Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Brain histamine is involved in the regulation of arousal state, brain energy metabolism, locomotor activity, autonomic and vestibular functions, feeding, drinking, sexual behavior, and analgesia (Hough, 1988; Schwartz et al., 1991; Wada et al., 1991). Identification of molecular mechanisms used by brain histamine is therefore necessary for a better understanding of these complex physiological functions. The histamine H₃ receptor is one of the three receptors that is considered responsible for the actions of the neurotransmitter histamine (Schwartz et al., 1991; Hill et al., 1997). Originally discovered in 1983 as a presynaptic autoreceptor (Arrang et al., 1983), numerous studies have since shown that the H₃ receptor also regulates the release of other important neurotransmitters, such as acetylcholine, dopamine, glutamate, noradrenaline, and serotonin in both the central nervous system (CNS) and peripheral nervous system (Schlicker et al., 1988, 1989; Clapham and Kilpatrick, 1992; Schlicker et al., 1993; Brown and Reymann, 1996).

In vitro and in vivo studies suggest that H₃ receptor ligands have potential therapeutic use (e.g., Bowel's disease, ADHD, Alzheimer's disease, obesity) (Leurs et al., 1998), but also led to the recognition of potential receptor heterogeneity. Results from both radioligand binding and functional studies have provided evidence for the existence of H₃ receptor subtypes (West et al., 1990; Cumming and Gjedde, 1994; Jansen et al., 1994; Schworer et al., 1994; Leurs et al., 1996; Schlicker et al., 1996; Harper et al., 1999). So far, no convincing proof for receptor heterogeneity has been presented, probably because of the lack of knowledge on the genetic information encoding the H₃ receptor protein(s). Recently, Lovenberg et al. (1999) showed that, like the H₁ and H₂ receptor, the H₃ receptor belongs to the large superfamily of G protein-coupled receptors (GPCRs). Using the genetic information of the human H₃ receptor, we set out a PCR-based strategy to establish the existence of H₃ receptor subtype(s). In this study we report the existence of at least three functional rat H₃ receptor isoforms (H_3A, H_3B, and H_3C) that are generated as a result of alternative splicing. The three isoforms have distinct CNS expression profiles and couple differentially to adenylate cyclase and MAP kinase signaling pathways.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cloning of H₃ Receptor Isoform cDNAs. Total RNA (5 µg) of rat brain (CLONTECH, Palo Alto, CA) was reverse transcribed with random hexamer primers (100 ng/µl) (Invitrogen, Carlsbad, CA) and Superscript II reverse transcriptase (200 U) (Life Technologies, Gaithersburg, MD) according to the manufacturer's protocol. The cDNA was amplified by PCR using 2.6 U Expand High Fidelity DNA polymerase (Roche Diagnostics, Nutley, NJ) and 15 pmol of different couples of primers based on the human cDNA sequence (Lovenberg et al., 1999). After a 10-min denaturation step at 95°C, 35 cycles (1 min at 96°C, 40 s at 66°C, and 3 min at 72°C) were followed by a final extension for 8 min at 72°C. The use of primers overlapping the third intracellular loop of the human H₃ receptor (GenBank accession number AF140538) (5'-TGAACATCCAGAGGCGCACCC-3' as forward primer and 5'GCAGAGCCCAAAGATGCTCAC-3' as reverse primer corresponding to amino acids 224 to 229 and 364 to 370, respectively) resulted in the amplification of three different products, which were cloned in pCRII-TOPO and sequenced. The full-length cDNAs were isolated with primers overlapping the full H₃ sequence. The forward primer was based on the human H₃ cDNA sequence (5'-GTCCCGGAGCCGCGTGAGCCTGC-3'), whereas the reverse primer (5'-TACAAGGGCCTGGCCGTAGAAGG-3') was based on a mouse expressed sequence tag sequence (GenBank accession number AI509395). Five clones of each of the three different cDNA isoforms were sequenced automatically (PRISM 310; ABI, Norwalk, CT) on both DNA strands. For cellular expression, cDNAs were amplified with a new rat forward primer including a Kozak sequence (underlined) (5'-CCGCCACCATGGAGCGCGCGCCGCCCGACGGGCTG-3') and the reverse mouse primer and subcloned in pcDNA3.

Characterization of H₃ Receptor Isoforms. COS-7 cells were grown and transfected as described previously (Wieland et al., 1999). After 48 h, cells were homogenized and binding of [¹²⁵I]iodophenpropit (IPP) (1900 Ci/mmol; Menge et al., 1992), or [³H]N-methylhistamine in 50 mM Tris, 5 mM MgCl₂ buffer, pH 7.4, was determined (Jansen et al., 1994). Binding data were evaluated by nonlinear regression analysis using GraphPad Prism (GraphPad Software, San Diego, CA). All binding data were analyzed according to one- and two-binding site models and evaluated statistically.

Experimental Animals and Preparation of Tissue. All experiments were approved by the Åbo Akademi Animal Care and Use Committee. Adult male Wistar rats (weight, 200-250 g) were decapitated, tissues were rapidly dissected, frozen in precooled isopentane, and stored at 70°C. All tissues were cut to 15-µm cryosections, thaw-mounted onto poly-L-lysine slides, and stored at 70°C until used.

In Situ Hybridization. The oligonucleotides used for in situ hybridization were designed so that they specifically recognized the different H₃ receptor isoform mRNAs. The sequences are indicated in Fig. 3, except for oligo X, which detects all characterized H₃ receptor isoforms and spans the nucleotides 496 to 540 (5'-GCCACCAGACAGGTACTCCCAACTCA GGATGGCAGGCCCATACAG-3'). As a control probe, we used a Staphylococcus aureus chloramphenicol acetyltransferase-specific oligonucleotide. As an additional control, we routinely used a normal hybridization mixture with a 100-fold excess of unlabeled specific probes. The hybridization procedure used has been described before and was used with minor modifications (Dagerlind et al., 1992; Lintunen et al., 1998). All probes were labeled with [³⁵S]deoxyadenosine 5'-(-thio) triphosphate (NEN) at their 3' ends using terminal deoxynucleotide transferase (Promega, Madison, WI). Nonincorporated nucleotides were removed by purification through Sephadex G-50 columns.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Cloning of cDNAs Encoding Rat H₃ Isoforms. Using RT-PCR on rat whole-brain total RNA with primers based on the human H₃ cDNA sequence (Lovenberg et al., 1999), we obtained evidence for the existence of receptor isoforms. RT-PCR with a primer pair overlapping the nucleotide sequences, encoding the ends of the intracellular loop 3 (I3), resulted in the amplification of three different DNA products. One fragment showed 85% identity with the I3 loop of the human H₃ receptor cDNA and represented part of the rat H₃ receptor homolog. The other two sequences were identical to the first PCR product, but contained deletions of 96 and 144 bp, corresponding to potential in-frame deletions of 32 and 48 amino acids, respectively. Subsequent RT-PCR on rat whole-brain total RNA resulted in the isolation of full-length cDNAs, encoding three isoforms of the H₃ receptor. The open reading frames of the different cDNAs encode for proteins with 445 (H_3A), 413 (H_3B), or 397 (H_3C) amino acids. The H_3A receptor isoform shows 93% identity with the corresponding human H₃ receptor (Fig. 1). The 32- and 48-amino-acid deletions of the H_3B and H_3C isoform are located in the middle of the I3 loop, resulting in the deletion of potential PKC and PKA phosphorylation sites in the H_3C isoform (Fig. 1). In addition to the H_3A, H_3B, and H_3C isoforms, sequence analysis of the full-length cDNA clones revealed a deletion of 4 bp of the cDNA sequence corresponding to transmembrane domain 2. This 4-bp deletion results in a shift of the open reading frame, resulting in a truncated H₃ receptor isoform (H_3T) of 94 amino acids (amino acids 1-83 and 11 new amino acids) (Fig. 1).

View larger version (67K): [in this window] [in a new window]   Fig. 1.   Alignment of the rat and human H3 receptor amino acid sequence. The asterisks indicate the observed differences. The boxed amino acids indicate the deletions in either the H3B (open box) or H3C isoform (open shaded box), whereas the arrow indicates the change in the open reading frame, resulting in the insertion of 11 additional amino acids (bold) and a stop codon (H3T). Potential glycosylation (), PKA (), or PKC () phosphorylation sites were identified using the PROSITE database (Hofmann et al., 1999) and are indicated.

Pharmacological Characterization of the Rat H3 Receptor Isoforms. The H₃ receptor isoforms were transiently expressed in COS-7 cells and assayed for [¹²⁵I]IPP or [³H]NAMH binding. Except for the H_3T receptor, all H₃ receptor isoforms specifically bound the agonist and antagonist radioligands (Table 1, data not shown). High-affinity binding of [¹²⁵I]IPP to the H_3A, H_3B, or H_3C receptor did not differ importantly for the three isoforms and was displaced by a variety of selective H₃ receptor agonists and antagonists. The agonists histamine, immepip, and (R)--methylhistamine show a 3- to 5-fold difference in affinity for the H_3A compared with the H_3B or H_3C receptor (Table 1). A similar difference was observed for impentamine, which behaves as an agonist at the three isoforms (see below). For the H₃ antagonists clobenpropit and thioperamide, only slight differences in affinity were noticed (Table 1). Coexpression of the H_3T receptor with each of the other isoforms affected neither the expression of the respective H₃ receptor isoform nor the affinity of the agonist immepip (data not shown).

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Signal transduction of the H3 isoforms after expression in COS-7 cells. A, transfected cells were stimulated with 10 µM forskolin in the absence or presence of impentamine. cAMP was determined as the amount of luciferase expression and expressed as relative light units (RLU). The inset shows the effect of 0.1 µM immepip at the H3A receptor on forskolin-induced cAMP production in the absence or presence of 0.1 µM thioperamide or after PTX treatment (100 ng/ml). Similar data were obtained for the H3B and H3C isoforms. Data shown are mean ± S.E.M. of four determinations. B, effect of different concentrations of immepip on the phosphorylation of p44/p42 MAPK as determined by Western blot analysis using specific anti-phospho-p44/p42 (P-p44/p42) antibodies. Phosphorylation was quantified by chemiluminescence imaging and corrected for total MAPK (T-p44/p42) expression on stripped blots. Data shown are expressed as fold basal and are from a representative experiment. Similar data were obtained in two other independent experiments.

CNS Expression of Rat H₃ Receptor Isoforms. Application of four different histamine H₃-receptor specific probes revealed the receptor isoform expression patterns in the rat brain. The signal intensities for the probes were generally highest for H_3X, followed by H_3C, H_3A, and H_3B (H_3X > H_3C > H_3A H_3B; Fig. 3). In situ hybridization with the H_3X probe gave the strongest signal, because it detects the unspliced RNA message as well as all the isoforms. H_3C receptor isoform expression pattern resembled that seen with the H_3X probe, but the signal intensity was weaker. It is important to note that the signal intensity does not directly and reliably indicate expression levels, although probes were designed for similar conditions. Comparisons are possible between various brain regions when each probe is applied. The H_3C signal was strong in the striatum, olfactory tubercle, cortical laminae V and VIb, pyramidal layers of hippocampal fields CA1 and CA2, dorsal thalamic nuclei, ventromedial hypothalamic nucleus, locus ceruleus, tuberomamillary nucleus, trapezoid body, and the cerebellar Purkinje cell layer (Figs. 3 and 4). Moderate expression was also evident in layer II of the cerebral cortex, but it was low in, for example, medial septum, diagonal band, and substantia innominata (data not shown)

View larger version (80K): [in this window] [in a new window]   Fig. 3.   Histamine H3 receptor isoform expression in rat brain. The expression pattern seen with the H3X probe resembles that seen with the H3C probe. The H3A and H3B expression pattern clearly differs from that of H3C and H3X. Control hybridization with a 100-fold excess of unlabeled H3X probe results in a completely abolished signal (Block). The brain map indicates some of the relevant areas, and a schematic picture of part of the nucleotide sequence shows the positions for the isoform-specific oligonucleotides (oligo A, B, and C, respectively). The positions for the mRNA splicing are also indicated on the H3A sequence (underlined). The micrographs with in situ hybridization pictures are images from exposed X-ray films.

View larger version (127K): [in this window] [in a new window]   Fig. 4.   Histamine H3 receptor expression in rat brain. The in situ hybridization was carried out with the H3X probe which detects all the cloned mRNAs. All micrographs are scanned images from X-ray films. The stereotaxic location for each section is indicated by giving the distance to the bregma (in millimeters). CA1, 2, and 3, CA1, CA2, and CA3 pyramidal layer of hippocampus; CA3c, CA3 caudal part; CPu, caudate putamen (striatum); Cx, cortex; DG, dentate gyrus; DGd, dentate gyrus dorsal; DGv, dentate gyrus ventral; DRD, dorsal raphe nucleus, dorsal part; DRVL, dorsal raphe nucleus, ventrolateral part; DTT, dorsal tenia tecta; Ent, entorhinal cortex; LC, locus ceruleus; LSI, lateral septal nucleus, intermediate part; MGD, medial geniculate nucleus, medial part; MGV, medial geniculate nucleus, ventral part; MSO, medial superior olive; PC, Purkinje cells; Pn, pontine nuclei; RMC, red nucleus, magnocellular part; RPC, red nucleus, parvicellular part; SNR, substantia nigra, reticular part; SPO, superior paraolivary nucleus; Th, thalamus; TM, tuberomamillary area; Tu, olfactory tubercle; Tz, trapezoid body; VMH, ventromedial hypothalamic nucleus; VTT, ventral tenia tecta.

View larger version (89K): [in this window] [in a new window]   Fig. 5.   Histamine H3 receptor isoform expression in rat brain. A, the differential expression patterns of H3X and H3A probes in rat hippocampus. The pyramidal layer of the CA1 area is indicated with arrows, and dentate gyrus with arrowheads. B, expression pattern with the H3X probe in the dorsal raphe nucleus, dorsal part, is identical to the pattern seen with H3B. In dentate gyrus, the H3X expression is also obvious, whereas the H3B expression is below the detection level. C, a prominent expression signal with the H3X probe is revealed in the locus ceruleus and in Purkinje cells, but in adjacent sections, the H3B expression is faint. The micrographs are X-ray film images from adjacent brain sections and the film exposure times are similar for both probes (A), darkfield images taken from emulsion-coated and exposed slides (B and C), except for the inset picture, which is a bright-field image from a hybridized and counterstained section. DRD, dorsal raphe nucleus, dorsal part; DG, dentate gyrus; LC, locus ceruleus; PC, Purkinje cells; GrCb, granular layer of cerebellum. [Scale bars, 2 mm (A), 500 µm (B and C), 30 µm (inset micrograph)].

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, we report the cloning, CNS expression, and functional characterization of three rat H₃ receptor isoforms, named H_3A, H_3B, and H_3C, that vary in I3 length, with H_3B and H_3C lacking 32 and 48 amino acids, respectively. Moreover, we identified a 4-bp deletion variant that would give rise to a truncated receptor protein with only one transmembrane domain (H_3T). The rat H_3A receptor protein is 93% homologous to its human counterpart (Lovenberg et al., 1999) and corresponds to the rat variant that was reported by Lovenberg et al. (2000) (GenBank accession number AF237919) while this work was in progress. Also, the sequences encoding the H_3B and H_3C isoforms are already known in GenBank, although they are characterized as cDNAs for orphan GPCRs (BAA88767 and BAA88768).

The H₃ receptor isoforms are likely to be generated by alternative splicing. Submission of the human H₃ receptor cDNA sequence to the GenBank database led us to locate the human H₃ receptor gene at human chromosome 20 (GenBank accession number 7263900). Comparison of the reported human cDNA (Lovenberg et al., 1999) and the genomic sequence reveals that the human H₃ receptor gene consists of at least three exons separated by two introns. Exon 1 encodes the first 84 amino acids of the human H₃ receptor. In the human gene, the first intron of 1063 bp is exactly located at the position of the identified 4-bp deletion, suggesting that the rat H₃ receptor gene has a similar organization. The rat H_3T variant is probably the result of alternative splicing of intron 1. Exon 2 encodes for the amino acids 85 to 139 and is separated from exon 3 by a second intron of 1564 bp, which encodes for the rest of the human H₃ receptor (amino acids 140 to 445). If we assume a similar genomic organization of the rat gene, the H_3B and H_3C variants are generated by an alternative splicing mechanism without the obvious presence of an intron. The presence of potential splice donor and acceptor sites in exon 3 of the rat cDNA sequence (see Fig. 3) are apparently responsible for the generation of these isoforms, as has been shown previously for the P_2X2 receptor (Simon et al., 1997), for example. These putative splice donor and acceptor sites of the rat H₃ receptor gene are conserved in the human cDNA (Lovenberg et al., 1999) and genomic sequence, suggesting the presence of human H₃ receptor isoforms as well.

The general distribution of the H₃ receptor mRNA as revealed with the H_3X probe resembled that described previously in some brain areas (Lovenberg et al., 1999). All isoforms were expressed in the tuberomamillary histamine neurons. Further studies are needed to find out whether the H₃ receptor-mediated effects on histamine synthesis (Arrang et al., 1987) and release (Arrang et al., 1983) are regulated by the same or different isoforms. It is likely that the H_3C isoform is important in regulation of striatal, thalamic and cortical functions. The relatively strong expression of the H_3A isoform in the hippocampus renders it a likely candidate for regulation of hippocampal functions. Histamine has been shown to depress synaptic transmission in the dentate gyrus through an H₃ receptor-mediated mechanism (Brown and Reymann, 1996). The current results demonstrate differential expression of H₃ receptor isoforms in the dentate gyrus and hippocampal subfields. Based on current results, it is obvious that H₃ receptors in the hippocampal formation are located on pyramidal neurons of CA1-3, dentate granule cells, and multiple long-axon afferent pathways. The low expression of H₃ receptor isoforms in dorsal CA3 areas was in contrast with the strong expression in the basal CA3 and CA1 area. Binding of [³H]NAMH is also low in dorsal hippocampus, whereas it is moderate in ventral hippocampus (Cumming et al., 1991). This, together with the heavier innervation of ventral compared with dorsal hippocampus by histaminergic afferents (Panula et al., 1989), suggests that the ventral hippocampal areas are primarily regulated by tuberomamillary histaminergic neurons through H_3A and H_3C receptor isoforms. Strong expression of H_3B and H_3C isoforms in the locus ceruleus and dorsal raphe nucleus suggests that these isoforms may be responsible for inhibition of noradrenaline (Schlicker et al., 1989) and serotonin (Schlicker et al., 1988) release, respectively. Lack of all isoforms in the pars compacta of the substantia nigra suggests that the H₃ receptor-mediated inhibitory effect on dopamine release as observed in the mouse (Schlicker et al., 1993) may be indirect. This is supported by the abundant presence of H₃ receptor mRNA in striatal cells, which may thus mediate the inhibitory effect on dopamine release from striatal dopaminergic terminals. In agreement with evidence that suggests that H₃ receptors are not located on cholinergic terminals (Arrang et al., 1995), very low expression of all isoforms was characteristic of the medial septum, diagonal band, and substantia innominata, areas in which cholinergic projection neurons are located. Regulation of acetylcholine release in vitro (Clapham and Kilpatrick, 1992) or in vivo (Blandina et al., 1996) by histamine through H₃ receptor may thus also involve indirect mechanisms. The cortical neurons in several cortical laminae that express H₃ receptor mRNA may mediate the effect. These cells, together with cholinergic terminals, are present in slice preparations used in some experiments, which makes it difficult to evaluate the release site in perfusion experiments. Cerebellar Purkinje cells expressed strongly the H_3C isoform, and H_3A suptype was found in granule cells. A direct hypothalamo-cerebellar pathway consists of long histaminergic axons that pass through the granule cell and Purkinje cell layer and enter the superficial portion of the molecular layer also in human brain (Panula et al., 1993). Histamine may thus participate in regulation of motor functions in cerebellum through H_3A and H_3C receptor isoforms.

Differences in H₃ receptor isoform expression were found in many other brain areas as well. The functions of H₃ receptor in many of these areas are in general poorly known. The heterogeneous distribution of the isoforms suggests that H₃ receptor isoform-specific functional histaminergic regulation may be important in several areas. The H₃ receptors displayed differential expression in key areas involved in regulation of the sensory, endocrine, and cognitive functions in the brain. Robust changes also occur in the brain histamine system during the hibernation cycle, in which the turnover is high during the hibernation bout when other transmitter systems are generally inactive (Sallmen et al., 1999). Hence, histamine may modulate many general functions (Hough, 1988; Schwartz et al., 1991; Wada et al., 1991) through the H₃ receptor isoforms.

As found for the human H₃ receptor (Lovenberg et al., 1999), the rat isoforms bind H₃ selective agonists and antagonists with high affinity and inhibit the production of cAMP via PTX-sensitive G proteins. In line with the idea that the I3 loop is important for GPCR-G protein coupling (Wess, 1997), reduced potencies for various H₃ receptor agonists at the H_3A receptor were observed in comparison with the H_3B or H_3C isoform. The histamine homolog impentamine, which has been reported as an antagonist at the H₃ receptor in the guinea pig jejunum and an agonist for the H₃ receptor in the rat cerebral cortex, showed full agonism at all three isoforms. Again, activity at the H_3A receptor was reduced. For all H₃ receptor isoforms, the inhibition of adenylate cyclase was completely PTX-sensitive, (i.e., G_i/o-mediated). The observed differences in agonist potency at the three isoforms point to differences in coupling efficiencies to the same G_i/o-subunit but can also be explained by an isoform-specific coupling to distinct G_i/o-subunits, as previously reported for the isoforms of the D₂ receptor (Monsma et al., 1989).

Interestingly, activation of the H₃ receptor isoforms also leads to activation of the MAP kinase signaling cascade via PTX-sensitive G proteins. The H_3A isoform seems to be more effectively coupled to the p44/p42 MAPK activation, further stressing the differences in G protein coupling of the different isoforms. This is the first report linking the H₃ receptor to the MAPK pathway, which is believed to be important in neuronal plasticity and is activated in hippocampal long term potentiation (English and Sweatt, 1996; Bhalla and Iyengar, 1999). The H₃ receptor is known to be involved in learning and memory processes (see Leurs et al., 1998) and histamine has also been implicated in long-term potentiation (Brown et al., 1995). The strong expression of the H_3A isoform in the hippocampus and the preferential linkage to the MAP kinase cascade will add a new level of complexity to our understanding of the role of the H₃ receptor(s) and histamine in this process.

In conclusion, in contrast to the H₁ and H₂ receptor (Hill et al., 1997) the presence of introns in the H₃ receptor gene gives rise to various H₃ receptor isoforms via alternative splicing. The three identified rat H₃ receptor isoforms have a distinct CNS distribution and show some differences in pharmacology and signaling. Moreover, all three isoforms couple to the MAPK cascade, a newly identified signaling pathway for the H₃ receptor. Our data are supported by the recent report of Tardivel-Lacombe et al. (2000). While this article was in preparation, the cDNAs of the guinea pig H₃ receptor and one shorter isoform (Tardivel-Lacombe et al., 2000) were reported. The shorter guinea pig H₃ isoform corresponds to the rat H_3B variant, but no radioligand binding data, signal transduction, or isoform-selective expression was reported. Because the identified rat H_3A, H_3B, and H_3C subtypes do not explain all pharmacological findings that gave rise to suggestions of H₃ receptor heterogeneity, it is possible that the complex genomic organization of the H₃ receptor can result in further isoforms.