Multi-Transcriptional Profiling of Melanin ... - Springer Link

C 2005) Cellular and Molecular Neurobiology, Vol. 25, No. 8, December 2005 ( DOI: 10.1007/s10571-005-8184-8

Multi-Transcriptional Profiling of Melanin-Concentrating Hormone and Orexin-Containing Neurons ´ 1 Micha Nethe,1 Lucien F. Harthoorn,1,2 Arseni San˜ e, 1 and Joop J. Van Heerikhuize Received June 7, 2005; accepted August 25, 2005 SUMMARY 1. Melanin-concentrating hormone (MCH) and orexin-containing neurons participate in hypothalamic circuits that control energy homeostasis. While these two systems have projections to widespread target areas within the central nervous system, little is known about intrinsic characteristics and the molecular composition of both the MCH and orexin neurons themselves. 2. By a combinatory approach of quantitative immunocytochemical identification and analysis with laser microdissection and semi-quantitative Real-time RT-PCR, here we present multi-transcriptional profiling of MCH and orexin neurons in the rat lateral hypothalamus. 3. Immunocytochemical analysis showed that orexin peptide expression was increased after fasting both during the activity and resting period of rats, whereas MCH peptide content was only clearly upregulated at resting phase. Subsequent transcriptional profiling showed distinct expression patterns of MCH, orexin and cocaine-amphetamine regulated transcript (CART) between MCH and orexin neurons. A low expression level of dynorphin was found both in MCH and orexin neurons. Receptor expression profiles, reflecting interaction with neuropeptide Y, melanocortins, leptin, glucocorticoids and GABA, showed approximately similar expression patterns among the MCH and orexin neuronal systems. Expression of glutamate- and GABA-markers revealed a possible contributory role of both glutamate and GABA in functional output of MCH and orexin neurons. 4. This method allowed differential screening at mRNA level after immunocytochemical neuron identification and analysis in heterogeneous brain regions, which can further specify functioning of the individual neurons. With respect to MCH and orexin neurons, this study emphasizes that these neurons are targets for stimulatory and inhibitory signals from other brain regions including the arcuate nucleus and the general circulation. Additionally, both glutamate and GABA appear to be involved in MCH and orexin neuronal functioning related to feeding and regulation of the energy balance. KEY WORDS: immunocytochemistry; gene expression; hypothalamus; food intake; diurnal; rat.

1 Netherlands

Institute for Brain Research, Amsterdam, The Netherlands.

2 To whom correspondence should be addressed at Netherlands Institute for Brain Research, Amsterdam,

The Netherlands; e-mail: [email protected]. 1209 C 2005 Springer Science+Business Media, Inc. 0272-4340/05/1200-1209/0

1210

Harthoorn, San˜ e, ´ Nethe, and Van Heerikhuize

INTRODUCTION Distinct areas of the hypothalamus have been implicated in regulation of homeostatic processes; they sense and integrate neural and endocrine metabolic signals that reflect the body’s nutritional state and engage distinct effector pathways to anticipate physiological changes so that energy homeostasis can be maintained (Schwartz et al., 2000). The lateral area of the hypothalamus has been shown to enclose two discrete but spatially overlapping neuronal populations that produce melaninconcentrating hormone (MCH) and orexins (hypocretins), both of which exert pronounced feeding-stimulatory (orexigenic) actions (Qu et al., 1996; Sakurai et al., 1998). The organisation ascribed to both MCH and orexin neurons indicate that they most likely represent targets for input from the arcuate nucleus, and thus together constitute components of a larger hypothalamic circuit controlling food intake and energy balance (Broberger et al., 1998; Elias et al., 1998). Notwithstanding the restricted localization and potentially specific input of MCH and orexin-containing neurons within the lateral hypothalamus, these two systems have projections to a large variety of target areas within the central nervous system. The extensive distribution and location of MCH and orexin fibers and receptors imply that both peptides do not only play a considerable role in feeding, but also affect a broad spectrum of non-ingestion related processes (Bittencourt et al., 1992; Peyron et al., 1998; Date et al., 1999; Van den Pol, 1999; Saito et al., 2001; Hill et al., 2001). Particularly the dense innervation and strong excitatory effects of orexins on noradrenergic locus coeruleus neurons demonstrates that this system most notably provides a functional link between energy homeostasis and wakefulness (Horvath et al., 1999; Hagan et al., 1999; Chemelli et al., 1999; Yamanaka et al., 2003). Despite their widespread target regions that can be associated with numerous important homeostatic functions, little is known about the intrinsic features of both the MCH and orexin neuronal systems themselves. In terms of receptor and co-transmitter expression, only morphological studies have shown potentially important features of MCH and orexin neurons, such as co-expression of cocaineamphetamine regulated transcript (CART) (Broberger, 1999; Vrang et al., 1999), dynorphin (Chou et al., 2001), glutamate (Abrahamson et al., 2001), vesicular glutamate transporters (VGLUTs) (Collin et al., 2003; Rosin et al., 2003), leptin receptor ˚ (Hakansson et al., 1998, 1999), pancreatic polypeptide/neuropeptide Y (NPY)-4 re¨ ceptor (Campbell et al., 2003) and GABA receptor subtypes (Backberg et al., 2003, 2004). Especially the role of CART, which was principally shown to be an anorectic peptide (Kristensen et al., 1998), and its relationship with MCH may appear paradoxical. In the present study, we aimed at gaining insight into intrinsic characteristics of MCH and orexin neurons by an approach of transcriptional profiling at the individual neuronal level. Hereto, individual MCH and orexin neurons in the rat lateral hypothalamus were first identified and analysed for their peptide content by means of quantitative immunocytochemistry, especially in relation to day–night changes and fasting, and were further isolated by single cell laser microdissection and pressure catapulting (LMPC) for subsequent semi-quantitative Real-time RT-PCR.

Expression Profiling of MCH and Orexin Neurons

1211

METHODS Animals and Tissue Preparation All experiments were conducted with male Wistar rats (250–300 g, Harlan, Zeist, The Netherlands) under the approval of the Animal Care Committee of the Royal Netherlands Academy of Arts and Sciences (DEC-KNAW), acting in accordance with the European Communities Council Directive (86/609/EEC). The animals were housed on a 12-h light, 12-h dark cycle (lights on at 7:00 a.m.). Water and standard food pellets were usually available ad libitum, except for the fasted animals, which were food-deprived during 48 h before sacrificing. Animals were sacrificed in the early phases of either the resting (light) or active (dark) period at Zeitgeber time (ZT) 2 or ZT14, respectively, with 50 mg sodium pentobarbital i.p., followed by transcardial perfusion with RNase-free (diethylpyrocarbonate (DEPC)-treated) phosphate-buffered saline (PBS) (pH 7.4), followed by a solution of 4% paraformaldehyde in PBS. Brains were removed, post-fixed for 12 h, and cryoprotected by immersion with DEPC-treated 30% sucrose in 0.2 M phosphate buffer (pH 7.4) for a further 48 h. Coronal sections (18 µm) were cut at the hypothalamic level by cryostat (−25◦ C), and were treated free-floating with DEPC-treated solutions. During all antibody incubations, RNase-inhibitor (40 U RNase-OUT, Invitrogen) was added. After cutting, the sections were extensively washed in PBS, and incubated overnight with rabbit anti-rat MCH (1:60,000) (Netherlands Institute for Brain Research, Amsterdam) or rabbit anti-rat orexin (1:40,000) (Institute for Protein Research, Osaka, Japan). The production and specificity of these antisera were previously reported (Bujs et al., 2001). The sections were then incubated for exactly 1 h with biotinylated goat anti-rabbit IgG, followed by exactly 1 h with ABC complex (Vector Laboratories, Burlingame, CA, USA), and finally stained for exactly 10 min with 0.025% 3,3 -diaminobenzidine (DAB) in PBS to which 0.05% H2 O2 was added. Sections were mounted on 1 mm thick silanized (2% 3-aminopropyl-triethoxysilane in acetone) glass slides, and enclosed with 85% glycerol in PBS (pH 9.0) under cover slips. For the purpose of quantitative determination of peptide amounts by optical density measurement of the final DAB-precipitate, we validated the relationship between immunostaining intensity within individual MCH and orexin cell bodies and first antibody concentrations, based on antibody dilutions of 1:8000, 1:16,000, 1:32,000, 1:64,000 and 1:128,000. Quantitative Image Analysis Quantitative image analysis was performed with an image analysis system consisting of a Zeiss Axioskop microscope, a Sony-XC77 camera and Image-Pro Plus 4.5 software (Media Cybernetics, Silver Spring, MD, USA). Densities of DABprecipitate were measured of immune positive cell profiles containing a nucleolus in a set of random systemic selected sections using a monochromatic 560 nm contrast filter as previously reported (Goncharuk et al., 2001). Outside the section 100% transmission of light and shading correction were adjusted, and background correction was done at random in a hypothalamic area where no neurons were present.

1212


Each sample consisted of the mean integrated optical density (IOD) of about 25 MCH- or orexin-containing cells. LMPC and Real-time RT-PCR After quantitative image analysis, cover slips were removed from the glass slides, and glycerol was rinsed out with DEPC-treated water. The immunocytochemical images were compatibly transferred to a PALM Microbeam System (PALM Microlaser Technologies, Bernried, Germany), which was used to carry out LMPC. Measured samples, containing either 25 MCH neurons or 25 orexin neurons, were collected selectively and individually by direct contact-free laser pressure catapulting into a PCR tube cap filled with 5 µL lysis buffer (4 M guanidine thiocyanate, 25 mM sodium citrate, 0.5% Sarcosyl, pH 7.8). In a first session, 45 samples of MCHand orexin-stained sections were processed for Real-time RT-PCR analysis, while for a second set of profiling 27 samples of both MCH- and orexin-containing neurons were analysed. Since only complete, full-sized neurons were collected and analysed, the risk of contamination with adjacent neurons was aimed to be minimal. In order to improve mRNA recovery (Fink et al., 2000), each sample was further treated with Proteinase K (1 mg/mL, RNase-free, Invitrogen) for 2 h at 65◦ C. After heat inactivation, total RNA was isolated using the Trizol Reagent method (Invitrogen) with volumes for small RNA quantities, and glycogen (Boehringer-Mannheim, Germany) as co-precipitant. Possible contamination of genomic DNA was prevented by DNasetreatment of the RNA samples (1 U DNase I, Amplification Grade, Invitrogen) for 15 min, after which DNase was heat-inactivated at 70◦ C for 10 min. RNA samples were reverse transcribed using 50 ng of oligo-(dT)21 , and 200 U reverse transcriptase (SuperScript II RT, Invitrogen) for 1 h at 42◦ C. For each transcript, a 1 µL aliquot of cDNA sample was used for Real-time RT-PCR using a GeneAmp 5700 sequence detector and the SYBR Green PCR master mix (Applied Biosystems). Almost all PCR primer pairs were designed with intron-spanning annealing properties closest to the poly(A)-tail using Primer Express 2.0 (Applied Biosystems), giving amplicons between 90 and 300 basepairs (Table I). Concentrations of primer pairs were optimised at 150 nM. Non-reverse transcribed samples (n = 3) and lysis buffer blanks (n = 4) were included as negative controls. Data Validation, Processing, and Analysis The outcomes of Real-time RT-PCR results were expressed as threshold cycle (Ct ), which was the number of PCR rounds needed to pass a point Rn of fluorescence intensity set at 0.500. The results were validated based on the quality of dissociation curves, generated at the end of the PCR runs by ramping the temperature of the samples from 60 to 95◦ C, meanwhile continuously collecting fluorescence data. The decisive criterion was that amplification of single product showed a melting temperature (Tm ) (see Table I) distinct from that of an aspecific signal or primer– dimer products, and corresponding to the dissociation curve of each of the primer pairs tested on reference cDNA. This reference cDNA was synthesized from RNA extracted from a total homogenate of rat hypothalamus. Because of the relatively

Forward

CCGCAGAAAGATCGGTTGTT GCCAGAAGACGTGTTCCTGC TGGATGATGCGTCCCATGA TGATGAATGATGAAGCCGCA GCCGAAGCATAAGCTGTGGAT TGATGGCGAGGCTTCACATT CAGAGCACCCAGGGAACCT TGCTCTGCTTTGCTCCTGATC GCAAAAGCGTGGTTCCAGAA GGTGCAATGACCAAGCACAAG GGAAGCCTCAGCACACAAATG ATGGTGAGCCTGAGCACACAA

MCH Orexin CART Dynorphin Y5R MC4R OB-Rb GR GABAARα1 VGLUT1 GAD-65 GAD-67

TGGTCCTTTCAGAGCGAGGTA GCTTTCCCAGAGTGAGGATGC CGGAATGCGTTTACTCTTGAGC TGAACTGACGCCGCAGAAA TTTTCTGGAACGGCTAGGTGC TGAGACATGAAGCACACGCAG CTGTTTTCACGTTGCTGACCA TGTCAGTTGGTAAAACCGTTGC TTAGCAATAGTGGCCAAGCCG TTCACTTTCGTCACTGCCAGC ACCATGCGGAAGAAGTTGACC TGAGGCTGGTAACCAACCATG

Reverse 174 96 119 266 237 199 92 299 136 201 190 163

bp

82 84 80 86 83 83 80 79 80 85 82 84

Tm

Primer Sequences for Semi-Quantitative Real-time RT-PCR, Amplicon Length (bp), and Observed Tm (◦ C)

Target

Table I.

M62641 AF041241 U10071 NM 019374 U66274 U67863 U60151 M14053 L08490 U07609 M72422 M76177

GenBank

Expression Profiling of MCH and Orexin Neurons 1213

1214


small amounts of RNA extracted from a minimum number of 25 LMPC-dissected neurons per sample, for each designed primer pair linearity of Real-time RT-PCR signalling was verified by determining correlations between the amount of cDNA and the Ct with wide-range serial dilutions of reference cDNA that covered the amount of target mRNA expected in the experimental samples. The resulting Ct values were plotted against the logarithm of the dilution factor. To show whether all amplifications developed in a log-linear manner, the LinRegPCR tool (Ramakers et al., 2003) was used to analyse the increase of SYBR Green fluorescence in the course of each individual PCR, which also allowed deriving primer-specific amplification efficiencies (Es ). For this purpose fit options were chosen with best linear correlation coefficients between 0.997 and 0.999. For absolute expression levels E−Ct values were calculated and, in relation to a constant number of molecules (C) synthesized at Ct set at 1012 following X0 = C × E−Ct , used as estimates of relative copy number of template molecules present at the start of the reaction (X0 ) (Meijerink et al., 2001; Kamphuis et al., 2001). IODs obtained were expressed as mean values ± SEM. At the level of transcriptional profiling, expression of individual transcripts was represented as absolute expression levels ± SD. For quantitative image analysis and expression levels between MCH and orexin neurons statistical significance was established using Student’s t-test. In all cases, differences were considered statistically significant when p < 0.05.

RESULTS Quantitative Immunocytochemistry of MCH and Orexin Neurons Similar to earlier studies, which describe quantitative immunocytochemistry as an appropriate quantitative determination of peptide amounts by optical DABdensity measurements (Van der Sluis et al., 1988; Goncharuk et al., 2001), we validated the relationship between immunostaining intensity within individual MCH and orexin perikarya and first antibody concentrations on rat hypothalamic sections. This was based on antibody dilutions of 1:8000, 1:16,000, 1:32,000, 1:64,000 and 1:128,000. The intensity of MCH and orexin staining was proportional to the antibody dilution representing linear correlations (MCH: r = 0.986, orexin: r = 0.973). Figure 1 shows that the majority of stained MCH and orexin perikarya were located within the lateral hypothalamus, according to previously reported data (Bittencourt et al., 1992; Sakurai et al., 1998; Elias et al., 1998; Peyron et al., 1998; Date et al., 1999). The peptide content of MCH and orexin, expressed as IOD of DAB-staining, is represented in Fig. 2A and B (each experimental group, n = 14), respectively, and shows an upregulation after 48 h of fasting in the first phase of the resting (light) period at ZT2, whereas at the beginning of the active (dark) period at ZT14 only the orexin content showed a significant upregulation after fasting. For orexin neurons, the intensity of staining after fasting at ZT2 and ZT14 increased by approximately 40 and 50%, respectively, compared to a 35% increase of MCH staining intensity at ZT2 after fasting.


1215

Fig. 1. MCH and orexin expression in the rat hypothalamus. (A, B) Distribution of MCH and orexin immunoreactive perikarya in the lateral hypothalamic area. (C, D) Immunostaining reveals specific MCH and orexin neurons within the zona incerta (ZI) and around the fornix (fnx) (perifornical region), respectively. Scale bars represent 100 µm.

Quality and Validation of LMPC and Real-time RT-PCR Laser microdissection techniques have been shown to be a proper and elegant method for isolation of individual cells or cell populations from heterogeneous tissue (Emmert-Buck et al., 1996). Since this innovative approach is free of laser-induced destruction in isolated cells, ensuring the integrity of nucleic acids isolated for downstream analysis, the combination of single neuron isolation with transcriptional profiling has been made relatively easily (Simone et al., 1998; Bahn et al., 2001). In the present study, illustrated in Fig. 3A and B, LMPC with the PALM Microbeam system showed to be an application with great spatial resolution for an exact dissection of individual cells, with minimal contribution of adjacent cellular structures, of which the peptide content was first measured by quantitative image analysis. For all designed primer sets, linearity of Real-time RT-PCR signalling was determined with wide-range serial dilutions of reference cDNA and clear linear correlations were found between the amount of cDNA and the Ct for the duration of at least 40 Realtime RT-PCR rounds. The sensitivity of the PCR detection was sufficient to study the LMPC-cDNA samples without the need for an RNA-preamplification step. In spite of great sensitivity of Real-time RT-PCR measurements, a number of samples was found showing contribution of an aspecific signal most likely by aspecific annealing

1216


Fig. 2. Effects of fasting at two diurnal time points on MCH peptide expression in MCH neurons and orexin peptide expression in orexin neurons. (A, B) MCH and orexin peptide content as IOD of staining intensity in rat hypothalamus after 48 h of fasting and in control animals measured both at light (ZT2) and dark (ZT14) period. Bars represent the mean values ± SEM of 25 stained neurons (each experimental group, n = 14). (∗ p < 0.05; Student’s t-test).

and primer–dimer products as validated by the quality of the dissociation curves. By the LinRegPCR tool (Ramakers et al., 2003), the results of all correct individual Real-time RT-PCRs were analysed to ensure log-linear patterns of amplification. By the use of this implementation, primer-specific Es were determined for each primer pair during these individual PCRs, and were further used for adequate calculation of absolute expression levels. Multi-Transcriptional Profiling of MCH and Orexin Neurons Figure 4A–D illustrates the first level of expression analysis, which involved screening of the number of positive and negative Real-time RT-PCR signals of MCH and orexin neuronal samples. A differential pattern was found for MCH, orexin and CART. Figure 4A shows that MCH mRNA was detected in 18 samples out of 45 neuronal samples identified as MCH-containing, whereas in Fig. 4B only three samples out of the 45 identified as orexin-containing showed MCH mRNA.


1217

Fig. 3. Two images showing the successive stages of LMPC of immunocytochemically identified and analysed neurons in an 18 µm-thick immunocytochemical section. (A) Typical example of MCH neurons that were first measured for staining intensity. (B) After uncovering the immunocytochemical section, the images were compatibly transferred to a PALM Microbeam system where the neurons were selectively harvested by contact-free laser pressure catapulting.

Orexin expression analysis by this approach appeared to be difficult; only six orexin samples showed correct orexin signaling as shown in Fig. 4B, but no MCH samples contained any orexin product as graphed in Fig. 4A. Most presumably this has been raised by difficulties in transcript properties for primer design, because the orexin transcript is relatively short that restricts ample possibilities for primer design. Figure 5 represents the absolute expression level of MCH and orexin where differences were found to be specific between the MCH and orexin neuronal populations. These specific expression patterns of MCH and orexin mRNAs in neurons identified both as MCH- or orexin-containing and vice versa provide evidence for the specificity of LMPC and Real-time RT-PCR at the individual neuron level. A different expression was also found for CART, which is shown in Fig. 4A where CART mRNA was found in 25 samples out of the 45 MCH samples, instead of only three out of the 45 identified as orexin-containing in Fig. 4B. Also the absolute level of expression as presented in Fig. 5 strongly points to a specific co-expression of CART within MCH neurons, whereas orexin cells showed negligible CART expression. Concerning dynorphin expression, Fig. 4A and B shows no differences in the number of samples with positive dynorphin signals between MCH and orexin neurons, and the expression analysis presented in Fig. 5 indicates a relatively low level of dynorphin both in MCH and orexin neurons. In this multi-transcriptional profiling other transcripts were measured to study whether there might be another differentiation between MCH and orexin neurons,

1218


Fig. 4. Multi-transcriptional profiling of MCH and orexin neurons. (A, B) The numbers of samples containing positive and negative Real-time RT-PCR signalling for mRNAs of MCH, orexin, CART, dynorphin, Y5R, MC4R, OB-Rb, GR and GABAA Rα1 in neurons immunocytochemically identified as either MCH- or orexin-containing (total number of samples tested = 45). (C, D) The numbers of samples containing positive and negative Real-time RT-PCR results for VGLUT1, GAD-65 and GAD67 in MCH and orexin neurons (total number of samples tested = 27). A remaining number of samples showed aspecific signalling as validated by the quality of the dissociation curves.

Fig. 5. Absolute expression levels (C × E−Ct ) of all specific peptides, receptors and primary neurotransmitter markers measured in MCH and orexin neuronal samples. mRNA levels are expressed as absolute mean levels ± SD of 25 stained neurons. (∗ p < 0.05; Student’s t-test). Vertical axis is a log scale.


1219

reflecting a possible differential input from other brain regions including the arcuate nucleus as well as the general circulation. Although in Fig. 4A and B, a relatively low but reliable score of positive Real-time RT-PCRs is demonstrated for NPY-5 receptor (Y5R), melanocortin-4 receptor (MC4R), leptin receptor (isoform b) OB-Rb, glucocorticoid receptor (GR) and GABA-A receptor α1 subunit (GABAA Rα1) in MCH and orexin neurons, no significant differentiation was established at expression level of these receptors as shown in Fig. 5. This shows that both MCH and orexin neurons are targets for stimulatory and inhibitory (metabolic) signals coming from the arcuate nucleus, other brain regions, and the general circulation. Further transcriptional analysis indicated that MCH and orexin neurons can be either glutamatergic or GABAergic as demonstrated by the presence of VGLUT1, glutamic acid decarboxylase (GAD)-65 and GAD-67 in Fig. 4C and D. Among VGLUTs, particularly VGLUT2 appears to be a prominently expressed glutamate transporter protein in these neurons (Collin et al., 2003; Rosin et al., 2003), however, by this approach we only succeed in designing appropriate primer pairs for VLGUT1. As graphed in Fig. 4C and D, and Fig. 5, no significant differences were found in molecular composition of these primary neurotransmitter markers between MCH and orexin neurons.

DISCUSSION The complex organisation and extreme differentiation of many brain structures such as the hypothalamus demands to answer questions about intrinsic regulation and function of their individual neurons. Even neurons present within a defined area may be functionally different, which indicates that analysis of individual neurons is essential to learn more about their functional role. The present study demonstrates multi-transcriptional analyses after immunocytochemical identification of functionally overlapping hypothalamic systems at the individual neuron level. Although, this approach of LMPC can never prevent tiny contribution of adjacently intermingled cellular structures possibly containing extrasomal mRNAs, the specificity and sensitivity of detection with the current approach has shown to be far sufficient to study fixed or archival brain tissue samples of phenotypically defined neurons without the need for preamplification. Herein qualitative and semi-quantitative expression of multiple genes was examined, allowing the analysis of the identity and diversity of the individual MCH and orexin neuronal systems, which are orexigenic key players in relation to their regulation of energy homeostasis. Since in this study, transcriptional profiling can only be expressed in a qualitative or semi-quantitative way, it can be of special importance to measure peptide contents, because in neurosecretory cells and peptidergic neurons changes in peptide content ˜ et al., 1997; Harthoorn and release do not necessarily reveal mRNA changes (Castano et al., 2001). Although peptide secretion takes place at the extrasomal projection level outside the areas of measured peptide content, we show quantitative differences between MCH and orexin peptide storage. Within orexin neurons there is an increase in peptide staining intensity after fasting both at ZT2 and ZT14. The increased peptide content in MCH neurons after fasting only at the resting period at ZT2 and

1220


not in the early phase of the activity period indicates that there might be either a high regulated secretion or a minor change in gene expression of MCH after fasting at ZT14. With respect to possible diurnal variances, the results of peptide levels of MCH and orexin show that the peptide contents remain the same between ZT2 and ZT14. Other studies (Yoshida et al., 2001; Kiyashchenko et al., 2002; Zeitzer et al., 2003) have described a positive correlation between high release of orexins into the cerebrospinal fluid and the state of activity, however, about MCH nothing has been reported yet about an activity-dependent MCH secretion. One may argue that, when comparing these two systems, the orexin system represents a straightforward system that consistently responds to fasting and MCH neurons on the other hand respond in a more differential manner. The present multi-transcriptional profiling of MCH neurons confirms the previously reported specific co-expression of CART (Broberger, 1999; Vrang et al., 1999), which was principally found to be an anorectic peptide (Kristensen et al., 1998). It could be interesting to investigate whether CART together with MCH represents a ‘multifaced’ system of two antagonizing peptides with a potential cooperative mechanism of functioning by which the MCH system can fulfil its orexigenic function. Although, dynorphin is relatively low expressed in MCH and orexin neurons, no specific coexpression restricted to one of these two peptidergic systems could be established. A high amount of co-localization of dynorphins with orexin has indeed been described previously (Chou et al., 2001), however, the opposite that almost all dynorphinproducing neurons in the lateral hypothalamus must also contain orexin (Chou et al., 2001), is not consistent with our results regarding the co-expression of dynorphin within MCH neurons. Our present results demonstrate also the expression of Y5R, MC4R, OB-Rb, GR and GABAA Rα1 in MCH and orexin cells, which emphasize an interaction of both of these lateral hypothalamic systems with the arcuate nucleus, signals from the general circulation and the classical transmitter GABA. Y5R and MC4R expression among the MCH and orexin systems implies that they possess sensitivity for stimulatory as well as inhibitory arcuate substrates and indeed constitute ‘second-order’ components of the hypothalamic circuit that controls food intake and energy balance as reported previously (Broberger et al., 1998; Elias et al., 1998). Together with ˚ earlier studies (Hakansson et al., 1998, 1999), we show OB-Rb expression in MCH and orexin neurons, suggesting that these cell groups, portrayed as ‘second-order’ sites, may, via leptin, also receive direct adiposity signals from the general circulation. According to studies that have proposed a modulatory role of glucocorticoids on MCH and orexin expression (Stricker-Krongrad and Beck, 2002; Drazen et al., 2004), our results demonstrate that both MCH and orexin neurons can be sensitive for glucocorticoids by direct expression of GR. This gives emphasis to an important link between the glucocorticoid status and a plausible role of MCH and orexin in anticipating the activity period. Altogether, this provides proof of an existent interaction between MCH and orexin neurons and substrates from the general circulation. Expression of GABAA Rα1 by both the MCH and orexin neurons, show that GABA, presumably produced by other hypothalamic regions including the suprachiasmatic nucleus (SCN) that houses the main circadian pacemaker, can achieve its inhibitory role on the MCH and orexin cell systems. It has been reported that the SCN sends


1221

projections to MCH and orexin cells (Abrahamson et al., 2001), but we have been unable to verify a possible influence of the SCN particularly on orexin neurons, due to the absence of markers or receptors that reflect specific SCN input. Finally, this multi-transcriptional analysis also allowed us to study the concept that MCH and orexins are expressed as neuropeptides in hypothalamic neurons that may also use small, fast-acting neurotransmitters, such as glutamate and GABA. As shown here, MCH and orexin neurons express both glutamatergic and GABAergic markers. This means that these primary neurotransmitters can account for a differential synaptic activity and intercellular signalling of these two hypothalamic peptidergic systems that regulate energy homeostasis. Apart from the well-known excitatory properties of orexins (Horvath et al., 1999; Hagan et al., 1999; Antunes et al., 2001), the orexin as well as the MCH neuronal populations convey an ambivalent character, revealed by expression of VGLUT1 and GADs, the latter responsible for the synthesis of the inhibitory neurotransmitter GABA. In conclusion, the neuronal featuring outlined here have provided a characterization of the transcriptional profile of specific peptides, receptors and primary neurotransmitters within these defined MCH and orexin cell systems, which consequently renders a functional link between input and output to serve their integrative role in energy homeostasis. Especially the presence of both glutamate- and GABAmarkers among the MCH and orexin cell populations most likely points towards the existence of subpopulations of MCH and orexin neurons, probably with opposite effects. Further studies on a differential regulation of each system dependent for instance on their projections would be of great importance. ACKNOWLEDGMENTS We thank Katsuya Nagai of the Institute for Protein Research of Osaka (Japan) for his generous gift of the orexin antibodies, and Elly M. Hol for her helpful suggestions concerning LMPC and Real-time RT-PCR protocols. We are indebted to Henk Stoffels for preparing the illustrations. REFERENCES Abrahamson, E. E., Leak, R. K., and Moore, R. Y. (2001). The suprachiasmatic nucleus projects to posterior hypothalamic arousal systems. NeuroReport 12:435–440. Antunes, V. R., Brailoiu, G. C., Kwok, E. H., Scruggs, P., and Dun, N. J. (2001). Orexins/hypocretins excite rat sympathetic preganglionic neurons in vivo and in vitro. Am. J. Physiol. Regul. Integr. Comp. Physiol. 281:1801–1807. ¨ Backberg, M., Collin, M., Ovesjo, M. L., and Meister, B. (2003). Chemical coding of GABA B receptorimmunoreactive neurones in hypothalamic regions regulating body weight. J. Endocrinol. 15: 1–14. ¨ Backberg, M., Ultenius, C., Fritschy, J. M., and Meister, B. (2004). Cellular localization of GABA receptor alpha subunit immunoreactivity in the rat hypothalamus: Relationship with neurones containing orexigenic or anorexigenic peptides. J. Endocrinol. 16:589–604. Bahn, S., Augood, S. J., Ryan, M., Standaert, D. G., Starkey, M., and Emson, P. C. (2001). Gene expression profiling in the post-mortem human brain—No cause for dismay. J. Chem. Neuroanat. 22:79–94. Bittencourt, J. C., Presse, F., Arias, C., Peto, C., Vaughan, J., Nahon, J. L., Vale, W., and Sawchenko, P. E. (1992). The melanin-concentrating hormone system of the rat brain: An immuno- and hybridization histochemical characterization. J. Comp. Neurol. 319:218–245.

1222


¨ Broberger, C., De Lecea, L., Sutcliffe, J. G., and Hokfelt, T. (1998). Hypocretin/orexin- and melaninconcentrating hormone-expressing cells form distinct populations in the rodent lateral hypothalamus: Relationship to the neuropeptide Y and agouti gene-related protein systems. J. Comp. Neurol. 402:460–474. Broberger, C. (1999). Hypothalamic cocaine- and amphetamine-regulated transcript (CART) neurons: Histochemical relationship to thyrotropin-releasing hormone, melanin-concentrating hormone, orexin/hypocretin and neuropeptide Y. Brain Res. 848:101–113. Bujs, R. M., Chun, S. J., Niijima, A., Romijn, H. J., and Nagai, K. (2001). Parasympathetic and sympathetic control of the pancreas: A role for the suprachiasmatic nucleus and other hypothalamic centers that are involved in the regulation of food intake. J. Comp. Neurol. 431:405–423. Campbell, R. E., Smith, M. S., Allen, S. E., Grayson, B. E., Ffrench-Mullen, J. M., and Grove, K. L. (2003). Orexin neurons express a functional pancreatic polypeptide Y4 receptor. J. Neurosci. 23:1487–1497. ˜ J. P., Faught, W. J., Glave, E. E., Russell, B. S., and Frawley, L. S. (1997). Discordance of prolactin Castano, gene transcription, mRNA storage, and hormone release in individual mammotropes. Am. J. Physiol. 272:E390–E396. Chemelli, R. M., Willie, J. T., Sinton, C. M., Elmquist, J. K., Scammell, T., Lee, C., Richardson, J. A., Williams, S. C., Xiong, Y., Kisanuki, Y., Fitch, T. E., Nakazato, M., Hammer, R. E., Saper, C. B., and Yanagisawa, M. (1999). Narcolepsy in orexin knockout mice: Molecular genetics of sleep regulation. Cell 98:437–451. Chou, T. C., Lee, C. E., Lu, J., Elmquist, J. K., Hara, J., Willie, J. T., Beuckmann, C. T., Chemelli, R. M., Sakurai, T., Yanagisawa, M., Saper, C. B., and Scammell, T. E. (2001). Orexin (hypocretin) neurons contain dynorphin. J. Neurosci. 21:RC168. ¨ Collin, M., Backberg, M., Ovesjo, M. L., Fisone, G., Edwards, R. H., Fujiyama, F., and Meister, B. M. (2003). Plasma membrane and vesicular glutamate transporter mRNAs/proteins in hypothalamic neurons that regulate body weight. Eur. J. Neurosci. 18:1265–1278. Date, Y., Ueta, Y., Yamashita, H., Yamaguchi, H., Matsukura, S., Kangawa, K., Sakurai, T., Yanagisawa, M., and Nakazato, M. (1999). Orexins, orexigenic hypothalamic peptides, interact with autonomic, neuroendocrine and neuroregulatory systems. Proc. Natl. Acad. Sci. U.S.A. 96:748– 753. Drazen, D. L., Coolen, L. M., Strader, A. D., Wortman, M. D., Woods, S. C., and Seeley, R. J. (2004). Differential effects of adrenalectomy on melanin-concentrating hormone and orexin A. Endocrinology 145:3404–3412. Elias, C. F., Saper, C. B., Maratos-Flier, E., Tritos, N. A., Lee, C., Kelly, J., Tatro, J. B., Hoffman, G. E., Ollmann, M. M., Barsh, G. S., Sakurai, T., Yanagisawa, M., and Elmquist, J. K. (1998). Chemically defined projections linking the mediobasal hypothalamus and the lateral hypothalamic area. J. Comp. Neurol. 402:442–459. Emmert-Buck, M. R., Bonner, R. F., Smith, P. D., Chuaqui, R. F., Zhuang, Z., Goldstein, S. R., Weiss, R. A., and Liotta, L. A. (1996). Laser capture microdissection. Science 274:998–1001. Fink, L., Kinfe, T., Stein, M. M., Ermert, L., Hanze, J., Kummer, W., Seeger, W., and Bohle, R. M. (2000). Immunostaining and laser-assisted cell picking for mRNA analysis. Lab. Invest. 80:327– 333. Goncharuk, V. D., Van Heerikhuize, J., Dai, J. P., Swaab, D. F., and Buijs, R. M. (2001). Neuropeptide changes in the suprachiasmatic nucleus in primary hypertension indicate functional impairment of the biological clock. J. Comp. Neurol. 431:320–330. Hagan, J. J., Leslie, R. A., Patel, S., Evans, M. L., Wattam, T. A., Holmes, S., Benham, C. D., Taylor, S. G., Routledge, C., Hemmati, P., Munton, R. P., Ashmeade, T. E., Shah, A. S., Hatcher, J. P., Hatcher, P. D., Jones, D. N., Smith, M. I., Piper, D. C., Hunter, A. J., Porter, R. A., and Upton, N. (1999). Orexin A activates locus coeruleus cell firing and increases arousal in the rat. Proc. Natl. Acad. Sci. U.S.A. 96:10911–10916. ˚ Hakansson, M. L., Brown, H., Ghilardi, N., Skoda, R. C., and Meister, B. (1998). Leptin receptor immunoreactivity in chemically defined target neurons of the hypothalamus. J. Neurosci. 18:559–572. ˚ Hakansson, M., De Lecea, L., Sutcliffe, J. G., Yanagisawa, M., and Meister, B. (1999). Leptin receptorand STAT3-immunoreactivities in hypocretin/orexin neurones of the lateral hypothalamus. J. Neuroendocrinol. 11:653–663. Harthoorn, L. F., Oudejans, R. C., Diederen, J. H., Van de Wijngaart, D. J., and Van der Horst, D. J. (2001). Absence of coupling between release and biosynthesis of peptide hormones in insect neuroendocrine cells. Eur. J. Cell. Biol. 80:451–457. Hill, J., Duckworth, M., Murdock, P., Rennie, G., Sabido-David, C., Ames, R. S., Szekeres, P., Wilson, S., Bergsma, D. J., Gloger, I. S., Levy, D. S., Chambers, J. K., and Muir, A. I. (2001). Molecular cloning and functional characterization of MCH2, a novel human MCH receptor. J. Biol. Chem. 276:20125–20129.


1223

Horvath, T. L., Peyron, C., Diano, S., Ivanov, A., Aston-Jones, G., Kilduff, T. S., and Van den Pol, A. N. (1999). Hypocretin (orexin) activation and synaptic innervation of the locus coeruleus noradrenergic system. J. Comp. Neurol. 415:145–159. Kamphuis, W., Schneemann, A., Van Beek, L. M., Smit, A. B., Hoyng, P. F. J., and Koya, E. (2001). Prostanoid receptor gene expression profile in human trabecular meshwork: A quantitative realtime PCR approach. Invest. Ophthalmol. Vis. Sci. 42:3209–3215. Kiyashchenko, L. I., Mileykovskiy, B. Y., Maidment, N., Lam, H. A., Wu, M. F., John, J., Peever, J., and Siegel, J. M. (2002). Release of hypocretin (orexin) during waking and sleep states. J. Neurosci. 22:5282–5286. Kristensen, P., Judge, M. E., Thim, L., Ribel, U., Christjansen, K. N., Wulff, B. S., Clausen, J. T., Jensen, P. B., Madsen, O. D., Vrang, N., Larsen, P. J., and Hastrup, S. (1998). Hypothalamic CART is a new anorectic peptide regulated by leptin. Nature 393:72–76. Meijerink, J., Mandigers, C., Van de Locht, L., Tonnissen, E., Goodsaid, F., and Raemaekers, J. (2001). A novel method to compensate for different amplification efficiencies between patient DNA samples in quantitative real-time PCR. J. Mol. Diagn. 3:55–61. Peyron, C., Tighe, D. K., Van den Pol, A. N., De Lecea, L., Heller, H. C., Sutcliffe, J. G., and Kilduff, T. S. (1998). Neurons containing hypocretin (orexin) project to multiple neuronal systems. J. Neurosci. 18:9996–10015. Qu, D., Ludwig, D. S., Gammeltoft, S., Piper, M., Pelleymounter, M. A., Cullen, M. J., Mathes, W. F., Przypek, R., Kanarek, R., and Maratos-Flier, E. (1996). A role for melanin concentrating hormone in the central regulation of feeding behaviour. Nature 380:243–247. Ramakers, C., Ruijter, J. M., Deprez, R. H., and Moorman, A. F. (2003). Assumption-free analysis of quantitative real-time polymerase chain reaction (PCR) data. Neurosci. Lett. 339:62–66. Rosin, D. L., Weston, M. C., Sevigny, C. P., Stornetta, R. L., and Guyenet, P. G. (2003). Hypothalamic orexin (hypocretin) neurons express vesicular glutamate transporters VGLUT1 or VGLUT2. J. Comp. Neurol. 465:593–603. Saito, Y., Cheng, M., Leslie, F. M., and Civelli, O. (2001). Expression of the melanin concentrating hormone (MCH) receptor mRNA in the rat brain. J. Comp. Neurol. 435:26–40. Sakurai, T., Amemiya, A., Ishii, M., Matsuzaki, I., Chemelli, R. M., Tanaka, H., Williams, S. C., Richarson, J. A., Kozlowski, G. P., Wilson, S., Arch, J. R., Buckingham, R. E., Haynes, A. C., Carr, S. A., Annan, R. S., McNulty, D. E., Liu, W. S., Terrett, J. A., Elshourbagy, N. A., Bergsma, D. J., and Yanagisawa, M. (1998). Orexins and orexin receptors: A family of hypothalamic neuropeptides and G proteincoupled receptors that regulate feeding behavior. Cell 92:573–585. Schwartz, M. W., Woods, S. C., Porte, D., Seeley, R. J., and Baskin, D. G. (2000). Central nervous system control of food intake. Nature 404:661–671. Simone, N. L., Bonner, R. F., Gillespie, J. W., Emmert-Buck, M. R., and Liotta, L. A. (1998). Laser-capture microdissection: Opening the microscopic frontier to molecular analysis. Trends Genet. 14:272–276. Stricker-Krongrad, A., and Beck, B. (2002). Modulation of hypothalamic hypocretin/orexin mRNA expression by glucocorticoids. Biochem. Biophys. Res. Commun. 296:129–133. Van den Pol, A. N. (1999). Hypothalamic hypocretin (orexin): Robust innervation of the spinal cord. J. Neurosci. 19:3171–3182. Van der Sluis, P. J., Pool, C. W., and Sluiter, A. A. (1988). Immunochemical detection of peptides and proteins on press-blots after direct tissue gel isoelectric focusing. Electrophoresis 9:654–661. Vrang, N., Larsen, P. J., Clausen, J. T., and Kristensen, P. (1999). Neurochemical characterization of hypothalamic cocaine-amphetamine-regulated transcript neurons. J. Neurosci. 19:RC5. Yamanaka, A., Beuckmann, C. T., Willie, J. T., Hara, J., Tsujino, N., Mieda, M., Tominaga, M., Yagami, K., Sugiyama, F., Goto, K., Yanagisawa, M., and Sakurai, T. (2003). Hypothalamic orexin neurons regulate arousal according to energy balance in mice. Neuron 38:701–713. Yoshida, Y., Fujiki, N., Nakajima, T., Ripley, B., Matsumura, H., Yoneda, H., Mignot, E., and Nishino, S. (2001). Fluctuation of extracellular hypocretin-1 (orexin A) levels in the rat in relation to the light–dark cycle and sleep–wake activities. Eur. J. Neurosci. 14:1075–1081. Zeitzer, J. M., Buckmaster, C. L., Parker, K. J., Hauck, C. M., Lyons, D. M., and Mignot, E. (2003). Circadian and homeostatic regulation of hypocretin in a primate model: Implications for the consolidation of wakefulness. J. Neurosci. 23:3555–3560.