Pituitary Adenylate Cyclase-Activating Polypeptide Regulates Brain-Derived Neurotrophic Factor Exon IV Expression through the VPAC1 Receptor in the Amphibian Melanotrope Cell

Abstract

In mammals, pituitary adenylate cyclase-activating polypeptide (PACAP) and its receptors PAC1-R, VPAC1-R, and VPAC2-R play a role in various physiological processes, including proopiomelanocortin (POMC) and brain-derived neurotrophic factor (BDNF) gene expression. We have previously found that PACAP stimulates POMC gene expression, POMC biosynthesis, and α-MSH secretion in the melanotrope cell of the amphibian Xenopus laevis. This cell hormonally controls the process of skin color adaptation to background illumination. Here, we have tested the hypothesis that PACAP is involved in the regulation of Xenopus melanotrope cell activity during background adaptation and that part of this regulation is through the control of the expression of autocrine acting BDNF. Using quantitative RT-PCR, we have identified the Xenopus PACAP receptor, VPAC1-R, and show that this receptor in the melanotrope cell is under strong control of the background light condition, whereas expression of PAC1-R was absent from these cells. Moreover, we reveal by quantitative immunocytochemistry that the neural pituitary lobe of white-background adapted frogs possesses a much higher PACAP content than the neural lobe of black-background adapted frogs, providing evidence that PACAP produced in the hypothalamic magnocellular nucleus plays an important role in regulating the activity of Xenopus melanotrope cells during background adaptation. Finally, an in vitro study demonstrates that PACAP stimulates the expression of BDNF transcript IV.

PITUITARY ADENYLATE cyclase-activating polypeptide (PACAP) belongs to the vasoactive intestinal peptide (VIP), glucagon, secretin, and GHRH family of peptides (1). PACAP binds to three G protein-coupled PACAP/VIP receptors, PAC1-R, VPAC1-R, and VPAC2-R. PAC1-R has a 100-1000 times higher affinity for PACAP than VIP, whereas VPAC1-R and VPAC2-R have the same affinities for both peptides (2). PACAP and its receptors are widely distributed in the vertebrate central nervous system and peripheral organs, and control various physiological processes (see Refs. 3,4,5). In the mouse, PACAP stimulates proopiomelanocortin (POMC) gene expression and secretory activity of pituitary melanotrope cells (6), and also activates the expression of brain-derived neurotrophic factor (BDNF) in primary cultures of cortical neurons and astrocytes (7), indicating a role of PACAP in the control of neuronal plasticity. However, the physiological significance of PACAP regulation of neuroendocrine secretory activity and of BDNF expression-dependent neural plasticity is unknown.

A suitable physiological model to study the role of PACAP in the regulation of neural and neuroendocrine secretion and plasticity is the well-characterized background-adaptation process in the frog Xenopus laevis. In frogs placed on a black background, the melanotrope cells in the pars intermedia of the pituitary gland release α-MSH. This peptide stimulates the dispersion of melanin pigment granules in dermal melanophores, giving the animal a black appearance. In animals on a white background, the release of α-MSH is inhibited, and consequently, the pigment granules in the melanophore cell are aggregated around its nucleus, and the animal takes on a pale appearance (8,9). On a black background, the melanotrope cell diameter is about twice as large as on a white background (10), and the production of POMC and the secretion of its end product, α-MSH, are strongly enhanced (11). POMC biosynthesis and α-MSH secretion by Xenopus melanotropes are regulated by a wide variety of neurochemical factors, both classical neurotransmitters and neuropeptides (e.g. see Refs. 12 and 13). Recently, PACAP has been added to the list of neuropeptides acting on Xenopus melanotrope cells because pharmacological studies with isolated cells have shown that PACAP directly stimulates both α-MSH secretion and POMC biosynthesis (14). In this same study, immunocytochemical analysis revealed that nerve terminals in the pituitary neural lobe are the likely source of PACAP for controlling melanotrope cell function. Besides α-MSH, Xenopus melanotropes also produce BDNF, as shown with in situ hybridization and immunocytochemistry (15). The expression of BDNF is considerably enhanced in black-adapted frogs. BDNF and α-MSH are cosequestered in and presumably coreleased from melanotrope secretory granules (16). BDNF stimulates POMC biosynthesis in Xenopus melanotrope cells, implying that it acts as a stimulatory autocrine factor (15). This idea is supported by the observation that Xenopus melanotropes express the BDNF full-length TrkB receptor (17).

In the present study, we have tested the hypothesis that PACAP is involved in the regulation of Xenopus melanotrope cell activity during background adaptation and that part of this regulation is through the control of the expression of autocrine acting BDNF. To this end we have studied the expression profiles of the PACAP receptors PAC1-R and VPAC1-R in Xenopus melanotropes, examined PACAP immunoreactivity in neural lobes of white- and black-background adapted animals, and determined the effect of PACAP on BDNF gene expression in the melanotrope cell. Our results show that Xenopus melanotropes express VPAC1-R mRNA and that PACAP stimulates expression of the BDNF gene.

Materials and Methods

Animals

Young adults of the South African clawed frog X. laevis, aged 6 months, were reared in our laboratory under standard conditions, kept in tap water at 22 C, and fed beef heart and trout pellets (Touvit; Trouw, Putten, The Netherlands). To ensure full adaptation to the state of background illumination, animals were kept on a black or white background under incident continuous light for 3 wk. Animal treatment was in agreement with the Declaration of Helsinki and the Dutch law concerning animal welfare, as verified by the committee for animal experimentation of Radboud University Nijmegen.

Molecular cloning of X. laevis VPAC1-R

We first identified a presumptive frog VPAC1-R sequence in the genome database (Joint Genome Institute version 4.1; Joint Genome Institute; Walnut Creek, CA) of Xenopus tropicalis (which genome is fully known, in contrast to that of its close relative X. laevis) by searching for sequences with similarity to Rana ridibunda VPAC1-R, the only known amphibian VPAC1-R sequence (18). This yielded sequences in the X. tropicalis genome on Scaffold 869. To locate strong homology regions, we conducted a Basic Local Alignment Search Tool (BLAST) search of the X. tropicalis genome using the R. ridibunda sequences. Primers with 100% identity were then designed, namely: forward, 5′-TTCATCATGAGAGCCATCGC-3′; and reverse, 5′-GGCGAACATGATATAATGAAC-3′. With these primers and RT-PCR, a VPAC1-R cDNA fragment from the X. laevis brain and pituitary neurointermediate lobe (NIL) was amplified and subsequently cloned into pGEM-T Easy vector following the manufacturer’s instructions (Promega Corp., Madison, WI). This vector was used for sequencing with the sequencing GenomeLab DTCS Quick start kit in a Beckman Coulter CEQ8000 genetic analysis system (Beckman Coulter, Inc., Fullerton, CA).

RNA extraction and cDNA synthesis

After decapitation, freshly dissected NILs of 15 black-adapted animals were collected in Xenopus L15 (XL15) culture medium containing 67% Leibowitz medium (L15; Life Technologies, Paisley, UK), 0.1% kanamycin (Life Technologies), and 0.1% antibiotic/antimyotic solution (Life Technologies) with 0.08 mg/ml CaCl₂ and 0.2 mg/ml glucose (pH 7.4). The NILs were then rinsed three times in XL15 and once in XL15 containing 10% fetal calf serum (Life Technologies). NILs were then individually incubated in 48-well plates (Nunclon, Roskilde, Denmark), for 2 d at 22 C, each well containing 300 μl XL15 with 10% fetal calf serum. During incubation, 10 NILs were treated with 10⁻⁶ m neuropeptide Y (NPY) (American Peptide Co., Sunnyvale, CA) to keep POMC biosynthesis at a low level. On d 2, 10⁻⁵ m frog PACAP-38 (AnaSpec, San Jose, CA) (we previously showed that both types of PACAP, PACAP-38 and PACAP-27, stimulate α-MSH release with similar potency; see Ref. 14) was added to five of the NPY-treated lobes for 16 h. All NILs were then individually collected in 500 μl ice-cold Trizol (Life Technologies) and homogenized by sonification. Total RNA was extracted with chloroform and precipitated with isopropyl alcohol, dissolved in 25 μl ribonuclease-free H₂O, and measured with an Eppendorf Biophotometer (Vaudaux-Eppendorf AG, Basel, Switzerland). First-strand cDNA synthesis was performed with 1 μg RNA and 5 mU/μl random primers (Roche, Mannheim, Germany) at 70 C for 10 min, followed by double-strand synthesis in strand buffer (Life Technologies) with 10 mm dithiothreitol, 20 U Rnasin (Promega), 0.5 mm deoxynucleotide triphosphates (dNTPs) (Roche), and 100 U Superscript II reverse transcriptase (Life Technologies) at 37 C for 75 min and 95 C for 10 min.

RT-PCR and quantitative RT-PCR

The hypothalamus was microdissected from the brain of three black-adapted animals, and this tissue, in addition to NILs, distal lobes, and samples of peripheral organs (heart, lung, muscle, kidney, liver, and spleen; Table 1) were individually collected in 500 μl ice-cold Trizol and homogenized by sonification. Total RNA was extracted with chloroform and precipitated with isopropyl alcohol, dissolved in 25 μl ribonuclease-free H₂O, and measured with an Eppendorf Biophotometer. First-strand cDNA synthesis was performed with 1 μg RNA and 5 mU/μl random primers (Roche) at 70 C for 10 min, followed by the addition of first-strand buffer (Life Technologies), 10 mm dithiothreitol, 20 U Rnasin, 0.5 mm dNTPs, and 100 U Superscript II reverse transcriptase at 37 C for 75 min and at 95 C for 10 min.

Table 1.

Real-time RT-PCR analysis of the expression of VPAC transcript in different tissues in X. laevis

	VPACR1	GAPDH
Hypothalamus	24.35	18.29
Heart	25.95	20.49
Lung	25.59	20.99
Muscle	25.59	16.89
Kidney	30.24	22.64
Liver	25.36	16.36
Spleen	28.07	21.87
PD	25.51	21.26
NIL	24.36	19.12
NTC	40	40

Real-time RT-PCR on total RNA from different tissues. The Ct values are after a PCR run of 40 cycles. The PCR determination included an analysis of the expression of the housekeeping gene GAPDH. A high Ct value represents a low expression. PD, pars distalis; NTC, nontemplate control. 

To determine whether there is expression of the PAC1-R, RT-PCR was performed on the NIL and hypothalamic samples of the three animals. Each determination was performed in a total volume of 25 μl buffer containing 5 μl template cDNA, 3 mm MgCl₂, 0.625 U FastStart Taq DNA polymerase (Roche), 0.25 mm dNTPs, and 0.3 mm of each primer. The following primers were used for PAC1-R (accession no. AF187878): forward, 5′-TGGCAATCACAATCAGAATC-3′; and reverse, 5′-GTCACAGGCTTCAGAGTAATG-3′ (product size: 398 bp). The optimum temperature cycling protocol, determined with a programmable thermal cycler (Mastercycler gradient; Eppendorf, Hamburg, Germany), was 95 C for 30 sec, 60 C for 30 sec, and 72 C for 2 min. After PCR, the reaction products were run on a 2% agarose gel and visualized with ethidium bromide to check the length of the amplified cDNA.

To obtain a general overview of the level of expression of the VPAC receptor in various tissues, real-time RT-PCR was performed for the hypothalamic and other tissues collected from the three animals. Each determination was in a total volume of 25 μl buffer solution containing 5 μl template cDNA, 12.5 μl SYBR Green Master Mix (Applied Biosystems Benelux, Nieuwerkerk aan den IJssel, The Netherlands), and 0.6 μm of each primer pair: VPAC1-R (GenBank accession no. EU547209), forward 5′-TTCATCATGAGAGCCATCGC-3 and reverse 5′-TGGCCATGATGCAGTACTGG-3 (product size: 135 bp); and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primer pair (GenBank accession no. U41753), forward 5′-GCTCCTCTCGCAAAGGTCAT-3′ and reverse 5′-GGGCCATCCACTG TCTTCTG-3′ (product size: 101 bp). The optimum temperature cycling protocol was 95 C for 10 min, followed by 35 reaction cycles at 95 C for 15 sec and at 60 C for 1 min, using a 7500 Real Time PCR System (Applied Biosystems). For each mRNA, the cycle threshold (Ct) was determined, i.e. the number of cycles needed to reach an arbitrary fluorescence value (0.2) where Ct values of all mRNAs to be compared were within the linear phase of amplification.

Quantitative RT-PCR was performed on NILs of four black-adapted and four white-adapted animals in three independent quantitative RT-PCR runs (total 12 black-adapted and 12 white-adapted animals). The following primer pairs were used: BDNF (GenBank accession no. EU363497), forward 5′-CTATATTATCCAGAGTTTCAG-3′ and reverse 5′-CACTCTTCTCACCTGATGGAA-3′ (product size: 101 bp); and for GAPDH and VPAC1-R, the same primer sets were used as for real-time RT-PCR. The same cycling protocol was used as for real-time PCR. To correct for possible variations in melanotrope cell content among different NIL samples, the Ct values of VPAC1-R mRNA and BDNF mRNA were then adjusted to ΔCt values by subtracting the corresponding Ct value of the mRNA of the housekeeping enzyme GAPDH. The relative amount of mRNA (A_mRNA) for VPAC1-R or BDNF was then calculated as 2^−ΔCt, expressed in arbitrary units. The quantitative RT-PCR was performed in three independent experiments, which gave essentially the same results.

Immunocytochemistry

For immunocytochemistry, four black-adapted and four white-adapted frogs were transcardially perfused with 100 ml ice-cold 0.6% NaCl solution for 5 min, followed by 250 ml ice-cold Bouin’s fixative for 15 min. After dissection, brains with the pituitary gland attached were postfixed in Bouin’s fixative for 16 h at 4 C, washed in 70% alcohol to eliminate excess of picric acid for 24 h, further dehydrated through a graded series of ethanol, and embedded in paraffin. Coronal sections (7 μm) were mounted on poly-l-lysine-coated slides (Sigma Chemical, St. Louis, MO), allowed to air-dry for 16 h at 45 C, deparaffinized, and rehydrated. For immunodetection, the tyramide signal amplification technique was applied as follows. Sections were pretreated with 0.01 m sodium citrate (pH 6.0), and to quench endogenous peroxidase, incubated with 1% hydrogen peroxide in 0.1 m sodium PBS (pH 7.4) for 30 min. After rinsing in PBS, sections were incubated for 30 min in PBS containing 0.5% Triton X-100 (Sigma Chemical). To prevent aspecific binding, sections were, after rinsing in PBS, incubated for 1 h in incubation buffer, consisting of PBS containing 0.5% Triton X-100 plus 0.5% tyramide signal amplification blocking reagent (New England Nuclear Life Science Products, Boston MA), 2.5% normal goat serum (Vector Laboratories, Burlingame, CA), and 2.5% normal horse serum (Vector Laboratories). They were then rinsed in avidine/biotine blocking solution (Vector Laboratories) for 15 min, and incubated with monoclonal mouse anti-PACAP (generous gift from Dr. J. Hannibal, Copenhagen, Denmark) diluted 1:50 in incubation buffer for 16 h at 20 C. After rinsing in PBS, sections were incubated in biotinylated secondary horse-antirabbit antiserum (Vector Laboratories) diluted 1:200 in incubation buffer, rinsed in PBS, and incubated in streptavidin conjugated to horseradish peroxidase (1:100; New England Nuclear Life Science Products) for 30 min. After rinsing in PBS, they were incubated in fluorescein-conjugated tyramide solution (1:200 in amplification diluent; New England Nuclear Life Science Products) for 30 min, rinsed in PBS, coverslipped in Fluorsave (Calbiochem, La Jolla, CA), and examined with a Leica DMRBE light microscope (Leica Microsystems, Heerbrugg, Switzerland).

Specificity of the antiserum

The mouse monoclonal PACAP antiserum has been raised against the PACAP amino acid sequence 6–16 present in both PACAP-38 and PACAP-27, and recognizes these PACAP types with high specificity (19,20). In the present study, we did not find any immunosignal in brain or pituitary gland when this serum was omitted from the immunocytochemical procedure or was used after its preadsorption with excess synthetic rat PACAP-27 (AnaSpec).

Morphometry

Digital images were taken in three consecutive coronal sections per animal in the middle part of the neural lobe at a resolution of 1200 × 1600 dots/in., with a Leica DMRBE optical system and Leica DC 500 digital camera (Leica Microsystems) connected to an IBM computer (IBM Corp., Armonk, NY) running Scion Image software (version 3.0b; National Institutes of Health, Bethesda, MD). The density of the immunofluorescence signal was measured using ImageJ software (version 1.37; National Institutes of Health) and corrected for the background density outside the neural lobe in the unstained intermediate lobe, yielding the specific signal density (SSD) expressed in arbitrary units.

Statistics

Data were expressed as mean and sem per experimental group, and analyzed by the Student’s t test (α = 5%) using Microsoft Excel software (Microsoft Corp., Redmond, WA).

Results

Cloning and distribution of VPAC1-R

With RT-PCR, no amplification signal was detected for PAC1-R mRNA in the NIL, whereas there was a strong amplification signal in the hypothalamus (Fig. 1). We next studied the presence of the other receptor of PACAP, VPAC1-R. Including the primer sequences, the isolated and sequenced X. laevis VPAC1-R mRNA fragment contained 525 bp, and corresponded to the transmembrane domains II–VI of VPAC1-R. The sequence of VPAC1-R is submitted to GenBank (accession no. EU547209). The sequenced fragment showed strong homology with the corresponding VPAC1-R fragment of X. tropicalis (91%), R. ridibunda (80%), chicken (77%), mouse (74%), and rat (74%). The distribution of VPAC1-R mRNA in different brain regions and in various peripheral organs was studied using real-time RT-PCR. We found mRNA expression in all regions and organs studied, with similar degrees of expression (Table 1).

Figure 1.

Agarose gel electrophoresis of PAC1-R mRNA reaction product of RT-PCR on total RNA from the X. laevis NIL and hypothalamus (Hyp), using primers for the X. laevis receptor. Note absence of visible PAC1-R mRNA in NIL. Molecular weight markers, in base pairs, are indicated at the left.

Background-related expression of VPAC1-R mRNA in the NIL

Using quantitative RT-PCR, VPAC1-R mRNA expression in the NIL was tested for its possible regulation by the background light condition. As Fig. 2 shows, the VPAC1-R mRNA expression in NILs of white-adapted frogs appeared to be approximately four times as high as in the NILs of black-adapted animals (n = 4; P < 0.005).

Figure 2.

Quantitative RT-PCR analysis of VPAC1-R mRNA expression in NILs of black- (B) and white- (W) adapted X. laevis. The A_mRNA is expressed in arbitrary units on the basis of 2^−ΔCt. The two group means (n = 4), presented with the sem, differ statistically at P < 0.005 (**).

PACAP immunoreactivity in the pituitary neural lobe

Only weak, dispersed PACAP immunoreactivity was observed in neurohemal axon terminals in the neural pituitary lobe of black-adapted animals (Fig. 3A). However, in white-adapted animals, PACAP immunoreactivity appeared to be very strong and extensive (Fig. 3B). No PACAP immunoreactivity was found in the intermediate lobe, which contains the melanotrope cells, or in the distal part of the pituitary gland. To quantify the effect of background light intensity on PACAP immunoreactivity in the neural lobe, the SSD of PACAP immunofluorescence was determined (Fig. 4). The SSD appeared to be seven times higher in neural lobes of white-adapted animals (9.40 ± 2.79) than in those of black-adapted ones (1.35 ± 0.09; P < 0.05).

Figure 3.

Fluorescence immunocytochemistry of PACAP in the pituitary gland showing low immunoreactivity in neural lobe (n) of black-adapted (A) and high immunoreactivity in the neural lobe of white-adapted (B) X. laevis. Bar, 50 μm. d, Distal lobe; i, intermediate lobe.

Figure 4.

SSD of PACAP immunoreactivity in the neural lobe of the pituitary gland of black- (B) and white- (W) adapted X. laevis. The two group means (n = 4), presented with the sem, differ statistically at P < 0.05 (*).

Effect of PACAP on BDNF mRNA expression

To study if PACAP can stimulate the expression of BDNF mRNA in the NIL, BDNF mRNA expression was first suppressed by incubating NILs in NPY, enabling the reduced expression to be stimulated again by PACAP. Quantitative RT-PCR revealed that, compared with untreated control lobes, 10⁻⁶ m NPY decreased the amount of BDNF mRNA by 4.8 times (P < 0.0001), whereas 10⁻⁶ m NPY plus 10⁻⁵ m PACAP increased the amount of BDNF mRNA compared with NPY-treated NILs by 2.7 times (P < 0.005; Fig. 5).

Figure 5.

Quantitative RT-PCR analysis of BDNF mRNA expression in control NILs (C) and NILs treated with 10⁻⁶ m NPY (N) or 10⁻⁶ m NPY plus 10⁻⁵ m PACAP₃₈ (N + P). The relative A_mRNA is expressed in arbitrary units on the basis of A_mRNA = 2^−ΔCt. Means (n = 4) between two groups, presented with the sem, differ statistically at P < 0.005 (*) or P < 0.0001 (**).

Discussion

We have previously shown that PACAP stimulates α-MSH secretion in isolated melanotrope cells of X. laevis. Pharmacological studies have indicated that this action of PACAP would be through VPAC1-R/VPAC2-R, rather than PAC1-R, because PACAP and VIP were equipotent in stimulating the secretory activity of the melanotropes (14). Extending these findings, we here demonstrate the presence of VPAC1-R mRNA in the NIL of X. laevis, whereas the NIL did not show any expression of PAC1-R mRNA. This latter observation extends an in situ hybridization study showing that the Xenopus intermediate pituitary lobe lacks a hybridization signal for PAC1-R mRNA (21). This is the first report of the presence of a sequence with a coding region for VPAC1-R in the X. laevis pituitary gland. This sequence is strongly homologous to that of VPAC1-R in other nonmammalian as well as in mammalian species, and, furthermore, it is expressed in the central nervous system and in all peripheral organs studied. In fact, its brain and peripheral occurrence is similar to that of VPAC1-R in R. ridibunda (18), zebra fish (22), and mammals (23). This similarity underlines the importance of the pleiotropic actions of this receptor throughout the vertebrate class. In this study we could only identify one VPAC1-R, despite the fact that X. laevis has undergone whole-genome duplication. Therefore, we do not know if the quantitative RT-PCR is measuring one form or both forms of VPAC1-R expression. In addition, we cannot exclude the possibility that Xenopus possesses a VPAC2 receptor because such a receptor is reported for the frog Rana tigrina ruglosa (24).

The VPAC1-R expression is under control of the background light condition because we have demonstrated a substantially higher amount of VPAC1-R mRNA in the NIL of white-adapted than black-adapted frogs. Although the A_mRNA in melanotropes of white-adapted Xenopus is generally down-regulated compared with black-adapted frogs (15,17,25,26,27), an observation in line with the smaller size and lower cellular activity of such cells, VPAC1-R mRNA is not the only receptor mRNA that is up-regulated under white background condition. Placing frogs on a white background up-regulates TRH-R3 mRNA and CRH-R1 mRNA (28,29). Interestingly, PACAP, TRH, and CRH all stimulate α-MSH secretion from Xenopus melanotropes (12,14,30). Possibly, up-regulation of VPAC1-R, TRH-R3, and CRH-R1 is part of a mechanism by which melanotropes of white-adapted animals are sensitized to stimulatory input so that as soon as the animal is placed on a black background, the cells can quickly start to secrete α-MSH to achieve rapid skin color adaptation to the new environmental light situation.

Nerve terminals in the neural lobe are the source of a number of stimulatory neuropeptides, including TRH, CRH, and urocortin. These peptides are presumed to diffuse from the neural lobe into the pars intermedia to act upon the melanotropes (for review, see Ref. 13). Retrograde tracing studies with a hydrophobic fluorescent dye and horseradish peroxidase showed that the nerve terminals in the neural lobe originate exclusively from magnocellular neurons (31). Previous studies have demonstrated PACAP-containing axon terminals in the Xenopus neural lobe (14), and in situ hybridization studies have shown a complete lack of PACAP mRNA in the Xenopus pituitary gland but show the presence of PACAP-positive neurons in the magnocellular nucleus (21). The present study reveals that the neural lobe of white-adapted X. laevis has a dramatically higher PACAP content than that of black-adapted frogs. The low PACAP content in black-adapted frogs is likely due to strong PACAP release from the neurohemal axon terminals, whereas in the neural lobe of white-adapted animals, PACAP would be stored. Because we know that PACAP stimulates α-MSH release from isolated melanotrope cells (14), the demonstration of background-dependent changes in PACAP dynamics in the neural lobe provides strong evidence that magnocellular PACAP, released from nerve terminals of the pars nervosa, stimulates α-MSH release when frogs are placed on black background. This physiological, stimulatory action of PACAP on Xenopus melanotropes is likely not restricted to α-MSH secretion because our earlier in vitro study showed that PACAP also stimulates POMC gene expression and POMC biosynthesis (14).

In mouse cortical neurons and astrocytes, PACAP stimulates the expression of BDNF (7). From this observation and our observation that the Xenopus melanotrope cell sequesters and releases α-MSH together with BDNF (16), we have hypothesized that in Xenopus melanotropes, PACAP is involved in the coregulation of POMC and BDNF gene expression. Recently, Kidane found that the Xenopus BDNF gene possesses seven promoters, and analysis of the expression of promoter-specific exons revealed that BDNF transcript IV is strongly up-regulated in NILs of frogs placed on a black background (Kidane, A. H., unpublished data). Therefore, in the present study, we focused our attention on BDNF transcript IV. Our results show that PACAP stimulates BDNF mRNA expression, supporting our hypothesis that PACAP regulates the expression of both POMC and BDNF genes. Like the POMC gene (32), the BDNF gene has been duplicated, and both the two POMC and two BDNF transcripts are up-regulated to a similar degree by PACAP treatment (Kidane, A. H., unpublished results).

In conclusion, we provide evidence that PACAP plays an important role in regulating the activity of Xenopus melanotrope cells as a function of adaptation to background illumination. Because the neural lobe releases a wide variety of neuroactive factors that are able to stimulate melanotrope cell activity in vitro (12,13,14,30), it will be of interest to assess the interplay between PACAP and these factors in the integrative control of melanotrope cell activity and plasticity in X. laevis.