Thyrotropin-releasing hormone (TRH; pGlu-His-Pro-NH2) has multiple, but transient, homeostatic functions in the brain. It is hydrolyzed in vitro by pyroglutamyl peptidase II (PPII), a narrow specificity ectoenzyme with a preferential localization in the brain, but evidence that PPII controls TRH communication in the brain in vivo is scarce. We therefore studied in male Wistar rats the distribution of PPII mRNA in the septum and the consequence of PPII inhibition on the analeptic effect of TRH injected into the medial septum. Twelve to 14% of cell profiles expressed PPII mRNA in the medial septum-diagonal band of Broca; in this region the specific activity of PPII was relatively high. Twenty to 35% of PPII mRNA-labeled profiles were positive for TRH-receptor 1 (TRH-R1) mRNA. The intramedial septum injection of TRH reduced, in a dose-dependent manner, the duration of ethanol-induced loss of righting reflex (LORR). Injection of the PPII inhibitor pGlu-Asn-Pro-7-amido-4-methylcoumarin into the medial septum enhanced the effect of TRH. The injection of a phosphinic TRH analog, a higher-affinity inhibitor of PPII, diminished the duration of LORR by itself. In contrast, the intraseptal injection of pGlu-Asp-Pro-NH2, a peptide that did not inhibit PPII activity, or an inhibitor of prolyl oligopeptidase did not change the duration of LORR. We conclude that in the medial septum PPII activity may limit TRH action, presumably by reducing the concentration of TRH in the extracellular fluid around cells coexpressing PPII and TRH-R1.
Thyrotropin-releasing hormone (TRH) is a small peptide (pGlu-His-Pro-NH2) that in mammals regulates anterior pituitary secretions, including that of thyrotropin (TSH) and prolactin, as well as multiple homeostatic brain responses (Gary et al., 2003). These homeostatic effects are transient in part because TRH has an elimination half-life of a few minutes in vivo either in serum or the brain (Jackson et al., 1979; Nagai et al., 1980; Spindel et al., 1981).
There is no definitive consensus on the pathway of TRH inactivation in the brain extracellular space in vivo. Various peptidases have been proposed as critical. TRH is hydrolyzed in vitro to pGlu-His-Pro by prolyl oligopeptidase (POP; EC 188.8.131.52), an enzyme with multiple in vitro substrates that is localized in the cytosolic and synaptosomal membrane compartments in the brain (O'Leary and O'Connor, 1995). However, it is not yet known whether the membrane-bound form of POP is an ectoenzyme or what the biological role of POP is in the brain in vivo (Myöhänen et al., 2009). Another enzyme, pyroglutamyl peptidase II (PPII; EC 184.108.40.206; TRH-degrading ectoenzyme), hydrolyzes the pyroglutamyl-histidyl bond of TRH in vitro. PPII specificity is narrow, with TRH possibly its main substrate in vivo; its expression and activity are higher in the brain than in any other organ. PPII is an ectoenzyme enriched in synaptosomes, thus apparently adequately localized to hydrolyze an extracellular substrate (Charli et al., 1998; Heuer et al., 1998b).
The in vivo relevance of PPII has been evaluated in the context of the regulation of TSH secretion. The intraperitoneal injection of PPII inhibitors does not change basal serum TSH levels, but it increases (or extends) TRH- or cold stress-induced serum TSH levels (Scalabrino et al., 2007; Sánchez et al., 2009). Given that PPII activity is low in the anterior pituitary (Vargas et al., 1987), it is not expressed on thyrotrophs (Cruz et al., 2008), and it does not regulate TRH effect on TSH secretion from anterior pituitary cells (Cruz et al., 2008), PPII may have a minor role in the adenohypophysis for the control of TSH secretion. In contrast, PPII is expressed in the median eminence of the hypothalamus on β2 tanycytes, thus being able to inactivate TRH once released into the proximity of portal capillaries, and its inhibition enhances the recovery of TRH released in vitro from the median eminence (Sánchez et al., 2009). Control of TSH secretion by PPII may thus occur preferentially at the site of release of TRH and not at the site of action.
Inhibition of PPII activity in brain slices enhances the recuperation of TRH in the extracellular medium (Charli et al., 1989; Scalabrino et al., 2007). The intraperitoneal injection of pGlu-Asn-Pro-d-Tyr-d-Trp-NH2, a dual PPII inhibitor/agonist of central TRH receptors, mimics and augments TRH action on rat spontaneous activity; its intravenous administration reverses pentobarbital-induced narcosis, mimicking a well known effect of TRH (Scalabrino et al., 2007). Yet, these data do not clarify the in vivo role of PPII in the brain, because the relative importance of PPII inhibition in the periphery and brain and TRH receptor activation cannot be disentangled. The efficient inactivation of TRH by PPII may explain why TRH analogs resistant to PPII action have improved potency in the brain compared with TRH (Kelly, 1995). However, the role of PPII in the brain extracellular space awaits in vivo confirmation.
In this study, we tested the hypothesis that PPII controls TRH effects in the medial septum (MS)-diagonal band of Broca (DBB). We focused on the MS-DBB, part of the arousal circuits (Berridge 2008), because TRH injection in this region produces a reproducible analeptic effect that can be used as a behavioral output of TRH activity (McCown et al., 1986). In the MS-DBB, TRH receptor 2 (TRH-R2) mRNA is concentrated in the medial part of the MS, which sends GABAergic projections to the hippocampus, whereas TRH receptor 1 (TRH-R1) mRNA levels are higher and concentrated in the lateral part of the MS-DBB, which generates cholinergic projections to the hippocampus (Heuer et al., 2000; O'Dowd et al., 2000). Levels of TRH binding sites are moderate and PPII mRNA levels are relatively low in the MS-DBB (Manaker et al., 1985; Heuer et al., 1998a). The analeptic effect of TRH involves cholinergic projection pathways, although other neurons may also be implicated (Horita, 1998). To understand the role of PPII in the MS-DBB, we determined whether cells coexpress PPII and TRH-R1 mRNAs, because MS TRH seems to regulate cholinergic neurons and the impact of PPII inhibitors on the analeptic effect of TRH, both injected into the MS-DBB.
Materials and Methods
Adult male Wistar rats, weighing 270 to 350 g and raised at the animal facility of the Instituto de Biotecnología, Universidad Nacional Autónoma de México, were naive to any previous treatment. Animals were maintained in standard environmental conditions (3–4 per cage; lights on between 7:00 AM and 7:00 PM; temperature 21 ± 2°C) and received ad libitum standard rat chow (2018S; Harlan, Indianapolis, IN) and tap water. Animal care and protocols were approved by the Animal Care and Ethics Committee of the Instituto de Biotecnología; animals were used according to the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, 1996).
In Situ Hybridization Histochemistry.
Serial 20-μm frozen brain coronal sections through the rostro-caudal extent of the septum were cut on a cryostat, adhered to slides treated with gelatin and poly-l-lysine, and stored at −70°C until prepared for in situ hybridization as described previously (de Gortari et al., 2006).
Sections were hybridized with a single-stranded [35S]UTP-labeled RNA probe complementary to the full-length rat TRH-R1 cDNA (Zhu et al., 2002) or a single-stranded [35S]UTP-labeled RNA probe complementary to the rat PPII cDNA (nucleotides 129–773; Schauder et al., 1994), each reduced to a 200- to 300-nucleotide size by alkaline hydrolysis. Hybridization was performed at 52°C as described previously (de Gortari et al., 2006). After posthybridization washes, sections were dipped into nitroblue tetrazolium autoradiography emulsion (Eastman Kodak Co,, Rochester, NY) diluted 1:1 in distilled water. After 30 days of exposure at 4°C, silver grains were developed, and sections were counterstained with hematoxylin.
In some experiments, we used a single-stranded digoxygenin-UTP-labeled RNA probe complementary to the rat PPII gene (2 μg/ml). Digoxygenin detection was performed essentially as described previously (Sánchez et al., 1997), except that the antidigoxygenin antibody was conjugated to alkaline phosphatase. Endogenous alkaline phosphatase was inactivated with 0.24 mg/ml levamisole (Sigma, St. Louis, MO), and a solution of nitroblue tetrazolium and 5-bromo-4-chloro-3-indolylphosphatase was used to develop the color (Roche Diagnostics, Mannheim, Germany).
For double in situ hybridization, sections were hybridized with both a single-stranded [35S]UTP-labeled RNA probe complementary to TRH-R1 gene and a single-stranded digoxygenin-UTP-labeled RNA probe complementary to the rat PPII gene (2 μg/ml), each reduced in size by alkaline hydrolysis. After posthybridization washes, slides were treated to detect the digoxygenin-UTP-labeled RNA probe, dried, treated with 2% parlodion (Thermo Fisher Scientific, Waltham, MA), dissolved in acetone, and dipped into autoradiography emulsion.
Specificity of hybridizations was confirmed by various criteria. Sense probes generated only background signals (see, for example, Fig. 1). With antisense probes, whatever the labeling used, signal distributions were consistent with those of PPII and TRH-R1 mRNAs (Heuer et al., 1998a). We detected TRH-R1 mRNA in the dentate gyrus (granular layer), but not in the thalamic reticular, ventral, and central nuclei; PPII mRNA was detected in the CA2 region of the hippocampus and the piriform cortex, but not in the thalamic reticular and paraventricular nuclei or layer 1 of the cerebral cortex (data not shown). In addition, for the digoxygenin-UTP labeled RNA probe, signals were abolished if an excess of unlabeled probe was added in the hybridization solution or the antidigoxygenin FAB fragment was omitted. We also observed that the double in situ hybridization procedure did not produce silver grains if the [35S]UTP-labeled RNA probe was omitted; this confirmed that the grains were not the result of a chemical reaction between tissue and emulsion.
Sections were visualized with a Zeiss Axioskop microscope (Carl Zeiss GmbH, Jenna, Germany) under bright-field or dark-field illumination and an ExwaveHAD video camera (Sony, Park Ridge, NJ). To count labeled profiles, the images obtained under bright-field illumination were analyzed with Mercator software (Explora Nova, La Rochelle, France), essentially as described previously (Sánchez et al., 2001). In brief, the contour of each nucleus was drawn manually with the mouse over the 5× light microscope image. A grid of identical squares was drawn automatically over each nucleus. A motorized stage moved to each consecutive square (at 40×). Each labeled profile was classified visually according to a defined category (see below) and given a distinctive mark. The marks in each square were registered by the computer, and the number of cells for each category was counted. The resulting photographic images and maps were stored.
For single ISH with a [35S]UTP-labeled RNA probe, we ascribed silver grains to a hematoxylin-stained nuclear profile if they localized on the nucleus or in close proximity. For sections hybridized with a digoxygenin-UTP-labeled RNA probe, labeled profiles were taken as positive if the blue purpura color was easily distinguished from the background. For double in situ hybridization, grains were ascribed to a color-labeled profile if they coincided with the color or were in close proximity. Profiles detected with a [35S]UTP-labeled RNA probe were considered positive if at least six grains were detected on the profile; they were separated in two populations: 6 to 10 grains (low levels of mRNA) or more than 10 grains (intermediate-high).
For each nucleus, we measured the percentage of profiles positive for PPII mRNA along the rostro-caudal extent of the septum, except for the septo-hippocampal nucleus in which we observed PPII mRNA-positive profiles but did not determine their proportion, because the number of sections was too small. Each septal nucleus was divided in 60 × 60 μm (or 120 × 120) squares, and we counted all labeled profiles in 20% of the squares taken randomly. For double ISH, all of the squares dividing the MS-DBB, and in the septo-hippocampal nucleus for comparison, were used to determine the percentage of PPII mRNA-positive profiles expressing TRH-R1 mRNA. For other nuclei, the number of sections was too small to give a reliable value. For each nucleus, values are mean of values at three to seven antero-posterior levels.
Induction of Narcosis by Ethanol and Reversal by TRH and/or Enzyme Inhibitors.
The injection of ethanol (Merck, Darmstadt, Germany) was performed during the light phase, between 10:00 AM and noon. The behavior of each animal was recorded on tape, and all assessments were based on this record. Time 0 is the time at which the intraperitoneal injection of ethanol terminated. Loss of righting reflex (LORR; animal in the prone position for more than 1 min) occurred in 2 to 3 min. We recorded the time from the termination of ethanol injection to the spontaneous recovery of the righting reflex (more than 15 s on the four legs); this was called the duration of LORR.
pGlu-His-Pro-NH2 was from Bachem Biosciences (King of Prussia, PA), and pGlu-Asn-Pro-7-amido-4-methylcoumarin (MCA) (U.S. patent 7,378,397; Kelly, 2008) and pGlu-Asp-Pro-NH2 were from Peptides International Inc. (Louisville, KY). Their purity (over 99%) was confirmed by reverse-phase high-performance liquid chromatography. The phosphinic analog of TRH [pGluΨ[P(O)(OH)]His-Pro-NH2; ΨTRH] was synthesized as described previously (Matziari et al., 2008); its purity was 93% in reverse-phase high-performance liquid chromatography. N-benzyloxycarbonyl prolyl prolinal (ZPP) was from Enzo Life Sciences, Inc. (Farmingdale, NY).
Intraperitoneal Injection of TRH.
Rats were handled once daily for 3 days. To induce narcosis a 2 g/kg of ethanol (12% w/v in pyrogen-free saline solution; Sigma) was injected intraperitoneally, and rats were kept in individual cages. TRH (3–30 mg /kg; 5 mg/ml saline) or saline was injected intraperitoneally 25 min after ethanol injection.
Intracerebroventricular Injection of TRH and/or pGlu-Asn-Pro-MCA.
One week before the experiment, under ketamine (100 mg/kg i.m.) (Anesket, Pisa Agropecuaria, Mexico City, Mexico)/xylazine (Rompun, Bayer Health Care, Leverkusen, Germany) (10 mg/kg i.m.) anesthesia, rats were placed in a stereotaxic device (David Kopf Instruments, Tujunga, CA), and a 22-gauge stainless-steel cannula (Plastics One, Roanoke, VA) was implanted into the right lateral ventricle through a burr hole in the skull. Stereotaxic coordinates in reference to interaural were: antero-posterior, +9.2 mm; lateral, −1.1 mm; depth, −3.6 mm (Paxinos and Watson, 2005). The cannula was secured to the skull with two stainless-steel screws and dental cement and temporarily occluded with a dummy cannula. The skin incision was disinfected with 2% chlorhexidine digluconate (Maver Labs, Mexico City, Mexico). A total of 1,200,000 units of ampicillin were injected. Rats were housed in individual Perspex cages (47 × 26 × 20 cm) and handled once daily for 3 consecutive days before the experimental day to reduce stress. If an animal seemed sick (reduced activity), we did not test it. One week later, rats were transported to the nearby experimental room, and ethanol was injected (3 g/kg i.p.). Thirty minutes later, a 28-gauge stainless-steel tube was inserted into the external guide cannula; it extended 1 mm below the cannula and was connected with polyethylene tubing (Plastics One) to a 20-μl syringe (Hamilton Co., Reno, NV). Intracerebroventricular injections of TRH (1–25 μg) and/or pGlu-Asn-Pro-MCA (5 μg) in 2 μl of saline solution were made over 2 min by a microprocessor-controlled infusion pump (MD-1001; BAS Bioanalytical Systems, West Lafayette, IN) via the indwelling cannula. The tip of the needle was left in place for 30 s before removal to prevent backward leakage.
Intraseptal Injection of TRH, TRH-like Peptides, and Inhibitors.
Rats were implanted and treated as described above, except for the following differences. A 26-gauge guide cannula was implanted at: antero-posterior, +9.2 mm; lateral, −1.7 mm; depth, −2.6 mm, with a 14° angle between the vertical and cannula axes. Twenty minutes after an intraperitoneal ethanol injection (2.7 g/kg) a 33-gauge stainless-steel tube was inserted into the external guide cannula; it extended 4 mm away from the tip of the external guide cannula. Either TRH (1–25 μg), pGlu-Asn-Pro-MCA (5–45 μg), pGlu-Asp-Pro-NH2 (5 μg), pGluΨ[P(O)(OH)]His-Pro-NH2 (1 μg), or ZPP (0.33 μg) or TRH combined with one of the inhibitors were injected in 2 μl of saline solution (0.5 μl/min) via the indwelling cannula. Where indicated, drugs were dissolved in artificial cerebrospinal fluid (aCSF; 140 mM NaCl, 3.35 mM KCl, 1.15 mM MgCl2, 1.26 mM CaCl2, 1.2 mM Na2HPO4, 0.3 mM NaH2PO4, and 0.1% bovine serum albumin, pH 7.4) or 10% DMSO in aCSF (DMSO-aCSF) instead of saline. To define the potential extent of drugs' distribution, a few animals naive to any treatment were injected with 2 μl of pontamine blue (4% in water and 0.02% Tween 20; Sigma); the maximum spread from the injection site was 500 μm; however, this value may be an underestimate. The intraseptal injection of 2 μl of saline solution (at 0.5 μl/min; a slow rate) did not produce any tissue lesion.
At the end of each experiment, animals were euthanized by decapitation. For intracerebroventricular and intraseptal injections, the brain was removed and rapidly frozen on powdered dry ice. Serial 20-to 25-μm frozen brain coronal sections through the rostro-caudal extent of the septum were cut on a cryostat (Brights, Huntingdon, UK); confirmation of cannula and microinjector position was made by observation of the tissue scar.
Tissue Dissections for Biochemical and RT-PCR Analyses.
The medial septum was obtained from a 0.8-mm (approximate thickness) frozen brain coronal slice, spanning a region from +9.2 to +10 mm [interaural coordinates (Paxinos and Watson, 2005)], using a sample corer (Fine Science Tools, Heidelberg Germany) with 0.5-mm internal diameter. Other brain regions (right and left lateral septum, frontal cortex, dorsal hippocampus, cerebellum, and ventral pons-medulla) were dissected by hand on the same or other coronal slices. Tissues were expelled into a microcentrifuge tube on dry ice.
Tissue was homogenized in 50 mM NaPO4 buffer, pH 7.5 (buffer A). Membranes were collected by centrifugation (12,000g; 15 min), the pellet was washed once with buffer A and 1 M NaCl, and centrifugation was repeated. Finally, the pellet was homogenized in buffer A and stored at −80°C until use. PPII activity was determined essentially as described previously (Vargas et al., 2000) except for a few changes: 40 μM pGlu-His-Pro-MCA (TRH-MCA) was used as substrate, in a coupled assay with excess dipeptidyl aminopeptidase IV (EC 220.127.116.11); assay buffer (buffer A) included 0.2 mM N-ethyl maleimide, an inhibitor of pyroglutamyl peptidase I (EC 18.104.22.168), and 0.2 mM bacitracin, an inhibitor of POP. POP activity was determined with 40 μM N-benzyloxycarbonyl-Gly-Pro-β-naphthyl amide (Bachem Biosciences) as a substrate in buffer A. For both assays, the activity was linear for at least 2 h and referred to membrane protein content.
Detection of PPII mRNA Species by RT-PCR.
Total RNA was isolated from brain tissue as described previously (Chomczynski and Sacchi, 1987). One microgram of total RNA was used in reverse transcription experiments. One-fifth of the reverse transcription reaction product was submitted to PCR amplification by using one unit of Taq polymerase (Boehringer Ingelheim GmbH, Ingelheim, Germany) and 10 pmol of each primer. The following primers were used: A, 5′-TCAATCAAACTGGCTACTTCAGA-3′, nucleotides 2051 to 2073 of rPPII cDNA sequence (numbers beginning with the first adenosine residue of the initiation codon atg); B, 5′-GCTCTTCATGTTGATAGGAAG-3′, nucleotides 2440 to 2420 of rPPII cDNA; C, 5′-GGAAGCTGGATCCTTGGAAAC-3′, nucleotides 2029 to 2049 of rPPII* cDNA; and D, 5′-GCAGTGTGTGCATTCACACAG-3′, nucleotides 2493 to 2473 of rPPII* cDNA. Primers A and B allowed the amplification of a 390-base pair rPPII RNA sequence spanning exons 11 to 15; primers C and D allowed the amplification of a 465-base pair rPPII* RNA sequence from exon 11 to intron 14. An annealing temperature of 62°C and 35 amplification cycles were used. DNA was fractionated by agarose gel electrophoresis and stained with ethidium bromide.
TSH or Prolactin Release from Primary Cultures of Anterior Pituitary Cells.
Cells were cultured as reported previously (Vargas et al., 1994) except that no thyroid hormone or insulin was added. On day 5 in culture, cells were washed twice with Dulbecco's minimum essential medium, and 500 μl of serum-free medium containing either vehicle (Dulbecco's minimum essential medium), TRH, pGlu-Asn-Pro-MCA, or ΨTRH were added. At the time points stated in Results, culture medium was collected to measure TSH or prolactin release by radioimmunoassay with National Hormone and Pituitary Program reagents (National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD).
Results are presented as mean ± S.E.M. Data were compared by analysis of variance followed by Bonferroni-Dunn post hoc test or an unpaired t test where indicated. Values of p < 0.05 were considered statistically significant.
Expression of PPII in the Septum.
PPII mRNA-positive profiles were detected with a digoxygenin-UTP-labeled antisense RNA probe in various septal nuclei. Their density (number of PPII mRNA profiles per area) was higher in the various lateral septum subdivisions (data not shown) than in the MS and horizontal DBB (hDBB) where many profiles were detected (Fig. 1A). PPII mRNA-positive profiles were not detected with the sense probe (Fig. 1A).
Additional experiments, performed with a [35S]UTP-labeled antisense RNA probe, produced data consistent with those obtained with the digoxygenin-UTP-labeled antisense RNA probe. The highest density of PPII mRNA (number of silver grains per area) was detected in the lateral septum (mainly in the intermediate and ventral regions). Levels were medium to high in the septo-fimbrial and septo-hypothalamic nuclei. Lower levels were detected in the dorsal lateral septum, and levels were low in the DBB and very low in the MS and septo-hippocampal nucleus (data not shown). In most of the septal nuclei, including the MS-DBB, a small proportion of hematoxylin-labeled profiles (10–20%) was PPII mRNA positive. These profiles were generally labeled with low levels of PPII mRNA (Table 1).
In the brain, among distinct PPII mRNA species at least one of them encodes PPII while another encodes a truncated version (PPII*), which may function as a dominant-negative form. PPII* is codified by an mRNA that is probably the result of an exon extension at the border between exon 14 and intron 14 (Chávez-Gutiérrez et al., 2005). Because the hybridization probe we used is complementary to exon 1, it does not discriminate mRNAs that code for PPII from those that code for PPII*; cells detected by in situ hybridization may not code for active PPII. We thus determined by RT-PCR which of these mRNAs is expressed in the septum. Both mRNAs were detected in the brain, as well as in the lateral and medial septum (Fig. 2A).
Whether the medial septum contains PPII activity was previously unknown. The distribution of the activity of PPII in various brain regions was determined. Consistent with previous reports, frontal cortex activity was high, and ventral pons-medulla activity was low (Vargas et al., 1987). Medial septum and lateral septum activities were intermediate between these values (Fig. 2B). Three independent experiments confirmed the value of medial septum PPII activity (data not shown).
Distribution of TRH-R1 mRNA in the Septum.
To determine the distribution of TRH-R1 mRNA, we used a [35S]UTP-labeled antisense RNA probe. We detected a very high density of TRH-R1 mRNA-positive profiles in the hDBB, a high density in the medial septum (Fig. 1B), vertical DBB (vDBB), and septo-hippocampal nucleus, and a low density in the intermediate and ventral lateral septum, septo-fimbrial nucleus, and dorsal lateral septum (data not shown).
Colocalization of PPII and TRH-R1 mRNAs in the Septum.
To analyze whether PPII and TRH-R1 mRNAs are coexpressed in the septum, sections were hybridized with both a [35S]UTP-labeled RNA probe complementary to the TRH-R1 mRNA and a digoxygenin-UTP-labeled RNA probe complementary to the rat PPII mRNA. PPII mRNA colocalized with TRH-R1 mRNA in various regions of the septum, including the DBB (Fig. 3, A and B). The percentage of PPII mRNA profiles colabeled with TRH-R1 mRNA varied according to the septal nucleus, from undetectable to a maximum of 49% in the septo-hippocampal nucleus; it was 20% in the MS-vDBB and 35% in the hDBB. TRH-R1 mRNA levels in the PPII mRNA-positive profiles were generally low; in the hDBB, they were equally distributed between low and high values (Table 1). Profiles colabeled with both mRNAs in the MS-DBB were segregated in two nonoverlapping zones, the hDBB or MS-vDBB. A representative map of these profiles in a section through the MS-DBB is shown in Fig. 3C.
Analeptic Effect of TRH.
An intraperitoneal 2 to 3 g/kg dose of ethanol produced rapid LORR that lasted for 60 to 120 min in a dose-dependent manner (data not shown). The intraperitoneal injection of TRH reduced the duration of LORR induced by 2 g/kg of ethanol; the effect was dose-dependent (Fig. 4A).
In intracerebroventricularly cannulated animals, an intraperitoneal 3 g/kg dose of ethanol, followed by an intracerebroventricular injection of saline solution, produced LORR that lasted for an average of 138 min. The intracerebroventricular injection of 5 μg of TRH 30 min after ethanol administration reduced the duration of LORR (Fig. 4B). An independent experiment gave a similar result (saline, 129 ± 8, n = 3; TRH, 97 ± 10 min, n = 4; t test, p = 0.06).
Injection of 1 to 25 μg of TRH into the medial septum 20 min after 2.7 g/kg ethanol reduced the duration of LORR, with a more potent and reproducible effect when the guide cannula and microinjector were oriented to avoid damage of the septo-hippocampal nucleus. The effect of TRH was dose-dependent (Fig. 4C). The positions of the tip of the cannula in this experiment are shown in Fig. 4D. In an independent experiment, 5 μg of TRH reduced the duration of LORR (saline, 106 ± 6; TRH, 70 ± 8.5 min, n = 4; p < 0.01, unpaired t test).
Analeptic Effect of PPII or POP Inhibitors.
pGlu-Asn-Pro-MCA is not a substrate of PPII, but it inhibits its activity (Ki 0.97 μM; Kelly et al., 2000). This was confirmed in vitro: 10 μM pGlu-Asn-Pro-MCA inhibited brain membrane PPII activity by 44%. We also determined whether pGlu-Asn-Pro-MCA is an agonist of TRH receptors. In primary cultures of anterior pituitary cells, which express TRH-R1, TRH had a time-dependent (between 15 and 180 min) and dose-dependent (between 5 ×10−10 and 10−7 M) stimulatory effect on the release of TSH or prolactin; there was no consistent effect of pGlu-Asn-Pro-MCA up to 5 × 10−6 M (data not shown).
The intracerebroventricular injection of 5 μg of pGlu-Asn-Pro-MCA had no analeptic effect by itself and did not modify the analeptic effect of 5 μg of TRH [saline, 138 ± 19 min, n = 4; TRH, 85 ± 16 *, n = 5; pGlu-Asn-Pro-MCA, 141 ± 8, n = 5; TRH + pGlu-Asn-Pro-MCA, 102 ± 11, n = 3; *, p < 0.05 (unpaired t test)].
The injection of pGlu-Asn-Pro-MCA (5, 15, or 45 μg) into the MS did not produce an analeptic effect (Fig. 5, A and B), suggesting that the analeptic effect of TRH probably was not caused by an osmotic, but rather by a receptor-mediated effect. The duration of LORR was significantly reduced by the combined treatment of 5 μg of TRH and 5 μg of pGlu-Asn-Pro-MCA compared with saline solution alone or 5 μg of TRH alone. The effect of TRH alone was not significant (Fig. 5B).
The TRH-like peptide [Asp2]-TRH does not inhibit PPII activity (Kelly et al., 2000). This was confirmed in brain membranes in vitro: control PPII activity, 100%; 1.28 mM [Asp2]-TRH, 108%. The intraseptal injection of 5 μg of [Asp2]-TRH increased slightly, but not significantly, the duration of LORR and did not enhance TRH action, whose effect was not significant (Fig. 5C).
ΨTRH, a phosphinic analog of TRH (Matziari et al., 2008), is a potent and specific PPII inhibitor (Cruz et al., 2008). At 2 μM it inhibited brain membrane PPII activity by 65%. We determined whether it is an agonist of TRH-R1 receptors. In primary cultures of anterior pituitary cells TRH had a stimulatory time-dependent (between 15 and 360 min) and dose-dependent (between 2 × 10−9 and 2 × 10−6 M) effect on prolactin release, but we did not observe any consistent effect of ΨTRH at the same doses (data not shown).
The intra-MS injection of ΨTRH (1 μg) produced an analeptic effect. If a small dose of TRH (1 μg), which did not produce a significant effect, was combined with ΨTRH, a significant reduction of the duration of LORR was observed (Fig. 5D). In two other independent experiments, we confirmed that ΨTRH (1 μg) had an analeptic effect by itself [pooled data, aCSF, 103 ± 9 min, n = 13; ΨTRH, 75 ± 8, n = 12; p < 0.05 (unpaired t test)].
To test whether the membrane-bound form of POP regulates TRH action, we used ZPP, a potent inhibitor of the soluble and membrane forms of POP (Tenorio-Laranga et al., 2008). ZPP was active in vitro; at a concentration of 400 nM, it inhibited the activity of POP in rat brain membranes by 100%. The injection of 0.33 μg of ZPP into the medial septum had no effect on the duration of LORR and did not change the effect of 2 μg of TRH (DMSO-aCSF, 129 ± 6.5; TRH, 77 ± 12*; ZPP, 110 ± 13; TRH + ZPP, 90 ± 18*; n = 4; *, p < 0.05 versus DMSO-aCSF).
TRH has many pharmacological actions in brain and can restore homeostatic perturbations (Yarbrough et al., 2007). Its in vivo effects probably are limited by peripheral and central degradation. Although it is well established that TRH has a short half-life in serum, and degradation-stabilized analogs are more potent than TRH, the mechanism of TRH inactivation in the brain extracellular space has not been definitively established. Biochemical and cell biological results strongly suggest that TRH is inactivated by PPII (Charli et al., 1998; Heuer et al., 1998b), but competing alternatives have been put forward, in particular the involvement of POP. The data presented in this article demonstrate that in the septum PPII is expressed in multiple nuclei, with 10 to 20% of the hematoxylin-labeled profiles positive for PPII mRNA; in the MS-DBB 20 to 35% of PPII mRNA-positive profiles were also TRH-R1 mRNA-positive. The data also suggest that PPII activity limits the analeptic action of TRH endogenously released from or injected into this region. Therefore, a pathway of TRH inactivation in the MS-DBB extracellular space may be its degradation by PPII, localized on the surface of TRH-R1 mRNA-positive cells.
A broad map of PPII mRNA distribution in the adult male Sprague-Dawley rat brain shows that levels are high in the lateral septum (especially in the dorsal part) but not in the medial septum-DBB, where a few scattered cells express low levels of PPII mRNA (Heuer et al., 1998a, 2000). Our data are in general agreement with these results, but we also show that in adult male Wistar rats the proportion of PPII mRNA-labeled profiles in the MS-DBB (12–14% of the hematoxylin-labeled profiles) is significant, although most have low levels of PPII mRNA. Because the focus of this study was not on profile numbers, we did not use unbiased stereology; thus, the values we report could have been affected by stereological factors. In addition, albeit unlikely, an unexpected cross-hybridization in the MS-DBB might have led to an overestimation of the number of PPII mRNA-labeled profiles. Furthermore, ambiguity in our and other PPII ISH data (Heuer et al., 1998a, 2000; de Gortari et al., 2006) stems from the fact that the probes used do not distinguish RNA species coding for full-length PPII from those coding for a truncated inactive and dominant negative variant (PPII*; Chávez-Gutiérrez et al., 2005). Because the amounts of both RNAs detected by RT-PCR are similar in the MS-DBB, the ISH results are probably a composite of signals from both species. Both might be expressed in the same cells, because the brain regional distribution of PPII mRNA detected by ISH with a nonselective probe matches that of PPII activity (Heuer et al., 1998a; de Gortari et al., 2006). Therefore, labeled profiles detected by ISH may express a mixture of the truncated and complete isoform of PPII. ISH with transcript-specific probes should be used to confirm this proposal. Because the activity of PPII in septum is intermediate between the lowest and highest brain region activities, it is likely that PPII* expression does not override full-length PPII expression, and most profiles detected by ISH express the active protein.
In agreement with the distribution of PPII and TRH-R1 mRNAs, we demonstrate that some PPII mRNA profiles are positive for TRH-R1 mRNA in the intermediate part of the lateral septum, the septo-hippocampal, and septo-fimbrial nuclei. In the MS-vDBB and the hDBB 20 to 35% of PPII mRNA profiles are TRH-R1 mRNA positive, and 40 to 50% of these have intermediate to high levels of TRH-R1 mRNA. The double-labeled profiles are localized in two anatomically separated groups. Whether they correspond to functionally distinct cell types remains to be clarified. These data fit with indirect evidence that PPII is postsynaptic in the brain, because TRH and PPII mRNAs are differentially distributed (Heuer et al., 1998a). However, it remains possible that for some profiles the apparent double-labeling represents signals originating from two cells that might partially overlay each other. In addition, many PPII mRNA-labeled profiles are negative for TRH-R1 mRNA in the septum; some may be TRH-R2 mRNA positive because it is expressed in the MS and DBB (Heuer et al., 2000; O'Dowd et al., 2000).
The acute injection of a large dose of ethanol affects the brain through multiple primary mechanisms, including enhancement of GABA-A receptor function, opening of G protein-activated inwardly rectifying K+ channels, inhibition of N-methyl-d-aspartate receptor activity, or L-type calcium channels (Spanagel, 2009). These actions lead to a generalized depression of neuronal activity in many brain regions and various behavioral effects, including sedation. The effects of ethanol may be opposed by homeostatic mechanisms, one of which may involve TRH, which can block G protein-activated inwardly rectifying K+ channels and tandem of P domains in a weak inwardly rectifying K+ channel-related acid-sensitive K+ channels (Yarbrough et al., 2007). Ethanol induces TRH release from septal tissue explants (Kucerová and Strbák, 2001) and enhances medial septum TRH levels at recovery from narcosis (Morzorati and Kubek, 1993), and we observed a nonsignificant down-regulation of PPII activity in the medial septum in animals sacrificed when they regain their righting reflex (100–120 min after ethanol injection) (I. Lazcano and J. L. Charli, unpublished work). These endogenous adjustments of TRH transmission may contribute to an exit from narcosis, because the exogenous application of TRH, either intraperitoneally, intracerebroventricularly, or into the medial septum, facilitates recuperation from ethanol-induced narcosis (McCown et al., 1986; Horita, 1998; our data). The cholinergic septo-hippocampal pathway may mediate the effect of TRH (Horita, 1998). Preliminary data suggest that some of the MS-DBB PPII cells are cholinergic (R. M. Uribe and J. L. Charli, unpublished work), in line with the cholinergic hypothesis. An alternative untested hypothesis is that TRH may activate the MS-DBB cholinergic cells that send projections to, and activate, a subpopulation of hypocretin/orexin neurons of the lateral hypothalamus, which are causal for awakening from sleep (Sakurai et al., 2005; Adamantidis et al., 2007).
A major finding is that the analeptic effect of exogenous TRH in the MS is potentiated by PPII inhibition. This conclusion is based on the following arguments: the PPII inhibitor ΨTRH, which is not an agonist of TRH-R1, enhances the effect of TRH; and the analeptic effect of TRH is also enhanced by pGlu-Asn-Pro-MCA, a PPII inhibitor that is not an agonist of the TRH-R1 receptor. This inhibitor is also poor at displacing [3H]-3-methyl-His2-TRH (a TRH-R agonist) binding in rat cortical homogenates (Kelly et al., 2002), a region rich in TRH-R2 mRNA (Heuer et al., 2000). Finally, an analog of pGlu-Asn-Pro-MCA, pGlu-Asp-Pro-NH2, which does not inhibit PPII activity, is unable to modify the analeptic effect of TRH.
We also observed that the intramedial septum injection of ΨTRH, a PPII inhibitor more potent than pGlu-Asn-Pro-MCA, significantly reverses the narcosis induced by ethanol. Because Phe2-TRH and Tyr2-TRH, in vitro substrates of PPII putatively present in brain, are not analeptic (Hinkle et al., 2002), this effect may be caused by an enhancement of endogenous TRH concentration in the MS-DBB extracellular space; additional experiments are required to confirm this interpretation.
Injection of TRH in other medial regions of the brain also produces analepsia, albeit with less potency than in the MS-DBB; this was clearly shown in the case of pentobarbital-induced narcosis (Kalivas and Horita, 1980). Because pontamine blue may underestimate the spread of injected drugs, it remains possible that TRH and PPII inhibitors diffused out of the MS and affected additional targets, but the significant levels of PPII activity in the MS, and failure of the intracerebroventricular injection of pGlu-Asn-Pro-MCA to enhance the analeptic effect of intracerebroventricular TRH, support the parsimonious interpretation that PPII inhibition in the MS may be sufficient to produce analepsia. Even if additional regions contributed to the effects of PPII inhibitors, our data support the concept that PPII activity controls TRH turnover in the brain extracellular space. This conclusion is consistent with a preliminary report that the intracerebroventricular injection of 5 μg of pGlu-Asn-Pro-MCA enhances the number of wet dog shakes induced by intracerebroventricular TRH (Kelly et al., 2001).
The spatial relationship between sites of peptide release, action, and inactivation is poorly understood, but is likely an important determinant of ectopeptidase capacity for regulating peptide action (see Introduction). We propose that in response to ethanol TRH release from the MS-DBB activates TRH-R1 receptors on the surface of cells expressing PPII. The proximity of PPII and TRH receptors may facilitate the capacity of PPII to limit TRH action and control TRH-R desensitization. It makes it more likely that PPII can control TRH action than if it is not expressed on TRH-R-positive cells. However, it remains possible that the colocalization of PPII with TRH-R1 (or TRH-R2) is not an essential prerequisite for the suggested role of PPII, and PPII expression by nearby cells that do not express TRH-R would be sufficient to control TRH action.
Although our data support the notion that PPII inactivates TRH once released into the brain extracellular space, there may be additional ways of efficient removal of TRH. Inhibition of POP enhances TRH levels in specific brain regions; this may be caused by a change in the intracellular turnover of TRH (Tenorio-Laranga et al., 2011). POP may hydrolyze TRH in the brain extracellular space because TRH injected intracerebroventricularly is rapidly degraded to pGlu-His-Pro (Spindel et al., 1981), and the oral administration of the POP inhibitor ((S)-2-[[(S)-2-(hydroxyacetyl)-1-pyrrolidinyl]carbonyl]-N-phenylmethyl)-1-pyrrolidinecarboxamide) (JTP-4819) together with subcutaneous TRH, at doses at which each agent alone has no effect, improves retention time in a passive avoidance test in scopolamine-treated rats (Toide et al., 1995). These data, however, contradict the evidence that the hydrolysis of exogenous TRH in hypothalamic slices produces a metabolite signature that is not consistent with POP but is with PPII activity (Méndez et al., 1999). Moreover, ZPP, a potent POP inhibitor, does not change the recuperation of TRH released from rat hypothalamic slices (Charli et al., 1987). Finally, membrane POP is possibly intracellular (Myöhänen et al., 2009). Because we observed that POP inhibition does not change narcosis time or TRH-induced recovery from narcosis, our data suggest that POP does not participate in TRH turnover in the medial septum extracellular space.
In conclusion, we have obtained data that suggest the analeptic effect of TRH is controlled by PPII activity in the medial septum-diagonal band of Broca. PPII activity may limit the capacity of TRH to activate TRH-R1 receptors involved in relaying homeostatic information to the circuits regulating exit from narcosis by controlling the extracellular turnover of TRH. PPII therefore seems relevant in both a neurohormonal (Sánchez et al., 2009) and neuromodulatory (this study) context. This result strengthens the rationale to develop PPII inhibitors or TRH-like receptor agonists resistant to hydrolysis by PPII for the exploration of central nervous system disorder therapies.
Participated in research design: Joseph-Bravo and Charli.
Conducted experiments: Lazcano, Uribe, Martínez-Chávez, and Vargas.
Contributed new reagents or analytic tools: Matziari.
Performed data analysis: Lazcano, Uribe, Martínez-Chávez, and Charli.
Wrote or contributed to the writing of the manuscript: Lazcano, Uribe, Joseph-Bravo, and Charli.
We thank Dr. Patricia de Gortari and Dr. Enrique Reynauld for useful discussions; Miguel Cisneros, Sergio Gonzalez, Karla Juarez, and Alan Jiménez for technical assistance; and Dr. Athanasios Yiotakis for the phosphinic analog of TRH.
This work was supported in part by the Consejo Nacional de Ciencia y Tecnología [Grant 61804] (to J.-L.C.); and the Dirección General de Asuntos del Personal Académico of the Universidad Nacional Autónoma de México [Grant IN221109] (to J.-L.C.).
I.L. is a student in the Programa de Posgrado en Ciencias Biológicas de la Universidad Nacional Autónoma de México.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- thyrotropin-releasing hormone
- TRH receptor
- artificial cerebrospinal fluid
- diagonal band of Broca
- horizontal DBB
- vertical DBB
- dimethyl sulfoxide
- in situ hybridization
- loss of righting reflex
- medial septum
- polymerase chain reaction
- prolyl oligopeptidase
- pyroglutamyl peptidase II
- recombinant PPII
- reverse transcription
- N-benzyloxycarbonyl prolyl prolinal
- Received January 20, 2012.
- Accepted April 20, 2012.
- Copyright © 2012 by The American Society for Pharmacology and Experimental Therapeutics