Neuronal somata and processes were observed after immunostaining against MCH and/or hypocretin. MCHergic neurons were located in the hypothalamus (Fig. 1). These neurons were intermingled with and were in close proximity to hypocretinergic neurons; however, these neuropeptides were not co-localized in the same cells (Fig. 2). The absence of co-localization was observed throughout the hypothalamus.#

MCHergic neurons were concentrated in the perifornical region of the lateral hypothalamic area, dorsal hypothalamic area, the posterior hypothalamic area and the zona incerta. An example of the distribution of both MCH and hypocretin-containing neurons of a representative cat is presented in Fig. 3. Both groups of neurons were located principally at the tuberal level, where there were 591.3±39.4 MCHergic vs. 197.9±12.1 hypocretinergic neurons per section (P<0.0001; two-tailed, paired, Student's t test, analyzed in seven cats). At this level, the peak density of MCHergic neurons was more lateral (Fig. 3F) and ventral (Fig. 3G) than that of hypocretinergic neurons. Compared with hypocretinergic neurons, the distribution of MCHergic neurons was less extensive in the anteroposterior axis; almost no MCHergic neurons were present at the mammillary, supraoptic and preoptic levels of the hypothalamus (Fig. 3). The total number of MCHergic and hypocretinergic neurons in the whole hypothalamus (bilateral) was estimated to be 25,000 and 11,200, respectively.#

Fusiform, bipolar and multipolar MCHergic and hypocretinergic neurons were observed. The mean largest diameter of the MCHergic and hypocretinergic neurons was similar (30.7±1.3 vs. 30.3±1.0 μm, respectively; analyzed in the perifornical area of the tuberal level). However, MCHergic neurons had a statistically significant smaller mean diameter (12.3±0.7 vs. 15.0±0.7 μm; two-tailed, unpaired, Student's t test, P<0.01).#

MCHergic and hypocretinergic varicose terminals were observed throughout the hypothalamus. Close apposition of hypocretinergic fibers onto MCHergic somata or proximal dendrites was frequently observed (Fig. 4A). MCHergic axodendritic and axosomatic fiber appositions onto hypocretinergic neurons were also present (Figs. 4B–E).#

The photomicrographs in Fig. 5 are examples of MCH and Fos immunoreactivity in the posterolateral hypothalamic area during different behavioral states. MCH was stained brown whereas Fos, which was stained black, concentrated in the nucleus. In view of published reports of Fos labeling in MCH-containing cells of the rat during AS rebound (Verret et al., 2003) or total sleep rebound paradigms (Modirrousta et al., 2005), it was an unexpected finding that, in our preparation, MCHergic neurons did not exhibit Fos immunoreactivity throughout the hypothalamus in any of the conditions studied (Fig. 5). This was true in spite of the presence of non-MCHergic Fos+ neurons that were intermingled with MCHergic cells. In very few sections, one or two MCHergic neurons presented Fos immunoreactivity (less than 1% of the MCHergic neurons of the sections); there were no statistical differences between the conditions studied. Fig. 6 presents data obtained from a cat that exhibited an AS induced by carbachol state (AS-carbachol). The AS-carbachol latency was 2 min and the episode lasted for almost 2 h (Fig. 6A); immediately thereafter, the animal was euthanized. A camera lucida drawing of a section of the hypothalamus of this animal shows that although a considerable number of Fos+ neurons were intermingled with MCH+ neurons, there were no double-labeled MCH+Fos+ neurons (Fig. 6B). Fig. 7 is an example of a cat euthanized after 1 h of QS rebound; in this period the animal state consisted of 90% QS and 10% quiet wakefulness (QW). A camera lucida drawing of a representative section illustrates that MCHergic neurons were not Fos immunoreactive (Fig 7B).#

Fos immunoreactivity during sleep and waking were analyzed in the hypocretinergic neurons in previous studies (Torterolo et al. 2001 2003ab). In the present report, we analyzed in sections double immunostained for hypocretin and Fos, the single-labeled non-hypocretinergic Fos+ neurons that were present in the hypocretinergic field (Fig. 8A), which was defined as the area that contained most hypocretinergic neuronal somata at the tuberal level (Fig. 8B). Because MCHergic neurons did not express c-fos, the single-labeled Fos+ neurons that were counted were neither hypocretinergic nor MCHergic. Fig. 8C shows that there was a larger number of non-MCHergic, non-hypocretinergic Fos+ neurons during active wakefulness with motor activity (AW with M) compared to other behavioral states. A reduction in the number of non-MCHergic, non-hypocretinergic Fos+ neurons during QS, compared with other behavioral states, was also observed. The following are the number of non-MCHergic, non-hypocretinergic Fos+ neurons per hemisection: AW with M, 199.7±21.1; AW without M, 80.8±12.8; QW, 86.8±13.4; QS, 40.7±6.9; AS-carbachol, 99.4±6.2 (*P<0.05; **P<0.01 compared with the other conditions; ANOVA Fisher).#

Details of the preparation of cats to monitor behavioral states and microinject drugs into the rostral pontine tegmentum have been previously reported (Torterolo et al., 2001). Briefly, under isoflurane anesthesia, screw electrodes were placed in the calvarium for the recording of the frontal and parietal electroencephalogram (EEG), and in the orbital bone to record the electro-oculogram; strut electrodes were directed stereotaxically to the lateral geniculate nuclei in order to monitor ponto-geniculo-occipital (PGO) waves. All electrodes were connected to a female Winchester plug. Two plastic tubes, which were used to maintain the animal's head in a stereotaxic position without pain or pressure, as well as the Winchester plug, were fixed to the skull with acrylic resin. In four animals, a 5-mm diameter hole was trephined overlying the posterior fossa in order to provide access for a microsyringe to microinject carbachol or saline into the NPO. The hole was filled with a bone-wax plug. At the completion of all surgical procedures, antibiotics were administered parenterally for three days. The skin margins surrounding the implant were kept clean and topical antibiotics were administered on a daily basis. After the animals had recovered from the preceding surgical procedures, they were adapted to the recording environment for a period of at least two weeks. After this period of adaptation, the animals did not show any signs of stress or discomfort in this restrained condition and if not disturbed rapidly fall asleep.#

The EEG, EOG, PGO and the electromyogram (EMG, recorded from two stainless steel wires inserted acutely into the posterior cervical muscles) were monitored, stored and employed for the determination of behavioral states (for details, see Torterolo et al., 2001).#

In order to study Fos immunoreactivity during sleep and waking states, the animals spent 1 to 2 h (i.e., the time needed for Fos proteins to reach an optimal degree of concentration; Dragunow and Faull 1989; Yamuy et al. 1993) prior to euthanasia either in a state of (1) quiet wakefulness (QW, n=3), (2) active wakefulness with motor activity (AW with M, n=3), (3) active wakefulness without motor activity (AW without M, n=3), (4) quiet sleep (QS, n=4) and (5) AS induced by carbachol (AS-carbachol, n=3). Details of the procedures required to maintain the animals in these behavioral states have been previously presented (Torterolo et al. 2001; Torterolo et al. 2003ab). Briefly, in experimental sessions held between 10 AM to 2 PM, cats were maintained in a head-restraining device, with continuous electrophysiological monitoring of their behavioral states during QW, AW-without M, QS and AS-carbachol. For the QW animals, if slow wave activity or spindles appeared, low noise and/or mild somesthetic stimulation (gentle touching of the animal's face) were immediately applied. In one animal, saline (0.2 μl) was microinjected in the NPO during quiet wakefulness, using the same procedures that are described below for the AS-carbachol animals. As previously noted, no clear differences were observed in the pattern of Fos expression in this cat compared with the other QW animals (Torterolo et al., 2001).#

The AW-without M group was maintained awake by continuous bilateral loud clicks. The QS group was sleep-deprived for 2 to 3 h; during the following 60–90 min period (prior to euthanasia) there was a sleep rebound, but AS was prevented from occurring. This group of animals spent about 90% (86 to 90%) of the time in QS during the 60–90 min prior to euthanasia. In the AS-carbachol group of cats, carbachol (0.8 μg in 0.2 μl of saline) was microinjected into the nucleus pontis oralis with a 2-μl Hamilton microsyringe during QS (NPO, at stereotaxic coordinates AP −2 to −3, L 1.5 to 2.5, H −3.5 to −5 according to Berman, 1968). The period of microinjection lasted for approximately 2 min. Following the administration of carbachol, AS episodes occurred with a latency of 6.7±2.2 min and 82.7±20.6 min of duration.#

The AW-with M animals (which were not restrained) were introduced to the experimental room for the first time on the day they were schedule to be euthanized; this room contained a variety of objects to attract their attention and induce motor (exploratory) activity. These animals were not recorded during this procedure.#

All animals were euthanized with an overdose of sodium pentobarbital (60 mg/kg) and perfused for subsequent immunohistochemical analyses (see Torterolo et al., 2001, for details). Thereafter, the brain was cryoprotected, frozen and serial sections of 20 μm were obtained. Single immunostaining procedures were performed to identify MCH or hypocretin-2 (Hcrt-2). Double immunohistochemical processes were undertaken to identify Fos and Hcrt-2, Fos and MCH, MCH and Hcrt-2, as well as MCH and Hcrt-1. Immunohistochemistry with Hcrt-1 or Hcrt-2 antibodies led, as expected, to similar results because these neuropeptides are co-localized in the cat (Zhang et al., 2002).#

Details regarding the immunohistochemical procedures that were used to identify Fos and Hcrt-2 have been presented previously (Torterolo et al., 2001). In order to identify MCH immunoreactivity, sections were incubated overnight with rabbit MCH antibody (Phoenix Pharmaceuticals, 1:4000) and normal donkey serum (NDS, 3%). Thereafter, the sections were rinsed and incubated for 90 min with biotinylated donkey anti-rabbit antibody (1:300) plus NDS. After another rinse, the tissue was incubated in the ABC complex (1:200) for 60 min and then exposed to diaminobenzidine tetrahydrochloride (DAB, 0.02%) and hydrogen peroxide (0.0015%); occasionally this reaction was enhanced using 0.6% nickel ammonium sulfate. Control omission experiments were performed by not exposing the tissue to the primary antibody. In addition, antiserum preabsorbed with an excess amount of MCH peptide (50 times) was also utilized as a control.#

Pyronin-Y counterstaining was utilized in immunolabeled sections, or thionin staining was utilized on adjacent sections to identify the cytoarchitecture of the analyzed areas.#

In order to examine the potential co-localization of MCH and Hcrt, the following double immunofluorescence procedures were performed. Free-floating sections were rinsed several times in PBST (0.1 M phosphate buffer saline with 0.25% Triton X-100). Thereafter, they were incubated simultaneously with goat anti-Hcrt-1 (Diagnostic Systems Laboratories, 1:500) and rabbit anti-MCH antibodies (Phoenix Pharmaceuticals; 1:2000) in PBST solutions for 3 days at 4 °C. After the sections were rinsed four times in PBST for a total duration of 30 min, they were simultaneously incubated for 3 h in PBST containing donkey anti-rabbit IgG conjugated with rhodamine (Jackson laboratories; 1:300) and donkey anti-goat IgG conjugated with fluorescein isothiocyanate (FITC, Jackson laboratories, 1:200). We also employed the reverse procedure; donkey anti-rabbit IgG conjugated with fluorescein isothiocyanate (FITC, Jackson Laboratories: 1:200) and donkey anti-goat IgG conjugated with rhodamine (Jackson Laboratories: 1:200). Finally, the sections were rinsed for 45 min with PBST and coverslipped with Vectashield fluorescent mounting media (Vector laboratories).#

Brain sections were examined either with conventional or fluorescent light microscopy; photomicrographs were obtained using a SPOT digital camera. Images were analyzed using Adobe PhotoShop software with a Power Macintosh G4 computer. The distribution of immunolabeled neurons was determined from drawings, which were obtained with a camera lucida.#

Hypocretinergic, MCHergic and MCHergic neurons that displayed Fos immunoreactivity (MCH+Fos+) were analyzed in two representative coronal sections from each cat at the mammillary (AP 8–9), tuberomammillary (AP 9–10), tuberal (AP 10–11), supraoptic (AP 13–14) and preoptic (AP 14–15) levels of the hypothalamus (according to Berman and Jones, 1982). Therefore, the analysis was performed in 10 representative sections, per cat and in 3 to 4 cats, per condition. Single labeled Fos+ neurons were also analyzed in sections that were double immunostained for both hypocretin and Fos at the tuberal level (AP 10–11), where MCHergic and hypocretinergic neurons were most highly concentrated (see Results); two representative sections per cat (four hemisections) and 3 to 4 cats per condition were utilized in this analysis. The results are presented as the mean±SEM of the number of neurons. Differences in the number of single-labeled Fos+ neurons or MCH+Fos+ neurons per condition were evaluated with the ANOVA and Fisher post hoc tests. Differences in the number of MCHergic and hypocretinergic neurons were determined by employing the paired two-tailed Student's t test in seven cats (the number of each group of cell per cat was paired); the differences in neuronal size were evaluated with the unpaired Student's t test. The criterion chosen to discard the null hypothesis was P<0.05. An estimate of the total number of MCHergic and hypocretinergic neurons was obtained utilizing the Abercrombie (1946) method; for this analysis, in one representative cat, counts were performed from coronal sections that were obtained every 100 μm for the study of hypocretinergic neurons, and every 200 μm for the examination of MCHergic neurons.#