SEARCH

SEARCH BY CITATION

Keywords:

  • cAMP ;
  • desensitization;
  • dopaminergic neurons;
  • NURR1;
  • phenotypic plasticity;
  • tyrosine hydroxylase

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Pituitary Adenylyl Cyclase-Activating Polypeptide (PACAP) is a neuroactive peptide present in the avian retina where it activates adenylyl cyclase (AC) since early in development via PACAP receptors. The synthesis of cAMP in response to PACAP is observed since embryonic day 8/9 (E8/9). After E12, signaling via PACAP receptors desensitizes, reaching very low levels in the mature tissue. We show here that chronic administration of PACAP in vitro desensitizes PACAP-induced cAMP accumulation, while the administration of the PACAP antagonist (PACAP 6-38) re-sensitizes PACAP receptor/cyclase system in vitro and in vivo. Moreover, a twofold increase in the number of tyrosine hydroxylase positive (TH+) cells is observed after in vivo injection of PACAP6-38. NURR1, a transcription factor associated with the differentiation of dopaminergic cells in the CNS, is present in the chick retina in all developmental stages studied. The presence of NURR1 positive cells in the mature tissue far exceeds the number of TH+ cells, suggesting that these NURR1-positive cells might have the potential to express the dopaminergic phenotype. Our data show that if PACAP signaling is increased in mature retinas, plastic changes in dopaminergic phenotype can be achieved.

Abbreviations used
BSA

Bovine serum albumin

CMF

Calcium and magnesium-free saline

MEM

Minimum Eagle Medium

PACAP

Pituitary adenylyl cyclase-activating polypeptide

PBS

Phosphate buffered saline

TH

Tyrosine hydroxylase

TTBS

Tween 20 Tris-buffered saline

During development, after oocyte fertilization, intrinsic genetic cues, together with environmental signaling, initiate a series of modifications of the primary totipotent egg cell that will lead to the different phenotypes of the mature organism (Schoenwolf 2001). The use of stem cells in modern medicine has created hope for the treatment of ailments that, a few years ago, were considered impossible to be properly handled. This reinforces the need to study processes related to the control of cell differentiation. In the nervous system, this aspect is even more remarkable because of the fact that the mature central nervous system (CNS) is relatively less plastic than other tissues (Weissman et al. 2001). All replacement therapies in CNS however have to fulfill at least three important requirements. The first is that stem cells have to reach the lesioned site in order to re-populate the region with ‘new neurons’. Then, it follows a more complex requirement in which ‘new neurons’ must arborize to allow the proper reconstitution of a synaptic network. And critically, these new cells will have to express the correct neurotransmitter phenotypes to closely mimic the synaptic functions operating before the lesion. During the development of nervous tissue, this is accomplished basically via a very dynamic exchange of information between genetic, epigenetic, and environmental cues that are substantially modified in time. Therefore, in the mature CNS, one could consider that for stem cell therapies to be successful, it would be better if the lesioned CNS sites could recapitulate the developmental environment that originated the morphological and functional complexity of the original synaptic network.

The retina is a portion of the CNS placed in the back of the eye, readily accessible to experimental manipulation. This tissue is composed of six cell types, including a major glial system represented by Müller cells. Retinal cells are organized in the tissue into layers that intercalate cell bodies with regions highly dense in synapses, named plexiform layers. Several neurotransmitter phenotypes are defined in the retinal tissue and, as in other parts of the brain, the majority of its synapses uses glutamate and GABA as neurotransmitters. However, several other neurotransmitter systems are active in the tissue, modulating locally the processing of visual information operating at the retinal level (Dowling 1987; Herrmann et al. 2011; Zhang et al. 2011).

One of these modulatory neurotransmitter systems is the dopaminergic network, represented by a subtype of amacrine cells and their processes that, together with interplexiform cells in some species, constitute the main catecholaminergic circuitry of the tissue. In the chick retina, with an embryonic period of 21 days, amacrine cells are born between embryonic days 3 and 7 (E3/E7) (Gardino et al. 1993). However, tyrosine 3-monooxygenase (EC 1.14.16.2), commonly known as tyrosine hydroxylase (TH), the rate limiting enzyme responsible for dopamine synthesis in these cells is not detected in the tissue before embryonic day 11/12 (Gardino et al. 1993; Kubrusly et al. 2003; Reis et al. 2007), although the capacity of dopamine uptake and DAT expression can be detected as early as E6/7 (Kubrusly et al. 2003). If retinas at E7/8 are dissociated and the resulting cell population is maintained in dispersed cell culture for more than 10 days, very few, if any, cells express the enzyme TH. However, if these cultures are exposed to agents that increase cAMP levels (i.e. forskolin) cells expressing the TH phenotype are greatly increased in number and complexity, reaching levels close to the full complement of TH amacrine neurons observed in the mature tissue (Guimaraes et al. 2001). One of the natural agents that promote the expression of the TH phenotype in the developing avian retina is PACAP (Borba et al. 2005). This peptide, among other signaling pathways that were already described, very frequently activates adenylyl cyclase via PAC1 and VPAC receptors (Peeters et al. 1998; Vaudry et al. 2000). PAC1 receptors are present in the avian retina since E6 and remain relatively constant throughout maturity. However, PACAP mediated cAMP accumulation in the embryonic tissue is far greater at the early stages of retinal development than in the mature tissue (Borba et al. 2005), revealing that PACAP signaling in the retina desensitizes as the tissue differentiates. Here, we show that by re-sensitizing PACAP receptors-adenylyl cyclase coupling in the adult chicken retina we induce a two-fold increase in the number of cells expressing TH, with intense TH-positive (TH+) neurite extensions. Our data reveal, for the first time, that partially recapitulating an embryonic environment (increased coupling between PACAP receptors and adenylyl cyclase) in the mature retina is sufficient to induce plastic changes in the dopaminergic system of mature and differentiated nervous tissue.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Animals

Experiments were performed in compliance to Brazilian laws for the use of animals and approved by the Ethics Committee on Animal Experimentation of the Health Sciences Center of the Universidade Federal do Rio de Janeiro (CEUA/CCS/UFRJ) under the protocol number IBCCF035. Fertilized white Leghorn eggs were obtained from a local hatchery and kept in an appropriate incubator under a 12-h light and dark cycle until the day of use. Post-hatched animals were maintained in a free running condition with water and food ad libitum. Embryos were staged according to Hamburger and Hamilton (1992), and all animals were killed by instantaneous decapitation. The number of retinas used for each experimental group in all experiments was included in the figure legends as n.

Materials

Minimum Eagle Medium (MEM), UltraPure DNase/RNase-Free water and Trizol were purchased from Invitrogen (Calsbad, CA, USA). First-strand cDNA synthesis kit was purchased from GE Healthcare (Little Chalfont, UK). DNA-free kit from Ambion (Austin, TX, USA) was used. Primers were purchased from Integrated DNA Technologies, USA. Immobilon Western Chemiluminescent horseradish peroxidase (HRP) Substrate was purchased from Millipore (Billerica, MA, USA). Ophthalmic anesthetic (Anestalcon) and antibiotic (ciprofloxacin) were purchased from Alcon Laboratories (Fort Worth, Texas, USA). PACAP 38, PACAP 6-38, and PACAP 38 (16-38) – RIA Kit (cat# S-2166) were from Bachem (Torrance, CA, USA). Biotinylated secondary antibodies and Vectastain ABC Kit were purchased from Vector Laboratories (Burlingame, CA, USA). OCT embedding medium was from Sakura Finetek (Torrance, CA, USA). 3-Isobutyl-1-methylxanthine (IBMX, cat# I5879) was from Sigma (St. Louis, MO, USA). All other reagents were of analytical grade. All information about the primary and secondary antibodies used is described in Table S1.

Desensitization and re-sensitization in vitro

For experiments of desensitization of PACAP receptor/cyclase system, retinas from 9-day-old embryos (E9) were cut into 2 mm² explants and incubated for 12 h or 24 h with PACAP (10 nM) in MEM at 37°C buffered with 20 mM Hepes at pH7.3. After that cAMP was determined as described below. For re-sensitization experiments, retinal explants from E14 or E19 were cut into 2 mm² segments and incubated for 48 h at 37°C with PACAP6-38 (100 nM) in MEM buffered with 20 mM Hepes at pH7.3.

To investigate the effects of PACAP on cell number and neurite expansion of embryonic retinas in vitro, two stages were analyzed: E11 retinas were dissected and incubated in 10 mL of Dulbecco's modified Eagle's medium (DMEM)/5% FCS under agitation (45 rpm) for 24 h then used for flat mount immunohistochemistry analysis. Alternatively, E16 retinas were maintained as above, in the presence or absence of PACAP6-38 (100 nM) for 24 h, washed to remove the antagonist and then further incubated with PACAP for an additional 24 h. The tissue was then processed as above for TH immunohistochemistry.

Re-sensitization in vivo

For in vivo re-sensitization, animals which ranged from 1 to 5 days post-hatch (PH) were first anesthetized with ether and a topic anesthetic was applied to the corneas. Then, 5 μL of PACAP6-38 (100 μM) were injected into one of the eyes using a Hamilton syringe (25 μL). Considering the eye volume around 600 μL, we assumed a PACAP6-38 final concentration of 0.8 to 1.0 μM into the eye. Contralateral eyes were used as control and were injected with 5 μL of the vehicle. Antibiotic eye drops were used to prevent infection. After 24 h, retinas were used for cAMP measurement, western blot or immunohistochemistry in whole mount retinas.

Measurement of cAMP accumulation

Retinas were cut into 2 mm² explants and incubated for 15 min at 37°C in MEM buffered with 20 mM Hepes at pH7.3 and containing 0.5 mM IBMX. Then, PACAP (10 nM) was added and tissue explants were further incubated for additional 15 min. cAMP was assayed as previously published (Gilman 1970; Matsuzawa and Nirenberg 1975).

Immunohistochemistry

For whole mount retinas, we used a protocol modified from Dos Santos and Gardino (Dos Santos and Gardino 1998). Briefly, immediately after enucleation, retinas were gently detached from the pigmented epithelium and fixed by immersion in 4% (wt/vol) paraformaldehyde (PA) in 0.16 M phosphate buffer (PB), pH 7.2, for 1.5 h. After this period, retinas were rinsed in PB several times and processed for immunohistochemistry. Retinas were first incubated with 5% (wt/vol) bovine serum albumin (BSA) in 0.25% (vol/vol) Triton X-100 in phosphate buffered saline (PBS) at 37°C for 1 h. Then, they were incubated for three days with tyrosine hydroxylase (TH) antibody in 0.25% (vol/vol) Triton X-100 in PBS. After several washes with PBS, retinas were incubated for 2 h with biotinylated goat anti-mouse IgG diluted 1 : 200 in 0.25% (vol/vol) Triton X-100 in PBS. Further washes in PBS were made and the tissue was incubated with a complex of biotin-avidin diluted 1 : 50 in 0.25% (vol/vol) Triton X-100 in PBS for 2 h at 37°C. Finally, sections were rinsed in PBS twice for 10 min each, and incubated with 0.05% (wt/vol) 3,3′-diaminobenzidine and 0.01% (vol/vol) hydrogen peroxide solution for another 10–30 min. After several rinses, the slides were mounted with glycerol 40% (vol/vol) in PB.

For NURR1 detection throughout development, retinal transverse sections from embryos at different stages were used. Fixation and cryoprotection was performed as described above for whole-mount preparation. After cryoprotection, retinas were mounted in OCT embedding medium, frozen and cryosectioned. Sections (8 μm) were collected on gelatinized slides and stored at −20°C. Immunostaining for NURR1 was analyzed in alternate transverse sections of the eyes. First, sections were incubated with 5% (wt/vol) BSA in PBS for 1 h. Following, they were incubated in rabbit polyclonal antibody against NURR1 in 0.25% (vol/vol) Triton-X100 in PBS, overnight. Controls were obtained by omission of the primary antibody. Retinal sections were then rinsed in PBS and incubated with goat anti-rabbit biotinylated antibody in PBS for 2 h followed by incubation with streptavidin-Cy3 (Sigma, cat# C2821) at the dilution of 1 : 400 in PBS for 2 h. After several washes, sections were mounted in a saturated solution of n-propyl gallate in PBS.

In some retinal sections, we performed double labeling experiments for TH and NURR1. First, non-specific binding sites were blocked for 1 h with 5% (wt/vol) BSA and 0.25% (vol/vol) Triton X-100 in PBS. Sections were incubated overnight with a mixture of primary antibodies against NURR1 and TH diluted in PBS containing Triton X-100. Sections were then rinsed in PBS and incubated for 4 h in goat anti-mouse FITC and goat anti-rabbit biotinylated antibody followed by incubation with streptavidin-Cy3 as previously described. After washing, the sections were mounted using a saturated solution of n-propyl gallate in PBS. Slides were imaged in an Imager.M2 Zeiss microscope, equipped with an ApoTome module. Images were analyzed with Adobe Photoshop 7.0 (Adobe Systems, San Jose, CA, USA) and Zeiss Axiovision 4.8.2 software (Carl Zeiss MicroImaging GmbH, Jena, Germany). Manipulation was restricted to linear adjustments to the final images.

Cell quantification

Density maps of TH-labeled amacrine cells were obtained from flattened whole-mounted retinas for cell quantification. Maps of the retinal contour were obtained under a photographic amplifier and the location of each sampling area was plotted on X–Y coordinates. Sampling areas of 0.060 mm2 were used to count immunolabeled cells in Standard 20 Zeiss microscope. We used the systematic sampling method, where sampling areas were obtained at 1 μm distance on both axes. A total of 950 to 2400 cells were counted for each group in each experiment.

Protein detection by western blot

Retinas of post-hatched chickens were dissected in Calcium and Magnesium-free Saline (CMF) and homogenized with a tissue grinder in protein extraction buffer [20 mM Tris base, 10 mM MgCl2, 600 μM CaCl2, 500 μM EGTA, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride (PMSF), 5 μg/mL aprotinin, 2 μg/mL leupeptin, and 0.05% (vol/vol) Triton X-100]. Protein concentration was determined with Bradford method. Extracts were diluted in sample buffer [10% (vol/vol) glycerol, 1% (vol/vol) β-mercaptoethanol, 3% (wt/vol) sodium dodecyl sulfate and 62.5 mM Tris base] and boiled for 5 min. Approximately, 30 μg of protein from each sample were electrophoresed in 10% (wt/vol) sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred for 45 min to nitrocellulose membranes. In the following day, membranes were washed with 0.1% (vol/vol) Tween 20 Tris-buffered saline (TTBS) and blocked for 3 h with 10% (wt/vol) skim milk and 3% (wt/vol) BSA in TTBS. Membranes were incubated with primary antibody against TH in 1% (wt/vol) BSA in TTBS overnight. Next day, membranes were incubated with secondary antibody anti-mouse HRP in 1% (wt/vol) BSA in TTBS for 2 h. Then, membranes were stripped with stripping solution at 70°C for 30 min, membranes were then profusely washed with TTBS and blocked for 1 h with 5% (wt/vol) skim milk in TTBS. Membranes were incubated with primary antibody against α-tubulin in 1% (wt/vol) BSA in TTBS for 1 h and then incubated with secondary antibody anti-mouse HRP in TTBS + 1% (wt/vol) BSA for 30 min. Western Blots were developed with Immobilon Western Chemiluminescent HRP according to manufacturer's instructions, and densitometry was analyzed with Image J software (NIH, Bethesda, MD, USA).

Measurement of endogenous PACAP peptide content

Retinas from 10-day-old embryos and 1 to 5 days post-hatched (PH) animals were dissected in CMF and homogenized in 2 mg/mL BSA and 1 mM PMSF in 5 M acetic acid. Extracts were then centrifuged, lyophilized, and re-suspended in RIA buffer with PMSF at 4°C for 24 h. Endogenous PACAP was quantified through a radioimmunoassay according to the protocol provided by the manufacturer of the kit. In brief, samples and standard peptide provided by the manufacturer were mixed with PACAP antiserum and incubated overnight at 4°C. Then, 125I-peptide was added; tubes were vortexed and once more incubated overnight at 4°C. Goat Anti-Rabbit IgG serum and normal rabbit serum were added and tubes were incubated for 90 min at 25°C. RIA buffer was then added, the tubes were vortexed and centrifuged for 20 min at 1700 g. Supernatant was aspirated (except Total counts tubes) and assay tubes were counted in a Gamma counter. Measurements were performed in duplicate.

RT-PCR

Total RNA from E10 and PE were extracted with Trizol following manufacturer's instructions. Then, samples were incubated with DNAse I-RNAse-free at 37°C for 30 min and RNA quality and final concentration were determined with Nanodrop spectrophotometer (Wilmington, DE, USA), 260/280 nm and 260/230 nm ratios were always above 1,9. In addition a PCR reaction with RNA was always performed before cDNA synthesis to make sure DNA contamination were not present in RNA preparation.

First-strand cDNA synthesis was performed with First-strand cDNA Synthesis Kit following strictly the protocol provided by the manufacturer. PACAP PCR reaction was performed with oligonucleotides described by Mirabella and coworkers (Mirabella et al. 2002) (5′-GAT GGG ATC TTC AGC AAA GC -3′ and 5′-AAT ACG CTA CTA CTC GGC GTC CT -3′). RT-PCR was analyzed at exponential phase of amplification. PACAP amplification was performed with the following conditions: 94°C 30 s + 58.3°C 30 s + 72°C 1 min. GAPDH PCR reaction was performed with the following oligonucleotides (5′-AAA GTC GGA GTC AAC GGA TTT GG-3′ and 5′-GGG TTG GCA CAC GGA AAG C -3′): 94°C 30 s + 65°C 1 min + 72°C 1 min. These primers were designed with NCBI/Primer blast tool. Final amplification products were resolved in 1.2% (wt/vol) agarose gels stained with ethidium bromide.

Statistical analysis

Statistical analyses were performed with unpaired t-test, one-way or two-way anova followed by Bonferroni's multiple comparison post-hoc test in GraphPad Prism (GraphPad Software, San Diego, CA, USA).

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Our group has previously shown that the response of embryonic chick retina to PACAP (determined by cAMP accumulation) is 7-fold to 10-fold higher in the early stages of development than in the mature tissue (Borba et al. 2005). As shown in Fig. 1a, while retinal explants from E9 embryos increase cAMP levels by 8–10-fold in response to a 15-min incubation with 10 nM PACAP, retinas obtained from post-hatched animals displayed cAMP levels in response to the peptide less than 1.5-fold higher than control. Even though PAC1 receptors (Borba et al. 2005) and PACAP precursor mRNA (Fig. 1b) are present in early embryonic stages and adulthood, PACAP content in the retina is 87% higher in post-hatched chicken as compared to E9/10 (Fig. 1c).

image

Figure 1. Pituitary adenylyl cyclase-activating polypeptide (PACAP) expression and cAMP response in developing retina. (a) Accumulation of cAMP in response to 10 nM PACAP in embryonic day 9 (E9) and post-hatched (PH) chick retinas. (b) PACAP mRNA and (c) PACAP peptide content in both stages. ***< 0.001 in two-way anova followed by Bonferroni's post-hoc test. Data were expressed as mean ± deviation of the individual values from the mean in a and c. n = 2.

Download figure to PowerPoint

This indicates that the PAC1-adenylyl cyclase system of the retina is desensitized as the tissue matures. To confirm this possibility, retinal explants from E9 embryos were pre-treated for 12 h (Fig. 2a) or 24 h (Fig. 2b) with 10 nM PACAP, explants were extensively washed with PACAP-free medium and then stimulated for 15 min with 10 nM PACAP to measure cAMP accumulation. In parallel, control explants were maintained without the peptide. Although control explants show a 6-fold and 11-fold higher cAMP levels upon short-term PACAP exposure, the explants that were subjected to PACAP treatment for 12 h or 24 h responded to exposure to the peptide with only fourfold and threefold increase in cAMP levels, respectively (Fig. 2).

image

Figure 2. Chronic treatment with pituitary adenylyl cyclase-activating polypeptide (PACAP) reduces cAMP accumulation in response to a short-term stimulation with the peptide. Retinal explants at E9 were incubated with PACAP for (a) 12 h or (b) 24 h and cAMP accumulation was then analyzed after a 15 min-stimuli. Data were expressed as mean ± deviation of the individual values from the mean in a (n = 2) and mean ± SE in b (n = 5). ***< 0.001, **p < 0.01, *p < 0.05 in one-way anova followed by Bonferroni's post-hoc test. E9/10- embryonic day 9 or 10; PH- post-hatched.

Download figure to PowerPoint

Therefore, long-term exposure of retinal tissue to PACAP (10 nM) accelerates the desensitization of the PAC1-cyclase system. This data are consistent with the loss of response displayed by the retinal tissue to PACAP after several days of embryonic development, until the tissue reaches its maturity. As a corollary of this observation, one would expect that the desensitized state of the PAC1-cyclase system of the mature retina could be restored to the highly coupled embryonic condition, if one could interrupt endogenous PACAPergic activity in mature retina for a reasonable period of time. Retinal explants from E14 and E19 embryos were incubated for 48 h in medium containing 100 nM of PACAP receptors antagonist, PACAP6-38. Control explants were also incubated for the same period of time in regular incubating medium. Also, control experiments using freshly dissected retina at E16 and E21 (hatching animals) were used to compare with E14 plus 48 h and E19 plus 48 hrs, respectively. Figure 3 shows that, as observed before (Borba et al. 2005) (Fig. 1a), freshly dissected retinas from hatching chicken responded to PACAP with less than a 1.5-fold increase in cAMP accumulation after a 15 min-stimulus with the peptide (Fig. 3bi) and, similarly, a small increment in cAMP levels is observed in E16 retinas (Fig. 3ai). On the other hand, when retinas from E14 or E19 embryos were maintained for 48 h in control medium a sixfold and twofold increase in cAMP levels in response to PACAP were already observed (Fig. 3a and b, respectively) and this effect was similarly observed when E14 or E19 explants were maintained for 48 h in the presence of 100 nM PACAP6-38, washed to remove the antagonist and then stimulated with 10 nM PACAP for 15 min (Fig. 3a and b). Thus, preventing the interaction of PACAP with its receptor for several hours do induce re-sensitization of PACAP receptors-cyclase system to levels similar to those observed at early embryonic stages of the retina. In addition, the incubation of explants in a relatively large volume of control medium seemed to favor an increased response of the tissue to exogenous PACAP (Fig. 3), suggesting that diffusion of the peptide out of the tissue during the 48 h-period was sufficient to re-sensitize the system (PACAP receptors-cyclase).

image

Figure 3. Response to pituitary adenylyl cyclase-activating polypeptide (PACAP) is re-sensitized in vitro after long-term exposure of retinas to the antagonist. Retinal explants at (a) E14 or (b) E19 were maintained in vitro in control medium or incubated with PACAP6-38 (100 nM) for 48 h. After this time, the explants were washed and cAMP production was determined after stimulation for 15 min with PACAP. Responses observed in freshly dissected retinal explants at E16 and H (hatching) are shown in the insets (a′ and b′, respectively). Data were expressed as mean ± deviation of the individual values from the mean in a (n = 2) and mean ± SE in b (n = 4). ***< 0.001, **< 0.01, *< 0.05 in one-way anova followed by Bonferroni's post-hoc test.

Download figure to PowerPoint

Finally, we asked whether we could induce resensitization of the PACAP receptors-cyclase system in mature retinas of relatively old chickens in vivo rather than in vitro, as shown above. Post-hatched chicken (1 to 5 days after hatching) were used. Intraocular injections were performed, in one eye, with 5 μL of 100 μM PACAP6-38 (in aqueous solution) to reach an estimate final concentration of 800 nM. The other eye of the same animal was injected with the vehicle. The animals were kept in appropriate cages for 24 h, and were then killed and the retinas were removed and used for measuring PACAP-induced cAMP accumulation. Retinas from the control eye when subjected to PACAP (10 nM) stimulation showed a typical, low cAMP accumulation that did not differ from the non-stimulated tissue (Fig. 4). When retinas from the eye treated for 24 hours with PACAP6-38 were exposed to PACAP for 15 min, the accumulation of cAMP reached levels that were higher than that observed in non-treated retinas (Fig. 4). Although in PACAP6-38 pre-treated retinas the difference between basal and PACAP-stimulated cAMP production displayed no statistical significance, an increment of about 30% could be observed (Fig. 4). However, a fourfold increment was observed when net cAMP accumulation, which refers to the difference between cAMP levels in the PACAP-stimulated retinas minus cAMP in control retinas, was compared, (Fig. 4 inset).

image

Figure 4. Re-sensitization of the pituitary adenylyl cyclase-activating polypeptide (PACAP) receptor/cyclase in vivo. Post-hatched chicken received intraocular injections of 5 μL of PACAP6-38 (estimated final concentration = 800 nM). After 24 h, retinal explants were incubated with PACAP 38 (10 nM) for 15 min and cAMP production was determined. Data were expressed as mean ± SEM (n = 9). Inset shows the net increase in PACAP-induced cAMP accumulation after in vivo re-sensitization compared with basal cAMP in control eyes. *< 0.05 in one-way anova followed by Bonferroni's post-hoc test.

Download figure to PowerPoint

NURR1 is a transcription factor associated with the dopaminergic phenotype and is expressed in most cells programmed to develop this neurochemical property (Zetterstrom et al. 1997; Jankovic et al. 2005). In the retina, NURR1 expression is detected in the inner nuclear layer of the tissue as early as E9. As expected, NURR1 immunostaining is concentrated in cell nuclei (Fig. 5). At E5, no immunostaining for NURR1 is observed (Fig. 5a). As the tissue differentiates, the number of cells expressing this factor increases substantially until E16, remaining relatively constant through adulthood (Fig. 5b–e). The immunolocalization of TH revealed that all cells expressing this enzyme were also positive for NURR1 since E12, period when TH+ cells start to be detected in the avian tissue (Fig. 5f and g). However, the majority of NURR1 positive cells in the mature as well as in the embryonic tissue were not labeled for TH, suggesting that in the mature retina a large number of cells could potentially express the TH+ phenotype (Fig. 5h–i). Therefore, we hypothesized that after re-sensitizing PACAP receptors-cyclase system in the mature retinal tissue, cells expressing NURR1 could be induced to become TH+.

image

Figure 5. Expression pattern of NURR1 in the developing chick retina. Retinal sections at (a) E5, (b) E9, (c) E12, (d) E16, and (e) PH were immunostained for NURR1. Retinal section of (f, g) E12 and (h, i) PH chick retinas were double labeled for NURR1 (red) and tyrosine hydroxylase (green). Calibration bar = 10 μm. E, embryonic day; PH, post-hatched.

Download figure to PowerPoint

With the intent to analyze if this potential reserve pool of Nurr1+ cells could lead to acquisition of the TH+ phenotype, cultures from E12 (embryonic day 12) retinas that are highly responsive to PACAP (not desensitized) were treated for 24 h with PACAP or forskolin. We observed in both cases not only an increased number of TH+ cells but also an extensive expansion of their neurite network (Figure S1). Thus, in a stage where the tissue is highly sensitive to PACAP and also responsive to forskolin (Guimaraes et al. 2001), we can promote changes in the dopaminergic network.

In addition, ex vivo cultures of E16 chick retinas, which are already desensitized to PACAP stimulation as previously reported (Borba et al. 2005), were pre-incubated in medium containing PACAP 6-38 (to antagonize PACAP) for 24 h. Thereafter, tissue samples were washed to remove the antagonist and then further exposed to PACAP for an additional 24 h. Parallel experimental group was carried out without the antagonist and then stimulated with PACAP. Figure S2 shows that although TH+ cells from E16 retinas not exposed to the antagonist, as expected for this stage, do reveal clear neurites expanding from cell bodies, retinas that were maintained in the presence of the antagonist (PACAP 6-38) for 24 h and then stimulated with PACAP for additional 24 h, reveal an increase in the total length of TH+ neurites (of about 40%, data not shown). Here, we show that through re-sensitizing E16 retinas to PACAP in vitro we can promote a similar expansion of neurites seen in vivo (Fig. 6, described below).

image

Figure 6. In vivo re-sensitization with pituitary adenylyl cyclase-activating polypeptide (PACAP) receptors antagonist increases the number and arborization of tyrosine hydroxylase (TH)+ cells. Post-hatched chicken received intraocular injections of 5 μL of PACAP6-38 (estimated final concentration = 800 nM) and whole-mount retinas were immunostained for TH. (a, b) Control and (c, d) PACAP6-38-treated retinas are shown in two different magnifications. (e) The number of TH+ cells was analyzed (950-2400 cells were counted for each group) and (f, g) content of TH was determined by western blot. Data were expressed as mean ± SEM in e (n = 3) and in g (n = 11). ***< 0.001, *< 0.05 in Student's t-test. Calibration bar = 20 μm.

Download figure to PowerPoint

Post-hatched animals were injected in one eye with PACAP 6-38 and in the other eye with vehicle. After 24 h, the retinas were dissected out and prepared for TH immunostaining in a flat mount preparation. TH+ cells were counted in both eyes as described in methods. Figure 6a–b show a typical distribution of TH+ neurons in the retinas obtained from the control eye, revealing a density of cells equivalent to 20 TH+ cells/mm2 (Fig. 6e). In PACAP6-38-treated retinas (Fig. 6c–d), one can observe a higher number of TH+ neurons, that reached more than 40 TH+ cells/mm2 (Fig. 6e). In addition, an exuberant extension of neurites was detected in PACAP6-38 retinas as compared with the control tissue (Fig. 6b compared to Fig. 6d). Neither a particular pattern of TH+ cells distribution nor prevalence of specific TH cell types were observed after treatment. It is worth to mention that these are avascular retinas, therefore, it is unlikely that blood vessels interfere with the analysis.

Immunoblot detection of TH revealed a significant increase in the content of this enzyme in re-sensitized retinas (Fig. 6f–g). The data indicate that restoring PACAPergic activity is sufficient to induce a substantial change in the number of cells expressing TH and to increase their arborization or distribution of this enzyme.

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Desensitization of receptor systems is a common and natural process in biology which may be a result of over-stimulation of receptors by their agonists, as has been previously shown for VIP and PACAP receptors (Laburthe et al. 2002). Desensitization usually occurs because of receptor internalization, conformational changes of the receptor or uncoupling of receptors from their effectors. In several cases, more than one event is associated with down-regulation of receptor function (Lefkowitz 2007). It was previously suggested that GRK3 is involved with PAC1 desensitization in human retinoblastoma Y79 cells (Dautzenberg and Hauger 2001). In fact, it was demonstrated that GRK3 as well as GRK2 and GRK5 are expressed in chick retinas at least in some post-hatched ages (de Almeida Gomes and Ventura 2004). We have previously observed that PACAP-adenylyl cyclase system of the avian retinal tissue is highly effective in promoting cAMP accumulation during early embryonic period, however, as development progresses, the capacity of PACAP to promote cAMP accumulation reduces substantially (Borba et al. 2005). This phenomenon was not because of reduction in the content of PAC1 receptors, suggesting that desensitization of PACAP receptors-adenylyl cyclase system might occur because of increased exposure to endogenous PACAP as tissue matures.

Here, we show that when explants of embryonic retinas are chronically exposed to PACAP, peptide receptors do desensitize, as the response to acute peptide treatment detected through cAMP measurement is substantially reduced. Moreover, we show that mature retinal tissue, that is relatively insensitive to PACAP, can recapitulate the response of the embryonic tissue if retinas are incubated for 48 h with PACAP receptors antagonist, PACAP6-38. An interesting observation is the fact that retinas from post-hatched chicken do contain PACAP in higher concentration than the embryonic tissue, while mRNA content for PACAP precursor is similar in both post-hatched retina as well as the embryonic tissue.

PACAP is effective in inducing the expression of tyrosine hydroxylase in cultured retinal cells (Borba et al. 2005). Also, PACAP increases the activity of TH through phosphorylation of serine 40 of the enzyme as shown in bovine adrenal chromaffin cells (Bobrovskaya et al. 2007). Co-localization of PAC1 receptors and TH containing cells in sheep hypothalamus also suggests a functional relationship between this neuroactive peptide and the dopaminergic phenotype (Anderson et al. 2005). Therefore, PACAPergic signaling seems to be important to TH expression and activity in different biological systems.

The data reported above show that in the avian retina the amount of cells expressing NURR1, one of the transcription factors associated with and essential to dopaminergic phenotype determination (Zetterstrom et al. 1997; Jankovic et al. 2005; Smidt and Burbach 2007; Kadkhodaei et al. 2009), far exceeds the TH-positive amacrine cell population in embryonic as well as in the mature tissue. This is consistent with the possibility that the mature retina might have the potential of increasing its dopaminergic processing if proper signaling involved with the expression of the dopaminergic phenotype could stimulate NURR1 positive cells to express TH. In fact by injecting the eyes of post-hatched chicken with relatively high concentration of the PACAP receptor antagonist, we could duplicate the total number of TH cell population of the tissue. This is in agreement with the observation that retinas obtained from these eyes are significantly re-sensitized to respond to PACAP, after long-term exposure to PACAP6-38 in vitro. The data suggest that an increase in PACAPergic communication can induce cells to express the TH phenotype. It is unlikely that new neurons are being generated under this manipulation of PACAP receptor sensitivity, because at the stage used for these experiments all cell types of the retina have been defined and no proliferation is observed (Mey and Thanos 2000; Martins and Pearson 2008). Moreover, we could not detect BrdU incorporation after in vivo treatment of post-hatched chicken with PACAP6-38 (data not shown). Recently, we have shown that Müller glial cells from avian and mouse retinas can develop functional dopaminergic properties in cultures from which neurons have been eliminated. These cultured Müller cells not only respond to PACAP with cAMP accumulation (Kubrusly et al. 2005), but also express functional TH, which is capable of converting tyrosine into dopamine (Kubrusly et al. 2008). Therefore, one possible explanation for the increased number of TH+ cells in re-sensitized retinas could be that Müller cells acquired the dopaminergic phenotype. However, even though one cannot rule out this possibility, Müller cells in the intact tissue do not express NURR1 transcription factor (Kubrusly et al. 2008). Therefore, most, if not all of NURR1-positive cells in the retina, are likely to be non-Müller cells. This also suggests that interaction of Müller cells with neurons, or factors released by them, maintains the non-dopaminergic state of Müller cell, which indicates that NURR1 expression is a regulated process and can be activated by interfering with cell–cell interactions.

The apparent ‘reservoir’ of NURR1-positive cells in the retina is consistent with an increased number of cells expressing TH after mature retina is re-sensitized to respond to PACAP. Whether these new TH+ cells might play a role in retinal physiology is not known. The intense new network of neurites associated with the increased number of TH+ cells, as evidenced by immunohistochemistry for TH, indicates that these neurites may reflect increased dopaminergic activity in the tissue.

The treatment of organotypic cultures of E12 and more mature E16 retinas with PACAP (Figures S1 and S2) substantially increases the neurite extension of dopaminergic cells. However, it does not significantly increase the number of TH-positive neurons as observed in re-sensitized mature retinas in situ (Fig. 4). The reason for that is not known and requires further investigation. One possible explanation for this apparent inconsistency may be attributed to the fact that the increase in the number of cells expressing TH requires other factors that, together with PACAPergic signaling, allows the reservoir of NURR1 cells to become fully dopaminergic. When we transfer more mature retinas to culture flasks containing relatively large amounts of medium we certainly dilute up the internal milieu of the tissue, interfering with the steady state of factors concentration of the extracellular space of the tissue. Therefore, the maintenance of the tissue in situ favors the increase in the number of TH+ cells. Further work is now in progress to establish which factors might be required to complement the PACAP effect described above.

The data reported above suggest that forcing NURR1 positive cells to respond to agents that may induce the expression of TH may activate plastic changes in the dopaminergic system.

Several lines of evidence indicate that NURR1 is essential for the maintenance of mesencephalic dopaminergic neurons and that it may play a role in the pathogenesis of Parkinson's disease (Jankovic et al. 2005; Smidt and Burbach 2007). Therefore, finding which factors are capable of regulating the continuous expression of NURR1 would be important to maintain functional dopaminergic integrity in the brain. Interestingly, it was recently suggested that in the retinal tissue NURR1 might also be involved with the determination of subtypes of GABAergic amacrine cells (Jiang and Xiang 2009).

Our data may represent what some research groups called neurotransmitter phenotype plasticity or re-specification (Trudeau and Gutierrez 2007; Demarque and Spitzer 2012), although in our case alteration in neuronal excitability was not directly analyzed. In previous studies, we observed that light deprivation in post-hatched chicken can re-sensitize dopaminergic receptors to dopamine (de Mello et al. 1982). Therefore, it is possible that environment might also influence PACAPergic plasticity in the retina, as was previously observed in Xenopus hypothalamus (Dulcis and Spitzer 2008).

The data reported in this study may represent an alternative approach to increase the expression of the dopaminergic phenotype in CNS. By restoring the responsiveness of cells to PACAP we could increase the number of TH+ cells in the retina. Apparently, it could occur because there is a large population of NURR1 positive cells not entirely committed to the dopaminergic phenotype in the adult tissue. It is not known whether the same phenomenon applies to other areas of the brain. Evidence that PACAP has neurotrophic effects on mesencephalic dopaminergic neurons (Takei et al. 1998; Reglodi et al. 2004, 2006) suggests that the approach used for retinal tissue to increase its PACAPergic communication may also display similar effects in other areas of the CNS. As shown by different groups PACAP is also neuroprotective in the retinal tissue (Silveira et al. 2002; Endo et al. 2011; Seki et al. 2011), including for dopaminergic neurons (Szabadfi et al. 2012).

Neurotransmitter phenotype plasticity may represent a new approach for treatment of neurological disorders. Restoration of neuronal circuit function could be achieved by neurotransmitter re-specification. Despite all potential benefits, phenotype plasticity may also be an underlying factor in the pathophysiology of neuropsychiatric disorders. Duplication of Vipr2 gene, one of the PACAP receptor subtypes, was significantly associated with risk for schizophrenia (Vacic et al. 2011). PACAP was also described to be critically involved with the regulation of the hypothalamic–pituitary–adrenal axis regulating responses to emotional stressors (Agarwal et al. 2005; Tsukiyama et al. 2011). PACAP signaling also controls the biosynthesis of catecholamine enzymes in stress responses (Stroth and Eiden 2010). Perturbations in the PACAP-PAC1 pathway were associated to abnormal stress responses underlying post-traumatic stress disorder (Ressler et al. 2011).

Therefore, the results reported in this study may indicate that neurochemical plasticity involving PACAP and dopaminergic circuitry could have implications in the physiology and pathophysiology of the nervous system.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

This investigation was supported by grants from CAPES, CNPq, FAPERJ, INCT/INNT (Instituto Nacional de Neurociência Translacional), and Pronex. There is no conflict of interest to declare.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
FilenameFormatSizeDescription
jnc12121-sup-0001-FigS1-S2-TableS1.pdfapplication/PDF551K

Figure S1. In vitro treatment with agents that increase cAMP levels in responsive retinas enhances the number and arborization of TH+ cells.

Figure S2. In vitro resensitization wiht PACAP recepter antagonist increases the neuritic arborization of TH+ cells.

Table S1. Primary and Secondary antibodies used for Western Blot or Immunohistochemistry.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.