ŽFig. 4.. Immunocytochemistry demonstrated that the preimmune, the 5-HTTN antibody-preabsorption, and the primary antibody-skipping controls, all stained negatively in the brain ŽFig. 6.. The antiserum after absorption against the fusion protein greatly decreased the immunostaining. The preabsorbed antiserum revealed very light and faint version of the non-absorbed antiserum, indicating a high titer of antiserum and remaining 5-HTTN antiserum after purification. The 1:3000 dilution of 5-HTTN antiserum immunostaining Ž5-HTTN-im. revealed a distinct morphology that is characteristics of 5-HT fibers ŽFig. 5b, hippocampus; Fig. 6a, cingulum cortex.. Curving, branching fine fibers with many fine varicosities were found in 5-HT fiber terminal regions, while thick, straight, and unbranching fibers were found in the 5-HT fiber pathways such as cingulum bundle and fornix. Many distribution pattern previously observed in 5-HT fibers were identified by the 5-HTTN antibody. A staining pattern of 5-HTT-im highly resembling that of 5-HT fibers in cingulum cortex is shown in Fig. 6. The 5-HTTN-im is densest in layers I and III and lower in layers II, IV–VI. Below layer VI and above the corpus callosum, the perpendicular 5-HTT-im fibers in cingulum bundle can clearly be seen. As we predicted, the 5-HTTN can also detect 5-HT neurons in raphe ŽFig. 7.. In contrast, no dopamine neurons in nigrarventral tegmental area, norepinephrine neurons in locus ceruleus, nor GABA neurons in any known GABA neurons distributed regions were detected by 5-HTTN Žnot shown.. The 5,7-DHT lesion in the hippocampus removed all 5-HTTN-im fibers in the ipsilateral and adjacent contralateral hippocampi ŽFig. 8.. Alternate sections of 5,7-DHT lesioned brains also stained negatively with 5-HT antiserum Žnot shown.. 5-HTTN-im fibers were greatly reduced with a degeneration profile in the brain regions distal to the injection site. Typical degeneration stumps, enlarged twisted and fragmented fibers were shown in frontal cortex millimeters away from the injection site in the hippocampus ŽFig. 8..
banks, especially specific against the known catecholamine and GABA transporter, Žb. high antigenic index w14x, and Žc. hydrophilic regions of the 5-HTT. The size of antigens varies among the three antisera. The 5-HTT55 and 5HTT315 antisera were made against MAP antigens. Each octomer-peptide has a molecular mass of approximately 13–15 kDa. Without additional conjugation of large molecules, the immunogen is sufficient to produce antiserum for Western blotting and immunocytochemistry. The advantage of MAP form antigen is that it has eight copies of the antigenic sites as compared with the same molecular mass of a single peptide conjugated antigen, yet, it still preserves the specificity of the small peptide. The disadvantage of MAP produced antibodies is that they can only detect a small region of the endogenous protein, which runs the risk, during immunocytochemistry, of the small portion of the peptide being masked by the folding of the protein itself or the surrounding tissue molecules. This seems to be the case with the 5-HTT55 antiserum against an N-terminal peptide as it detected only punctate staining of 5-HTT-im fibers, while the 5-HTTN antiserum against similar regions of 5-HTT but with greater peptide regions of 5-HTT detected the full fiber profile. However, the 5-HTTN antiserum which reacted against a 37 kDa fusion protein has a greater molecular mass than the other two. In general, the 5-HTTN provides a good morphological detection of 5-HTT fibers. In addition, it seems that the 5-HTT315 does not require Triton-X100 treatment, while the 5-HTTN and 5-HTT55 are more immunoreactive in the presence of Triton X-100. 5-HTT315 recognizes the extracellular loop while 5-HTT 55 and 5-HTTN recognize the intracellular loop region. Therefore, the use of detergent may render the accessibility of antibodies to the antigens across the membrane. The three antibodies against different domains of 5-HTT can be used for biochemical characterization of the 5-HTT. The sequence specific and high antigenic 71-amino acid N-terminal peptide is located in the first inner cytoplasmic loop which contains a putative protein kinase C binding cite ŽS.. This peptide is distal to the two glycosylation sites ŽN. in the second extracellular loop. An alteration in glycogen structure to hinder the antibodies binding to our peptide antigen is slim. This peptide is approximately 80% homologous with the human 5-HT serotonin sequence in this region. The 5-HTT315 antiserum provides an alternative antibody to detect the third extracellular loop, when characterization of 5-HTT transmembrane domains is desired. 4.2. Characterization of the antibodies
4. Discussion 4.1. The antigens The three antigenic epitopes were selected based on Ža. specificity against a known published sequence in the gene
In Western blots of the fusion protein, a single band was detected by our crude and purified 5-HTTN antiserum. The 5-HTT antibody recognized the fusion protein induced from the recombinant expression construct. In Western blots of the brain membrane preparation, SDS-PAGE indi-
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
cated that the mobility of 92 kDa is only seen with 5-HTT315 antiserum, the 74 kDa band is seen with the 5-HTT55 and 5-HTT315 antisera, while the 64 kDa protein is commonly observed in all 5-HTT55, 5-HTT315, and 5-HTTN antisera, in the brain. The 92 kDa and 74 kDa proteins could be glycosylated 5-HTT, and the 64 kDa commonly existing as a deglycosylated fragment of 5-HTT. The 92 kDa species is similarly detected by CT-2 antibody made against the 5-HTT C-terminal peptide, which detects an approximately 60 kDa molecule after enzymatic deglycosylation w23x. None of the three antisera recognized liver or muscle membranes. The 5-HTTN detected a single protein band and showed suitable staining for the immunocytochemical detection of the 5-HTT in the tissue. The 5-HTT55 and 5-HTT315 can be used when detection of a special domain of the 5-HTT is desired. Together, three pieces of immunocytochemical evidence support the specificity of the antisera. The staining of 5-HTT315, 5-HTT55, and 5-HTTN antisera consistently detected the 5-HT fibers observed in many brain regions. Three regions, cingulum cortex, hippocampus, and raphe are shown in Figs. 5–7. The distribution density of 5-HTT in many brain regions including the cingulum cortex, hippocampus and raphe indicated by the 5-HTTN antiserum is very similar to that of 5-HT innervation density. The distribution of 5-HT-im Žfrom our previous and others observation. and 5-HTTN fibers closely correlates to the detail cortical layer Žlayers I, II versus III–V in cingulum cortex. and many fiber bundles Žmedial forebrain bundle, stria terminalis, fornix, and fimbria.. The characteristic 5-HT immunocytochemical fiber morphologies-fine fibers highly decorated with varicosity, curving and twisting in the terminal regions, but short straight and thick in the pathway such as cingulum bundle, are also revealed by the 5-HTTN-im fibers. Furthermore, the 5-HTTN-im staining in the cell bodies in three ascending raphe nuclei, but not in substantia nigra, or locus ceruleus lends further support of the specific detection of the 5-HT neurons. Finally, the strongest evidence was derived from the study of 5-HTspecific neurotoxin lesion. The 5,7-DHT toxin under the protection of desipramine will specifically cause degeneration of 5-HT fibers w3,32,33,21x. The 5,7-DHT lesion removed all 5-HTTN fibers at the injection site, and caused a 5-HTTN-im degeneration profile within a millimeter distance from the center of the injection. The adjacent sections at regions where 5-HTTN-im disappeared were also 5-HT-im negative. This suggests that the 5-HTTN-im fibers are 5-HT in nature, unless a new large population of fibers are also 5,7-DHT sensitive and transport 5,7-DHT across the membrane. 4.3. Significance The 5-HTTN antibody serves as a good marker for 5-HT fibers in the case where 5-HT is depleted and 5-HT fibers are still intact. This maker is capable of distinguish-
277
ing depletion from degeneration. The 5-HTTN antibody should provide a practical application to distinguish the effects of abusive drugs which have different levels of effects on 5-HT neurons. In addition, 5-HTT seems intensely distributed in wide regions of the 5-HT fibers, and the degeneration profiles can be detected by the 5-HTTN antiserum. This antibody provides a very definitive marker for the degeneration of the 5-HT fibers. In conclusion, antisera against three specific regions of the serotonin transporter recognized antigen on fiber structure that is highly associated with 5-HT fibers in the brain. The slower migrating species in the SDS-PAGE may be various glycosylation forms of the transporter protein. The fast migration species is common to all three antisera, and is the unique single protein band when 5-HTTN antiserum was used. The three antisera provide options to identify different regions of the serotonin transporter. The 5-HTTN antiserum is suitable for immunocytochemically detecting minute amounts of serotonin transporter. The characteristic distribution and the degeneration in response to specific 5,7-DHT lesion of 5-H TTN-im fibers indicates that the 5-HTTN-im fibers are 5-HT fibers in nature. The 5-HTTN is a good marker to identify 5-HT fibers independent of 5-HT content. In addition to the use of the 5-HTTN antiserum in basic neurobiology research of 5-HT transporter, 5-HTTN antiserum has the ability to distinguish degeneration versus depletion, which will have many applications for verifying toxic effects of abusive drugs. For example, it will be possible to determine critical thresholds of drug dosage levels which may be maintained so that they will lead to depletion but not degeneration in experimental animals.
Acknowledgements We would like to thank Dr. R.D. Blakely for providing the 5-HTT cDNA template, the Wells Center oligonucleotide synthesis facility which is supported by the Riley Memorial Association for synthesis of the oligonucleotides used for the construction of the 5-HTT fusion protein, and Ms. Denise Scroggins, Laura Miller and Teresa Wilson for their technical assistance. This study is supported by NIH MH 50602 to F.C.Z. R.L. is supported by NIH AA06991. M.R.K. is supported by NIH RR09884 and Riley Memorial Association Research Grant.
References w1x Austin, M.C., Bradley, C.C., Mann, J.J. and Blakely, R.D., Expression of serotonin transporter messenger RNA in the human brain, J. Neurochem., 62 Ž1994. 2362–2367. w2x Balkovetz, D.F., Tiruppathi, C., Leibach, F.H., Mahesh, V.B. and Ganapathy, V., Evidence for an imipramine-sensitive serotonin transporter in human placental brush-border membranes, J. Biol. Chem., 264 Ž1989. 2195–2198.
278
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
w3x Baumgarten, H.G., Jenner, S. and Klemm, H.P., Serotonin neurotoxins: recent advances in the mode of administration and molecular mechanism of action, J. Physiol. (Paris), 77 Ž1981. 309–314. w4x Blakely, R.D., Berson, H.E., Fremeau, R.T. Jr., Caron M.G., Peek, M.M., Prince, H.K. and Bradley, C.C., Cloning and expression of a functional serotonin transporter from rat brain, Nature, 354 Ž1991. 66–70. w5x Bledsoe, S., Chong, S.K.F. and Zhou, F.C., Immunocytochemical detection of serotonin and serotonin transporter in the human small intestine, Soc. Neurosci. Abstr., 21 Ž1995. 540.6. w6x Coccaro, E.F. and Murphy, D.L. ŽEds.., Serotonin in Major Psychiatric Disorders, American Psychiatric Press, Washington, 1990. w7x Cool, D.R., Leibach, F.H. and Ganapathy, V., Modulation of serotonin uptake kinetics by ions and ion gradients in human placental brush-border membrane vesicles, Biochemistry, 29 Ž1990. 1818– 1822. w8x Corey, J.L., Quick, M.W., Davidson, N., Lester, H.A. and Guastella, J. a cocaine-sensitive Drosophila serotonin transporter: cloning, expression, and electrophysiological characterization, Proc. Natl. Acad. Sci. USA, 91 Ž1994. 1188–1192. w9x Demchyshyn, L.L., Pristupa, Z.B., Sugamori, K.S., Barker, E.L., Blakely, R.D., Wolfgang, W.J., Forte, M.A. and Niznik, H.B., Cloning, expression, and localization of a chloride-facilitated, cocaine-sensitive serotonin transporter form Drosophila melanogaster, Proc. Natl. Acad. Sci. USA, 91 Ž1994. 5158–5162. w10x Faraj, B.A., Olkowski, Z.L. and Jackson, R.T., Expression of a high-affinity serotonin transporter in human lymphocytes, Int. J. Immunopharmacol., 16 Ž1994. 561–567. w11x Hoffman, B.J., Mezey, E. and Brownstein, M.J., Cloning of a serotonin transporter affected by antidepressants, Science, 254 Ž1991. 579–580. w12x Holzwarth, M.A., Sawetawan, C. and Brownfield, M.S., Serotoninimmunoreactivity in the adrenal medulla: distribution and response to pharmacological manipulation, Brain Res. Bull., 13 Ž1984. 299– 308. w13x Kuhar, M.J., Roth, R.H. and Aghajanian, G.K., Synaptosome from fore-brains of rats with midbrain raphe lesions: selective reduction of serotonin uptake, J. Pharmacol. Exp. Ther., 181 Ž1972. 36–45. w14x Kyte, J. and Doolittle, R.F., A simple method for displaying the hydropathic character of a protein, J. Mol. Biol., 157 Ž1982. 105– 132. w15x Lesch, K.P., Wolozin, B.L., Ester, H.C., Murphy, D.L. and Diederer, P., Isolation of a cDNA encoding the human brain serotonin transporter, J. Neural Transm., 91 Ž1993. 67–73. w16x Lesch, K.P., Wolozin, B.L., Murphy, D.L. and Diederer, P., Primary structure of the human platelet serotonin uptake site: identity with the brain serotonin transporter, J. Neurochem., 60 Ž1993. 2319–2322. w17x McBride, W.J., Murphy, J.M., Lumeng, L. and Li, T.K., Serotonin and alcohol preference wreviewx, Recent DeÕ. Alcoholism, 7 Ž1989. 187–209. w18x Molliver, M.E., Berger, U.V., Mamounas, L.A., Molliver, D.C., O’Hern, E. and Wilson, M.A., Neurotoxicity of MDMA and related compounds: anatomic studies wReviewx, Ann. NY Acad. Sci., 600 Ž1990. 649–661. w19x Moret, C. and Briley, M., Platelet 3 H-paroxetine binding to the
w20x
w21x
w22x
w23x
w24x
w25x
w26x
w27x w28x
w29x
w30x
w31x
w32x
w33x
w34x
serotonin transporter is insensitive to changes in central serotonergic innervation in the rat, Psychol. Res., 38 Ž1991. 97–104. O’Reilly, C.A. and Reith, M.E.A., Uptake of w 3 Hxserotonin into plasma membrane vesicles from mouse cerebral cortex, J. Biol. Chem., 263 Ž1988. 6115–6121. Patel, T.D., Azmitia, E.C. and Zhou, F.C., Increased 5-HT1A receptor immunoreactivity in the rat hippocampus following 5,7-dihydroxytryptamine lesions in the cingulum bundle and fimbria-fornix, BehaÕ. Brain Res., 73 Ž1996. 319–323. Posnett, D.N., McGrath, H. and Tam, J.P., A novel method for producing anti-peptide antibodies. Production of site-specific antibodies to the T cell antigen receptor b-chain, J. Biol. Chem., 263 Ž1988. 1719–1725. Qian, Y., Melikian, H.E., Rye, D.B. and Levey, A.I., Blakely, R.D., Identification and characterization of antidepressant-sensitive serotonin transporter proteins using site-specific antibodies, J. Neurosci., 15 Ž1995. 1261–1274. Ramamoorthy, S., Bauman, A.L., Moore, K.R., Han, H., Yang-Feng, T., Chang, A.S., Ganapathy, V. and Blakely, R.D., Antidepressantand cocaine-sensitive human serotonin transporter: molecular cloning, expression, and chromosomal localization, Proc. Natl. Acad. Sci. USA, 90 Ž1993. 2542–2546. Ricaurte, G.A. and McCann, U.D., Neurotoxic amphetamine analogues: effects in monkeys and implications for humans wReviewx, Ann. NY Acad. Sci., 648 Ž1992. 371–382. Scheffel, U., Kim, S., Cline, E.J. and Kuhar, M.J., Occupancy of the serotonin transporter by fluoxetine, paroxetine, and sertraline: in vivo studies with w 125 IxRTI-55, Synapse, 16 Ž1994. 263–268. Sternberger, L.A., Immunocytochemistry, 2nd edn., Wiley, New York, 1979, pp. 104–169. Wade, P.R., Tamir, H. and Kirchgessner, A.L., Gershon, M.D., Analysis of the role of 5-HT in the enteric nervous system using anti-idiotopic antibodies to 5-HT receptors, Am. J. Physiol., 266 Ž1994. G403–416. Wilson III, D.M., Deutsch, W.A. and Kelley, M.R., Drosophila ribosomal protein S3 contains an activity that cleaves DNA at apurinicrapyrimidinic sites, J. Biol. Chem., 269 Ž1994. 25359– 25364. Zhou, F.C. and Bledsoe, S., Methamphetamine causes rapid varicosis, perforation and definitive degeneration of serotonin fibers: an immunocytochemical study of serotonin transporter, NeuroscienceNet, V.1 Ž1996. Article a10009, Oct. 28, 1996; http:rrwww.neuroscience.com Zhou, F.C., Bledsoe, S., Lumeng, L. and Li, T.K., Reduced serotonergic immunoreactive fibers in the forebrain of alcohol-preferring rats, Alcohol. Clin. Exp. Res., 18 Ž1994. 571–579. Zhou, F.C. and Azmitia, E.C., Induced homotypic collateral sprouting of serotonergic fibers in the hippocampus of rat, Brain Res., 308 Ž1984. 53–62. Zhou, F.C. and Azmitia, E.C., Induced homotypic sprouting of serotonergic fibers in hippocampus. II an immunocytochemistry study, Brain Res., 373 Ž1986. 337–348. Zhou, F.C., Pu, C.F., Murphy, J.M., Lumeng, L. and Li, T.K., Serotonergic neurons in the alcohol preferring rat, Alcohol, 11 Ž1994. 395–403.
Research report
Serotonin transporter antibodies: production, characterization, and localization in the brain Feng C. Zhou b
a,)
, Yi Xu b, Sharon Bledsoe a , Renee Lin c , Mark R. Kelley
b
a Department of Anatomy and Medical Neurobiology Program, Indiana UniÕersity School of Medicine, Indianapolis, IN 46202, USA Department of Pediatrics Section of Pediatric Endocrinology, and Biochemistry and Molecular Biology, Herman B. Wells Center for Pediatric Research, Indiana UniÕersity School of Medicine, Indianapolis, IN 46202, USA c Department of Biochemistry and VA Hospital, Indiana UniÕersity School of Medicine, Indianapolis, IN 46202, USA
Accepted 9 July 1996
Abstract Serotonin Ž5-HT. transporter, the mechanism for 5-HT high affinity uptake, is the essential component for the termination of 5-HT transmission. In order to identify transporter sites on 5-HT neurons or on other 5-HT uptaking cells, three rabbit antisera against cocaine sensitive-serotonin transporter Ž5-HTT. were produced. Antisera 5-HTT55 Žagainst amino acid sequence 55–68 in cytoplasmic N-terminal. and 5-HTT315 Žagainst amino acid sequence 315–325, a 3rd external loop peptide. were produced against synthetic multiple-antigenic peptides ŽMAP.. Antiserum 5-HTTN was produced against a fusion protein of the first 71 amino acids of N-terminal peptide expressed in recombinant DNA transformed bacteria. SDS-PAGErWestern blots indicate that 5-HTT55 and 5-HTT315 recognized bands of 74 and 64 kDa in rat brains with densities in the order of cortexG hippocampus) cerebellum, but not in liver, or muscle. The 5-HTTN recognized the fusion protein expressed in the bacteria, and the 64 kDa band with a similar density profile in the rat brain regions, and negative staining in liver and muscle. The immunocytochemistry of all three antisera revealed 5-HTT-immunostaining Ž5-HTT-im. in a pattern similar to 5-HT fiber distribution. 5-HTT55 and 5-HTT315 stainings were punctate in appearance; while 5-HTTN outlined the fibers in the 5-HT fiber areas, and neurons in raphe but not in substantia nigra or locus ceruleus. The preimmune serum and immune serum preabsorbed with 5-HTTN showed negative or diminished staining. Specific neurotoxin, 5,7-dihydroxytryptamine lesion removed all of the 5-HTTN fibers from the injection site, indicating that 5-HTTN-im fibers are 5-HT fibers in nature. Our study indicates that the three antibodies we produced recognize various domains of the 5-HTT. Our 5-HTT antibodies could be used as new markers of 5-HT fibers, and are particularly useful for the study of the plasticity of 5-HT fibers free of the complications involved with 5-HT content. Keywords: Site specific peptide; Multiple antigenic peptide; Polymerase chain reaction; Immunocytochemistry; 5,7-Dihydroxytryptamine; Fusion protein; Dorsal raphe
1. Introduction Serotonin Ž5-hydroxytryptamine; 5-HT. transporter Ž5HTT. plays an important role in the reuptake of 5-HT and subsequently terminates 5-HT transmission w13,20x. Imbalances in 5-HT transmission have been proposed to be involve in mood and personality disorders, obsessive– compulsive disorders, anxiety, insomnia, and eating disorders Žfor review see w6x., and alterations of 5-HT transmission are speculated to underlie many drug Že.g., cocaine, )
Corresponding author at: Department of Anatomy, Indiana University School of Medicine, 635 Barnhill Drive, Indianapolis, IN 46202, USA. F a x : q 1 Ž 3 1 7 . 2 7 4 - 4 9 3 4 , o r 2 7 8 - 2 0 4 0 ; E - m a il: [email protected]
LSD. and alcohol abuses w17,31x. 5-HTT plays a pivotal role in the presynaptic regulation of 5-HT transmission. 5-HTT is itself a target for many drugs of abuse such as cocaine, 3,4-methylenedioxymethamphetamine ŽMDMA, ‘ecstasy’. and methamphetamine w18,25,30x. Blocking 5-HT uptake at one of the 5-HTT sites prolongs 5-HT transmission in the synaptic junction, which is one of the most effective ways of alleviating 5-HT deficiency disorders, such as depression and alcoholism. Several structural related chemicals labeled as Selective Serotonin Reuptake Inhibitors ŽSSRI., were developed on 5-HTT for clinical use in 5-HT related disorders. 5-HTT cDNAs have recently been cloned in rat Žfrom brain w4x, from basophilic leukemia cells w11x., human Žfrom brains w15x, from placenta w24x, from platelet w16x.,
0169-328Xr96r$15.00 Copyright q 1996 Elsevier Science B.V. All rights reserved. PII S 0 1 6 9 - 3 2 8 X Ž 9 6 . 0 0 2 0 9 - 4
268
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
and Drosophila w8,9x, and are highly conserved among species. In the rat and human species; each 5-HTT cDNA codes approximately 630 amino acids and shares more than 90% common sequence. 5-HTT proteins Žin human and rats. are commonly featured with 12 transmembrane domains, cytoplasmic N- and C-terminals, glycosylated sites in the second external loop, a potential kinase C phosphorylation site in the N- and C-terminals, and a leucine zipper motif in the second transmembrane domain. The N- and C-terminals and the third external loop are the least conserved regions and have been used for discrimination among 5-HTTs ŽFig. 1..
Three methods have been used to detect 5-HTT sites and 5-HT uptake activity: Ža. 5-HT uptake has been demonstrated by high affinity uptake of radioactive 5-HT through synaptosomes derived from tissue homogenates; Žb. distribution of 5-HTT uptake sites have been indicated by autoradiography of specific affinity ligand binding with 5-HT, citalopram, paroxetine, b-CIT, and RTI-55 w26x; Žc. the existence of 5-HTT has also been indicated by the detection of 5-HTT mRNA w4x. 5-HT uptake has been found in such selective tissue types as brain w1x, lymphocytes w10x, placenta chorionic brush border w2,7x, blood platelet w19x, adrenal medulla w12x, and small intestine
Fig. 1. The three specific antigen peptides in rat 5-HT transporter. The 5-HTT55 and 5-HTT315 antisera were made against the synthetic octomer peptide in the multiple antigenic peptide ŽMAP. forms. The 14-amino acid 5-HTT55 begins at 55th amino acid sequence located in the cytoplasmic N-tail, while the 11-amino acid 5-HTT315 begins at 315th amino acid sequence located in the 3rd external loop. The 5-HTTN antibody is made against a fusion protein with a 71-amino acid epitope specific and highly antigenic located in the first inner cytoplasmic N-tail which contains a putative phosphorylation protein kinase C binding cite ŽS or T.. All peptides are distal to the two glycosylation sites ŽN. in the second extracellular loop. The chance of a glycogen structure to hinder antibody binding to our peptide antigens is slim. These peptides all have greater than 80% homology with the human sequence. The 5-HTT55 MAP antigen is shown in the insert. The Multiple Antigenic Peptide ŽMAP. is an octomer polypeptide with a unique oligopeptide sequence which is linked by the core cysteine. The MAP provides sizeable molecular mass and does not compromise the specificity of the sequence as it increases in size. It also presents eight epitope copies.
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
w5,28x. However, the study of 5-HTT cell biology and the cellular distribution would not be achievable without direct visualization of the transporter protein w23x. In this study, three antibodies immunospecific to 5-HTT protein were produced by immunization against three sitespecific 5-HTT peptides, two against multiple antigenic peptides ŽMAP. form, one against fusion protein Žsee Section 2.. The antibodies were characterized by immunoblotting, immunocytochemistry, compared to the known distribution of 5-HT neurons and fibers, and verified by immunostaining after neurotoxic lesion. We produced and characterized these antibodies hoping to provide additional and unique markers to study the neuroanatomy of 5-HTT. This report indicates that these antisera can be used for detecting a specific 5-HTT transmembrane domain. 5HTTN antiserum can be used as a new marker for 5-HT neurons and their extensive fibers, and has the potential of detecting 5-HTT plasticity independent of 5-HT content.
2. Materials and methods 2.1. Synthetic multiple antigenic peptides 5-HTT55 and 5-HTT315 The multiple antigenic peptides ŽMAP. were made in the Molecular and Biochemical Technology Facilities at Indiana University School of Medicine. The selected peptides were named according to the number of the first amino acid in their coding sequences. 5-HTT315 has 11 residues beginning at position 315 of the 5-HTT coding sequence; it is in the third external loop. The 5-HTT315 was made as a multiple antigenic peptide ŽMAP, 8 copies of peptides branch-linked by lysine in the core.. 5-HTT55 is a 14-residue peptide; it is proximal to the first transmembrane domain near the N-terminal and is located in the cytoplasm. 5-HTT55 was made in MAP form ŽFig. 1, w22x. Žtotal Mr approximately 17 000.. MAP is multiple antigenic, and the octomer size is large enough to be immunogenic; no carrier protein coupling was used. The sequences and key characters of the two peptides are shown in Fig. 1. 2.2. 5-HTTN fusion protein 2.2.1. The construction of PGEX-5HTTN Two oligo-primers Ž5X-GATCGTAGGATCCCCTTGAATTCTCAG and 3X GCGCTCTGAATTCTATTGGCGAATCTC, containing added BamHI and EcoRI sites respectively. flanking the first 213 bp of the N-terminal coding region of the serotonin transporter were synthesized at the Wells Center for Pediatric Research Oligonucleotide Synthesis Facility. These oligonucleotides were used in a standard PCR reaction using 1 m g of the 5-HTT cDNA as a template Žfrom Dr. R.D. Blakely. and 25 cycles. The
269
PCR products were purified, digested with the restriction enzymes, BamHI and EcoRI, repurified and ligated into pGEX3X, a glutathione S-transferase ŽGST. fusion vector ŽPharmacia.. Competent HB101 bacteria cells were transformed with the ligation reactants and transformants screened for inserts using the colony PCR procedure. Potential positive clones were over-expressed by isopropyl-b-D-thiogalactopyranoside ŽIPTG. induction and proper protein expression verified. Furthermore, the clone used for antigen production, hereafter called PGEX5HTTN, was sequenced to confirm correct reading frame and lack of mutations. 2.2.2. OÕer-expression of the PGEX-5HTT construct and antigen purification The overexpression of the PGEX-5HTT was performed as previously described w29x. Briefly, bacterial cultures Ž5 ml. containing the PGEX-5HTT fusion construct were grown overnight at 378C in Luria broth ŽLB. plus 100 m grml ampicillin. Overnight cultures were diluted 1:10 in fresh, pre-warmed Ž378C. LB medium supplemented with the appropriate antibiotics and grown for 1 h at 378C. Expression of the PGEX-5HTT fusion protein was induced by the addition of IPTG to a final concentration of 0.1 mM and growing the cells for an additional 3 h at 378C. Cells were harvested by centrifugation at 1000 = g for 10 min and washed once with PBS, pH 7.4. Packed cells were resuspended in 2 ml of PBS and lysed by mild sonication Žthree 15-s bursts. on ice. The insoluble proteins were pelleted by centrifugation at 10 000 = g for 10 min at 48C; the PGEX-5HTTN fusion protein was soluble. The soluble fusion protein in the supernatant was mixed with 1% Triton X-100 and loaded onto a glutathione Sepharose 4B column pre-equilibrated with PBS plus 1% Triton X-100. Binding of the PGEX-5HTTN protein was carried out within the column overnight on a nutator at 48C. The column was subsequently washed with 20 column volumes of PBS containing 1% Triton X-100 ŽTBST.. Following the wash, the fusion protein was eluted with 50 mM Tris, pH 7.5, containing 10 mM glutathione. The sample was collected from the column and analyzed by SDS–polyacrylamide gel electrophoresis ŽSDS-PAGE. and Western blotting. This column purified protein was used for antibody production and for biochemical analysis. This procedure resulted in a single protein band upon SDS-PAGE analysis. 2.3. Antiserum production and purification For 5-HTTN antigen preparation, the fusion protein was further purified by transferring the protein onto nitrocellular membrane ŽNC., and stained with phenol red. The NC at the single band of 36 kDa Mr was cut out, sealed in dialysis tubing with TBS elution buffer, and eluted out of the polyacrylamide gel in a submarine electrophoresis
270
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
chamber for 20–30 min. The elute was then dialyzed overnight against distilled water. This final dialyzed elute was used as antigen for antibody production. The 5-HTT315, 5HTT55, and 5-HTTN antigens were used to individually immunize rabbits over three months to produce antibodies in our Laboratory Animal Research Facilities or at HTI, Ramona, CA, using standard protocol. In brief, 1 ml of antigen complex in PBS was mixed with an equal amount of complete Freund’s adjuvant to obtain a 0.5 mgrml mixture. This solution was injected subcutaneously Žtotal 300 m l. on the back of pathogen-free adult,
female white rabbits under sedation. Booster injections were given with incomplete adjuvant at 3–4 week intervals. The first bleed was obtained 9–10 weeks after the first injection, the second bleed at 11 weeks, and the final bleed at 12 weeks. The 5-HTT315 and 5-HTTN antisera were further affinity purified by incubating the antisera with 5-HTT315-MAP peptides or PGEX-5HTTN fusion protein respectively on strips of nitrocellulose membrane overnight in a 15 ml conical tube, washed three times, 10 min each, with TBST, and eluted with 250 m l antibody elution buffer for 30 min
Fig. 2. Western blotting of brain protein immunoprecipitated by 5-HTT315 Župper panel. and 5-HTT55 Žlower panel. antisera in 10% SDS-PAGE. The 5-HT315 recognizes the mobility of species in 92, 74 and 64 kDa. The 92 kDa and 74 kDa protein could be glycosylated 5-HTT, with 64 kDa being a fragment of deglycosylated 5-HTT. The 5-HT55 recognizes 74 and 64 kDa species. Both antisera quantitatively distinguish the low level of 5-HTT in the least 5-HT innervated cerebellum, to a high abundance in 5-HT dense brain regions, such as brainstem, cortex, and hippocampus. The numbers on the top of lane represent different animals. No binding was seen with liver or muscle Žnot shown.. The numbers on top of the lane represent different animals.
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
271
X Fig. 3. Ža. The construct p GE X.5-HTTN DNA Žlane 3. digested with enzyme BamHI and EcoRI yields a predicted size of 0.2 kb fragment Ž5 first 213 bp of coding region. and a large fragment which has a size equal to the vector Žlane 1. in 1.5% agarose gel. The p GE X.5-HTTN DNA was verified with correct sequence in frame. Žb. The HB101 bacteria expressed fusion protein before ŽBacterial pellet. and after purification with GST-column or GST-resin tube. All detectable eluted proteins show a single band with predicted mobility between 43 and 35 kDa in 10% SDS-PAGE.
Fig. 4. Western blot analysis using affinity purified 5-HTTN antibody. Total protein extracts Ž20 m g. from cortex ŽCX., hippocampus ŽHIPP., cerebellum ŽCB., liver, muscle Žlanes 1 to 5., and GST-5-HTTN fusion protein Ž10 ng, lane 6. were run on a 10% SDS–polyacrylamide gel, blotted to nitrocellolose and reacted with purified 5-HTTN antibody. The antibody only recognized the GST-5-HTTN fusion protein Ž; 37 kDa. and a 64 kDa single band in the brain, but not protein in liver, nor in muscle.
272
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
ŽPierce Immunopure, Rockland, IL.. The antibody was aliquoted and stored in y808C. 2.4. Western blotting Frontal cortex, hippocampus, cerebellum, and raphe were dissected from normal rat brains and homogenized by sonication in 0.1 mM PBS. Proteins were separated on
12% SDS–polyacrylamide gels and transferred to nitrocellulose membrane w29x. Gels, Whatman filter paper, and membranes were soaked in electroblotting buffer Ž25 mM Tris-HCl; 193 mM glycine; 20% methanol. for 15 min prior to transferring. Proteins separated on SDS–polyacrylamide gels, were transferred to nitrocellulose sheets by electroblotting in a Transblot BioRad transfer apparatus in 25 mM Tris, 192 mM glycine, 20% methanol at 150 mA
Fig. 5. 5-HTT315 Ža. and 5-HTTN Žb. immunocytochemistry in the dentate gyrus of hippocampus. The punctate pattern of 5-HTT315 immunostaining Ž5-HTT315-im. is characteristic of this antibody. Punctate type 5-HTT315-im is shown in infragranule layer where 5-HT fibers accumulate. The 5-HTTN characteristically shows fiber type staining and more intensively stain in infragranular layer, and hippocampal CA regions. The 5-HTTN is a very sensitive antibody for the fiber staining. All illustrations are darkfield microscopy. Scale bars: a, b s 300 m m.
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
Ž70 V.. The transfer was carried out for 1 h at 48C. Following protein transfer, the membrane was blocked with Blotto Ž1 = TBST Ž10 = TBSTs 1.5 M NaCl; 100 mM Tris-HCl, pH 8.0; 0.5% Tween-20; 2% NP-40; 0.2% SDS.; 5% nonfat dry milk; 0.02% sodium azide. for 1 h at room temperature on a rotator. The 5-HTT antibody was diluted 1:1000 in Blotto and incubated with the membrane at 48C overnight. The membrane was then washed three times in 1 = TBST for 10 min at 48C. Detection of antigen–antibody complex was accomplished using chemiluminescence method ŽBoehringer Mannheim, Indianapolis, IN.. 2.5. 5,7-Dihydroxytryptamine lesion Animals under anesthesia were pretreated with desipramine Žto protect dopamine and norepinephrine fibers, 10 mgrkg, Merrill Labs, Cincinnati, OH. 45 min prior to the lesions to protect the dopaminernorepinephrine fibers. The serotonin neurotoxin 5,7-dihydroxytryptamine Ž5,7DHT, 4 m gr0.4 m l of 0.01% ascorbic normal saline. was unilaterally delivered by two injections Žeach 0.4 m l. into the hippocampus of rats Žcoordinates to Bregma: 2.5 mm posterior, 2.8 mm lateral, 3.7 mm below. to remove serotonin innervation in the hippocampal region. Five normal rats and 4 rats 7 days after the 5,7-DHT injections were
273
sacrificed, their brains removed, sectioned and immunocytochemically stained with 5-HTTN antiserum Žwhich is more prominent for fiber staining than 5-HTT55 or 5HTT315.. Alternate sections were used for 5-HT immunocytochemical staining. 2.6. Immunocytochemistry All animals were perfused intracardially with formaldehyde ŽReagent grade, Fisher Scientific. freshly made from 4% paraformaldehyde and 0.1 M phosphate-buffered saline ŽPBS. under deep ketamine cocktail Ž120 mgrkg. anesthesia. Their brains were removed, left in the same fixative overnight, and coronally sectioned at 40 m m with a vibratome for immunocytochemical staining w30,34x. The 5-HTT315, 5-HTT55, 5-HTTN, and control preimmune sera Žsee control below. were used to stain the normal brain tissue. The 5-HTTN serum was used to stain the 5,7-DHT lesion tissue. A pilot study was used to characterize the dilution profile and Triton-X100 concentrations Ž0, 0.03%, and 0.3%.. In general, a 1:2000–3000 dilution of 5-HTT antisera, and 0.05% Triton were used in the later process. Sternberger’s peroxidase-anti-peroxidase ŽPAP. indirect-enzyme method was used for staining w27x. The PAP reaction was performed with 0.003% H 2 O 2 and 0.05% 3X ,3-diaminobenzidine. The primary, secondary and
Fig. 6. Examples of 5-HTTN immunostaining Ž5-HTTN-im. in the 5-HT terminal regions in cingulate cortex. The 5-HTTN antiserum has greater sensitivity in detecting the 5-HTT in the brain. A characteristic fibrous staining pattern is shown in the brain regions where 5-HT terminals are located. The 5-HTT-im fibers share many similarities in fiber morphology, distribution pattern, and density in regions of the brain with the 5-HT-im fibers. Characteristic dense 5-HTTN-im fibers in layer I and II near the surface of the brain, less dense in layer III–IV, and dense layer V highly resemble the distribution of 5-HT fibers in the correspondent subregions Ža.. Few, if any fibers were stained in the preimmune serum with same dilution Žb.. Few fibers were stained with antigen-preabsorbed 5-HTTN serum control Žc.. Straight 5-HTT-im fibers are seen in cingulum bundle ŽCB. dorsal to corpus callosum ŽCC. in Ža. but not in Žb. or Žc.. All illustrations are darkfield microscopy. Scale bars: a, b, c s 300 m m.
274
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
marker antibodies were diluted with PBS containing 0.05% Triton X-100 and 1% normal sheep serum. The primary antibodies were incubated overnight; the second and third, for 1 h each. Three washes with PBS each for 5 min were used between each antibody. All the sections were Nisslcounterstained with methyl green to reveal non-5-HTTN-im cells, and profile or brain structure. Control. Three sets of controls were used for immunocytochemistry, the preimmune serum of each antiserum, the 5-HTTN antiserum preabsorbed with fusion protein, and skipping of the primary antiserum. All controls were processed under the same conditions and in parallel with the 5-HTTN staining.
3. Results 3.1. 5-HTT55 and 5-HTT315 antisera In Western blots from 12% SDS–polyacrylamide gels and detection of antigen–antibody complexes using the
alkaline phosphatase method, 5-HTT55 and 5-HTT315 recognized rat frontal cortex, hippocampus, cerebellum ŽFig. 2., but not liver and muscle Žnot shown.. The apparent molecular masses were based on migration in the gel. 5-HTT315 recognized the mobility of species at 92, 74 and 64 kDa. The 5-HTT55 recognizes 74 and 64 kDa species. The possible glycosylation-rendered diversity in size is speculated in Section 4. The 5-HTTN has a single band at 64 kDa. The antisera quantitatively distinguished the low level of 5-HTT in the least 5-HT innervated cerebellum, to a high abundance in 5-HT dense brain regions, such as brainstem, cortex, and hippocampus. The apparent binding density was observed with cortex G hippocampus ) cerebellum ŽFig. 2.. This is consistent with the immunocytochemically stained 5-HT Ž5-HT-im. fiber density in these three regions. The brainstem was also observed with high density of 5-HTT55 binding as compared with the above three regions. Immunocytochemical staining using 5-HTT55 and 5HTT315 in the rat brain revealed similar punctate type staining along fibers in many forebrain regions with den-
Fig. 7. 5-HTTN-im neurons in the dorsal raphe nucleus ŽDRN. and median raphe nucleus ŽMRN.. 5-HTTN-im neurons distributed in a distinct central group Žarrows. and two lateral wings ŽL, arrowheads. of the dorsal raphe Ža, b., and in midline neuronal groups in median raphe Žcrossed arrows, c, d.. Many 5-HTTN-im punctates are seen surrounding the neurons and outline the somatodendritic profiles. The immunocytochemical DAB signals are brown in color; the background cells are green in color revealed by Nissl-counterstaining with methyl green. mlf, medial longitudinal fascicularis. All illustrations are brightfield microscopy. Scale bars: Ža,c. s 250 m m; Žb,d. s 20 m m.
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
sity and distribution patterns similar to 5-HT-im fibers. 5-HTT55 with 0.3% Triton-X was the optimal staining condition, while the 5-HTT315 was not affected by the Triton-X condition. Immunostaining using 5-HTT315 is shown in the hippocampus ŽFig. 5a.; the 5-HTT315-im is distributed predominately in the infragranular regions and the hippocampal fissure. Perisomatic staining and staining along membrane of primary dendrites was seen in the septal nucleus Žnot shown.. This type of staining is seen in 5-HT-im fibers in the septal nucleus of rats. The above observations indicate that antisera, 5-HTT55 and 5-HTT315 Ža. recognize a protein in the brain with highrlow density in brain regions of 5-HT, but not in liver
275
or muscle, and Žb. immunocytochemically stained a diffuse pattern Žin punctate form. along 5-HT-like fibers. 3.2. 5-HTTN antiserum PCR amplification of the 5-HTT cDNA with primers of a 210 bp regions at the amino end resulted in a single band which was digested with BamHI–EcoRI and cloned into pGEX.3X Ž4.9 kb. giving a correct 5.1 kb construct ŽFig. 3a.. The correct sequence of the construct and confirmation of the correct reading frame for the coded protein was confirmed, and hereafter called pGEX-5-HTTN. Three transformants with enhanced expression of fusion
Fig. 8. The effect of 5,7-DHT on the 5-HTTN-im fibers in the brain. Under desipramine pretreatment, the 5,7-DHT specifically damage the 5-HT fibers, but not dopamine or norepinephrine. The 5,7-DHT injection in the dorsal hippocampus removed 5-HTTN-im fibers in the entire ipsilateral side Žb. and majority of the contralateral side. A contrast was seen as compared with the normal hippocampus of non-lesioned animals Ža.. Degeneration extended rostrally to the anterior septum Žc. where fibers were also removed ipsilateral to the lesion Žstars. as compared with that of contralateral side. Adjacent to the degenerated anterior septal area where fibers survived, degeneration profile Žarrows. is seen with large degeneration stumps Žd, arrows.. a, b, c: darkfield microscopy; d: brightfield microscopy. The immunocytochemical DAB signals are brown in color; the background cells are green in color revealed by Nissl-counterstaining with methyl green. Scale bars: a, b s 500 m m; c s 500 m m; d s 20 m m.
276
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
protein with IPTG were obtained. They were selected for large scale production, and the purified protein revealed a distinct 37 kDa band as shown in Fig. 3b. Western blotting of 5-HTTN showed the fusion protein expressed in bacteria as a single 37 kDa band ŽFig. 3b and Fig. 4.. Western blotting analysis of rat frontal cortex, hippocampus, and cerebellum showed a single 64 kDa band. No staining was seen in liver or muscle protein ŽFig. 4.. Immunocytochemistry demonstrated that the preimmune, the 5-HTTN antibody-preabsorption, and the primary antibody-skipping controls, all stained negatively in the brain ŽFig. 6.. The antiserum after absorption against the fusion protein greatly decreased the immunostaining. The preabsorbed antiserum revealed very light and faint version of the non-absorbed antiserum, indicating a high titer of antiserum and remaining 5-HTTN antiserum after purification. The 1:3000 dilution of 5-HTTN antiserum immunostaining Ž5-HTTN-im. revealed a distinct morphology that is characteristics of 5-HT fibers ŽFig. 5b, hippocampus; Fig. 6a, cingulum cortex.. Curving, branching fine fibers with many fine varicosities were found in 5-HT fiber terminal regions, while thick, straight, and unbranching fibers were found in the 5-HT fiber pathways such as cingulum bundle and fornix. Many distribution pattern previously observed in 5-HT fibers were identified by the 5-HTTN antibody. A staining pattern of 5-HTT-im highly resembling that of 5-HT fibers in cingulum cortex is shown in Fig. 6. The 5-HTTN-im is densest in layers I and III and lower in layers II, IV–VI. Below layer VI and above the corpus callosum, the perpendicular 5-HTT-im fibers in cingulum bundle can clearly be seen. As we predicted, the 5-HTTN can also detect 5-HT neurons in raphe ŽFig. 7.. In contrast, no dopamine neurons in nigrarventral tegmental area, norepinephrine neurons in locus ceruleus, nor GABA neurons in any known GABA neurons distributed regions were detected by 5-HTTN Žnot shown.. The 5,7-DHT lesion in the hippocampus removed all 5-HTTN-im fibers in the ipsilateral and adjacent contralateral hippocampi ŽFig. 8.. Alternate sections of 5,7-DHT lesioned brains also stained negatively with 5-HT antiserum Žnot shown.. 5-HTTN-im fibers were greatly reduced with a degeneration profile in the brain regions distal to the injection site. Typical degeneration stumps, enlarged twisted and fragmented fibers were shown in frontal cortex millimeters away from the injection site in the hippocampus ŽFig. 8..
banks, especially specific against the known catecholamine and GABA transporter, Žb. high antigenic index w14x, and Žc. hydrophilic regions of the 5-HTT. The size of antigens varies among the three antisera. The 5-HTT55 and 5HTT315 antisera were made against MAP antigens. Each octomer-peptide has a molecular mass of approximately 13–15 kDa. Without additional conjugation of large molecules, the immunogen is sufficient to produce antiserum for Western blotting and immunocytochemistry. The advantage of MAP form antigen is that it has eight copies of the antigenic sites as compared with the same molecular mass of a single peptide conjugated antigen, yet, it still preserves the specificity of the small peptide. The disadvantage of MAP produced antibodies is that they can only detect a small region of the endogenous protein, which runs the risk, during immunocytochemistry, of the small portion of the peptide being masked by the folding of the protein itself or the surrounding tissue molecules. This seems to be the case with the 5-HTT55 antiserum against an N-terminal peptide as it detected only punctate staining of 5-HTT-im fibers, while the 5-HTTN antiserum against similar regions of 5-HTT but with greater peptide regions of 5-HTT detected the full fiber profile. However, the 5-HTTN antiserum which reacted against a 37 kDa fusion protein has a greater molecular mass than the other two. In general, the 5-HTTN provides a good morphological detection of 5-HTT fibers. In addition, it seems that the 5-HTT315 does not require Triton-X100 treatment, while the 5-HTTN and 5-HTT55 are more immunoreactive in the presence of Triton X-100. 5-HTT315 recognizes the extracellular loop while 5-HTT 55 and 5-HTTN recognize the intracellular loop region. Therefore, the use of detergent may render the accessibility of antibodies to the antigens across the membrane. The three antibodies against different domains of 5-HTT can be used for biochemical characterization of the 5-HTT. The sequence specific and high antigenic 71-amino acid N-terminal peptide is located in the first inner cytoplasmic loop which contains a putative protein kinase C binding cite ŽS.. This peptide is distal to the two glycosylation sites ŽN. in the second extracellular loop. An alteration in glycogen structure to hinder the antibodies binding to our peptide antigen is slim. This peptide is approximately 80% homologous with the human 5-HT serotonin sequence in this region. The 5-HTT315 antiserum provides an alternative antibody to detect the third extracellular loop, when characterization of 5-HTT transmembrane domains is desired. 4.2. Characterization of the antibodies
4. Discussion 4.1. The antigens The three antigenic epitopes were selected based on Ža. specificity against a known published sequence in the gene
In Western blots of the fusion protein, a single band was detected by our crude and purified 5-HTTN antiserum. The 5-HTT antibody recognized the fusion protein induced from the recombinant expression construct. In Western blots of the brain membrane preparation, SDS-PAGE indi-
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
cated that the mobility of 92 kDa is only seen with 5-HTT315 antiserum, the 74 kDa band is seen with the 5-HTT55 and 5-HTT315 antisera, while the 64 kDa protein is commonly observed in all 5-HTT55, 5-HTT315, and 5-HTTN antisera, in the brain. The 92 kDa and 74 kDa proteins could be glycosylated 5-HTT, and the 64 kDa commonly existing as a deglycosylated fragment of 5-HTT. The 92 kDa species is similarly detected by CT-2 antibody made against the 5-HTT C-terminal peptide, which detects an approximately 60 kDa molecule after enzymatic deglycosylation w23x. None of the three antisera recognized liver or muscle membranes. The 5-HTTN detected a single protein band and showed suitable staining for the immunocytochemical detection of the 5-HTT in the tissue. The 5-HTT55 and 5-HTT315 can be used when detection of a special domain of the 5-HTT is desired. Together, three pieces of immunocytochemical evidence support the specificity of the antisera. The staining of 5-HTT315, 5-HTT55, and 5-HTTN antisera consistently detected the 5-HT fibers observed in many brain regions. Three regions, cingulum cortex, hippocampus, and raphe are shown in Figs. 5–7. The distribution density of 5-HTT in many brain regions including the cingulum cortex, hippocampus and raphe indicated by the 5-HTTN antiserum is very similar to that of 5-HT innervation density. The distribution of 5-HT-im Žfrom our previous and others observation. and 5-HTTN fibers closely correlates to the detail cortical layer Žlayers I, II versus III–V in cingulum cortex. and many fiber bundles Žmedial forebrain bundle, stria terminalis, fornix, and fimbria.. The characteristic 5-HT immunocytochemical fiber morphologies-fine fibers highly decorated with varicosity, curving and twisting in the terminal regions, but short straight and thick in the pathway such as cingulum bundle, are also revealed by the 5-HTTN-im fibers. Furthermore, the 5-HTTN-im staining in the cell bodies in three ascending raphe nuclei, but not in substantia nigra, or locus ceruleus lends further support of the specific detection of the 5-HT neurons. Finally, the strongest evidence was derived from the study of 5-HTspecific neurotoxin lesion. The 5,7-DHT toxin under the protection of desipramine will specifically cause degeneration of 5-HT fibers w3,32,33,21x. The 5,7-DHT lesion removed all 5-HTTN fibers at the injection site, and caused a 5-HTTN-im degeneration profile within a millimeter distance from the center of the injection. The adjacent sections at regions where 5-HTTN-im disappeared were also 5-HT-im negative. This suggests that the 5-HTTN-im fibers are 5-HT in nature, unless a new large population of fibers are also 5,7-DHT sensitive and transport 5,7-DHT across the membrane. 4.3. Significance The 5-HTTN antibody serves as a good marker for 5-HT fibers in the case where 5-HT is depleted and 5-HT fibers are still intact. This maker is capable of distinguish-
277
ing depletion from degeneration. The 5-HTTN antibody should provide a practical application to distinguish the effects of abusive drugs which have different levels of effects on 5-HT neurons. In addition, 5-HTT seems intensely distributed in wide regions of the 5-HT fibers, and the degeneration profiles can be detected by the 5-HTTN antiserum. This antibody provides a very definitive marker for the degeneration of the 5-HT fibers. In conclusion, antisera against three specific regions of the serotonin transporter recognized antigen on fiber structure that is highly associated with 5-HT fibers in the brain. The slower migrating species in the SDS-PAGE may be various glycosylation forms of the transporter protein. The fast migration species is common to all three antisera, and is the unique single protein band when 5-HTTN antiserum was used. The three antisera provide options to identify different regions of the serotonin transporter. The 5-HTTN antiserum is suitable for immunocytochemically detecting minute amounts of serotonin transporter. The characteristic distribution and the degeneration in response to specific 5,7-DHT lesion of 5-H TTN-im fibers indicates that the 5-HTTN-im fibers are 5-HT fibers in nature. The 5-HTTN is a good marker to identify 5-HT fibers independent of 5-HT content. In addition to the use of the 5-HTTN antiserum in basic neurobiology research of 5-HT transporter, 5-HTTN antiserum has the ability to distinguish degeneration versus depletion, which will have many applications for verifying toxic effects of abusive drugs. For example, it will be possible to determine critical thresholds of drug dosage levels which may be maintained so that they will lead to depletion but not degeneration in experimental animals.
Acknowledgements We would like to thank Dr. R.D. Blakely for providing the 5-HTT cDNA template, the Wells Center oligonucleotide synthesis facility which is supported by the Riley Memorial Association for synthesis of the oligonucleotides used for the construction of the 5-HTT fusion protein, and Ms. Denise Scroggins, Laura Miller and Teresa Wilson for their technical assistance. This study is supported by NIH MH 50602 to F.C.Z. R.L. is supported by NIH AA06991. M.R.K. is supported by NIH RR09884 and Riley Memorial Association Research Grant.
References w1x Austin, M.C., Bradley, C.C., Mann, J.J. and Blakely, R.D., Expression of serotonin transporter messenger RNA in the human brain, J. Neurochem., 62 Ž1994. 2362–2367. w2x Balkovetz, D.F., Tiruppathi, C., Leibach, F.H., Mahesh, V.B. and Ganapathy, V., Evidence for an imipramine-sensitive serotonin transporter in human placental brush-border membranes, J. Biol. Chem., 264 Ž1989. 2195–2198.
278
F.C. Zhou et al.r Molecular Brain Research 43 (1996) 267–278
w3x Baumgarten, H.G., Jenner, S. and Klemm, H.P., Serotonin neurotoxins: recent advances in the mode of administration and molecular mechanism of action, J. Physiol. (Paris), 77 Ž1981. 309–314. w4x Blakely, R.D., Berson, H.E., Fremeau, R.T. Jr., Caron M.G., Peek, M.M., Prince, H.K. and Bradley, C.C., Cloning and expression of a functional serotonin transporter from rat brain, Nature, 354 Ž1991. 66–70. w5x Bledsoe, S., Chong, S.K.F. and Zhou, F.C., Immunocytochemical detection of serotonin and serotonin transporter in the human small intestine, Soc. Neurosci. Abstr., 21 Ž1995. 540.6. w6x Coccaro, E.F. and Murphy, D.L. ŽEds.., Serotonin in Major Psychiatric Disorders, American Psychiatric Press, Washington, 1990. w7x Cool, D.R., Leibach, F.H. and Ganapathy, V., Modulation of serotonin uptake kinetics by ions and ion gradients in human placental brush-border membrane vesicles, Biochemistry, 29 Ž1990. 1818– 1822. w8x Corey, J.L., Quick, M.W., Davidson, N., Lester, H.A. and Guastella, J. a cocaine-sensitive Drosophila serotonin transporter: cloning, expression, and electrophysiological characterization, Proc. Natl. Acad. Sci. USA, 91 Ž1994. 1188–1192. w9x Demchyshyn, L.L., Pristupa, Z.B., Sugamori, K.S., Barker, E.L., Blakely, R.D., Wolfgang, W.J., Forte, M.A. and Niznik, H.B., Cloning, expression, and localization of a chloride-facilitated, cocaine-sensitive serotonin transporter form Drosophila melanogaster, Proc. Natl. Acad. Sci. USA, 91 Ž1994. 5158–5162. w10x Faraj, B.A., Olkowski, Z.L. and Jackson, R.T., Expression of a high-affinity serotonin transporter in human lymphocytes, Int. J. Immunopharmacol., 16 Ž1994. 561–567. w11x Hoffman, B.J., Mezey, E. and Brownstein, M.J., Cloning of a serotonin transporter affected by antidepressants, Science, 254 Ž1991. 579–580. w12x Holzwarth, M.A., Sawetawan, C. and Brownfield, M.S., Serotoninimmunoreactivity in the adrenal medulla: distribution and response to pharmacological manipulation, Brain Res. Bull., 13 Ž1984. 299– 308. w13x Kuhar, M.J., Roth, R.H. and Aghajanian, G.K., Synaptosome from fore-brains of rats with midbrain raphe lesions: selective reduction of serotonin uptake, J. Pharmacol. Exp. Ther., 181 Ž1972. 36–45. w14x Kyte, J. and Doolittle, R.F., A simple method for displaying the hydropathic character of a protein, J. Mol. Biol., 157 Ž1982. 105– 132. w15x Lesch, K.P., Wolozin, B.L., Ester, H.C., Murphy, D.L. and Diederer, P., Isolation of a cDNA encoding the human brain serotonin transporter, J. Neural Transm., 91 Ž1993. 67–73. w16x Lesch, K.P., Wolozin, B.L., Murphy, D.L. and Diederer, P., Primary structure of the human platelet serotonin uptake site: identity with the brain serotonin transporter, J. Neurochem., 60 Ž1993. 2319–2322. w17x McBride, W.J., Murphy, J.M., Lumeng, L. and Li, T.K., Serotonin and alcohol preference wreviewx, Recent DeÕ. Alcoholism, 7 Ž1989. 187–209. w18x Molliver, M.E., Berger, U.V., Mamounas, L.A., Molliver, D.C., O’Hern, E. and Wilson, M.A., Neurotoxicity of MDMA and related compounds: anatomic studies wReviewx, Ann. NY Acad. Sci., 600 Ž1990. 649–661. w19x Moret, C. and Briley, M., Platelet 3 H-paroxetine binding to the
w20x
w21x
w22x
w23x
w24x
w25x
w26x
w27x w28x
w29x
w30x
w31x
w32x
w33x
w34x
serotonin transporter is insensitive to changes in central serotonergic innervation in the rat, Psychol. Res., 38 Ž1991. 97–104. O’Reilly, C.A. and Reith, M.E.A., Uptake of w 3 Hxserotonin into plasma membrane vesicles from mouse cerebral cortex, J. Biol. Chem., 263 Ž1988. 6115–6121. Patel, T.D., Azmitia, E.C. and Zhou, F.C., Increased 5-HT1A receptor immunoreactivity in the rat hippocampus following 5,7-dihydroxytryptamine lesions in the cingulum bundle and fimbria-fornix, BehaÕ. Brain Res., 73 Ž1996. 319–323. Posnett, D.N., McGrath, H. and Tam, J.P., A novel method for producing anti-peptide antibodies. Production of site-specific antibodies to the T cell antigen receptor b-chain, J. Biol. Chem., 263 Ž1988. 1719–1725. Qian, Y., Melikian, H.E., Rye, D.B. and Levey, A.I., Blakely, R.D., Identification and characterization of antidepressant-sensitive serotonin transporter proteins using site-specific antibodies, J. Neurosci., 15 Ž1995. 1261–1274. Ramamoorthy, S., Bauman, A.L., Moore, K.R., Han, H., Yang-Feng, T., Chang, A.S., Ganapathy, V. and Blakely, R.D., Antidepressantand cocaine-sensitive human serotonin transporter: molecular cloning, expression, and chromosomal localization, Proc. Natl. Acad. Sci. USA, 90 Ž1993. 2542–2546. Ricaurte, G.A. and McCann, U.D., Neurotoxic amphetamine analogues: effects in monkeys and implications for humans wReviewx, Ann. NY Acad. Sci., 648 Ž1992. 371–382. Scheffel, U., Kim, S., Cline, E.J. and Kuhar, M.J., Occupancy of the serotonin transporter by fluoxetine, paroxetine, and sertraline: in vivo studies with w 125 IxRTI-55, Synapse, 16 Ž1994. 263–268. Sternberger, L.A., Immunocytochemistry, 2nd edn., Wiley, New York, 1979, pp. 104–169. Wade, P.R., Tamir, H. and Kirchgessner, A.L., Gershon, M.D., Analysis of the role of 5-HT in the enteric nervous system using anti-idiotopic antibodies to 5-HT receptors, Am. J. Physiol., 266 Ž1994. G403–416. Wilson III, D.M., Deutsch, W.A. and Kelley, M.R., Drosophila ribosomal protein S3 contains an activity that cleaves DNA at apurinicrapyrimidinic sites, J. Biol. Chem., 269 Ž1994. 25359– 25364. Zhou, F.C. and Bledsoe, S., Methamphetamine causes rapid varicosis, perforation and definitive degeneration of serotonin fibers: an immunocytochemical study of serotonin transporter, NeuroscienceNet, V.1 Ž1996. Article a10009, Oct. 28, 1996; http:rrwww.neuroscience.com Zhou, F.C., Bledsoe, S., Lumeng, L. and Li, T.K., Reduced serotonergic immunoreactive fibers in the forebrain of alcohol-preferring rats, Alcohol. Clin. Exp. Res., 18 Ž1994. 571–579. Zhou, F.C. and Azmitia, E.C., Induced homotypic collateral sprouting of serotonergic fibers in the hippocampus of rat, Brain Res., 308 Ž1984. 53–62. Zhou, F.C. and Azmitia, E.C., Induced homotypic sprouting of serotonergic fibers in hippocampus. II an immunocytochemistry study, Brain Res., 373 Ž1986. 337–348. Zhou, F.C., Pu, C.F., Murphy, J.M., Lumeng, L. and Li, T.K., Serotonergic neurons in the alcohol preferring rat, Alcohol, 11 Ž1994. 395–403.