Weaver cerebellar granule neurons show altered expression of NMDA receptor subunits bothin vivo andin vitro

RM. Expression of the N-methyl-D-aspartate receptor NR1 splice variants and NR2 subunit subtypes in the rat colon

N-methyl-D-aspartate (NMDA) receptors and the expression of their different splice variants and subunits were previously characterized in the brain and spinal cord. However, knowledge on the NMDA receptor expression and function in the enteric nervous system is limited. Previous work suggested that NMDA receptors were involved in a rat model of visceral hypersensitivity. The aim of this study was to characterize the expression of the NMDA receptor NR1 splice variants and the NR2 subunit subtypes in the rat colon. We visualized the expression of NR1 protein in the rat submucosal and myenteric plexuses. The NR1 splice variants found in the colon of rats lacked the N1 and C1 cassettes and contained the C2 and C2' cassettes (NR1(000) and NR1(001)). The NR2B and NR2D subunits were also found in the rat colon. Moreover, NMDA receptors in the rat colon were heteromeric, since NR1 was co-localized with NR2B and NR2D subunits using fluorescent immunohistochemistry. The identification of the NMDA receptors in the enteric nervous system could lead to the development of drugs that selectively modulate bowel function.

Selective up-regulation of NMDA-NR1 receptor expression in myenteric plexus after TNBS induced colitis in rats

Background

N-methyl-D-aspartic acid (NMDA) spinal cord receptors play an important role in the development of hyperalgesia following inflammation. It is unclear, however, if changes in NMDA subunit receptor gene expression in the colonic myenteric plexus are associated with colonic inflammation. We investigated regulation of NMDA-NR1 receptor gene expression in TNBS induced colitis in rats. Male Sprague-Dawley rats (150 g–250 g) were treated with 20 mg trinitrobenzene sulfonic acid (TNBS) diluted in 50% ethanol. The agents were delivered with a 24 gauge catheter inserted into the lumen of the colon. The animals were sacrificed at 2, 7, 14, 21, and 28 days after induction of the colitis, their descending colon was retrieved for reverse transcription-polymerase chain reaction; a subset of animals' distal colon was used for two-dimensional (2-D) western analysis and immunocytochemistry.

Results

NR1-exon 5 (N1) and NR1-exon 21 (C1) appeared 14, 21 and 28 days after TNBS treatment. NR1 pan mRNA was up-regulated at 14, 21, and 28 days. The NR1-exon 22 (C2) mRNA did not show significant changes. Using 2-D western analysis, untreated control rats were found to express only NR1₀₀₁ whereas TNBS treated rats expressed NR1₀₀₁, NR1_011, and NR1₁₁₁. Immunocytochemistry demonstrated NR1-N1 and NR1-C1 to be present in the myenteric plexus of TNBS treated rats.

Conclusion

These results suggest a role for colonic myenteric plexus NMDA receptors in the development of neuronal plasticity and visceral hypersensitivity in the colon. Up-regulation of NMDA receptor subunits may reflect part of the basis for chronic visceral hypersensitivity in conditions such as post-infectious irritable bowel syndrome.

Long-term NR2B expression in the cerebellum alters granule cell development and leads to NR2A down-regulation and motor deficits

N-methyl-D-aspartate receptor (NMDAR) composition in granule cells changes characteristically during cerebellar development. To analyze the importance of NR2B replacement by NR2C and NR2A subunits until the end of the first month of age, we generated mice with lasting NR2B expression but deficiency for NR2C (NR2C-2B mice). Mutant phenotype was different from NR2C knock-out mice as loss of granule cells and morphological changes in NR2C/2B cerebellar architecture were already evident from the second postnatal week. Increased NR2B subunit levels led also to a gradual down-regulation of cerebellar NR2A levels, preceding the development of motor impairment in adult animals. Therefore, cerebellar NR2A is important for proper motor coordination and cannot be replaced by long-term expression of NR2B. Consequently, the physiological exchange of NMDA receptor subunits during cerebellar granule cell maturation is important for accurate postnatal development and function.

Regulated appearance of NMDA receptor subunits and channel functions during in vitro neuronal differentiation

The schedule of NMDA receptor subunit expression and the appearance of functional NMDA-gated ion channels were investigated during the retinoic acid (RA) induced neuronal differentiation of NE-4C, a p53-deficient mouse neuroectodermal progenitor cell line. NR2A, NR2B, and NR2D subunit transcripts were present in both nondifferentiated and neuronally differentiated cultures, while NR2C subunits were expressed only transiently, during the early period of neural differentiation. Several splice variants of NR1 were detected in noninduced progenitors and in RA-induced cells, except the N1 exon containing transcripts that appeared after the fourth day of induction, when neuronal processes were already formed. NR1 and NR2A subunit proteins were detected both in nondifferentiated progenitor cells and in neurons, while the mature form of NR2B subunit protein appeared only at the time of neuronal process elongation. Despite the early presence of NR1 and NR2A subunits, NMDA-evoked responses could be detected in NE-4C neurons only after the sixth day of induction, coinciding in time with the expression of the mature NR2B subunit. The formation of functional NMDA receptors also coincided with the appearance of synapsin I and synaptophysin. The lag period between the production of the subunits and the onset of channel function suggests that subunits capable of channel formation cannot form functional NMDA receptors until a certain stage of neuronal commitment. Thus, the in vitro neurogenesis by NE-4C cells provides a suitable tool to investigate some inherent regulatory processes involved in the initial maturation of NMDA receptor complexes.

Involvement of non-NMDA receptors in the rescue of weaver cerebellar granule neurons and sensitivity to ethanol of cerebellar AMPA receptors in oocytes

The cellular mechanism responsible for the death of cerebellar granule neurons in the weaver mutant mouse is still being intensely investigated. To determine if alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) glutamate receptors are involved in producing the weaver phenotype or are altered by the weaver gene, we used (1) reverse transcription and polymerase chain reaction (RT-PCR) to detect transcripts of glutamate receptors (GluR1-4) from wild-type and mutant cerebella; (2) immunocytochemistry to establish the types of glutamate receptors present in granule neurons cultured from normal and homozygous weaver postnatal day 5-6 (P5-6) cerebella; (3) 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), a blocker of glutamate (AMPA/Kainate/NMDA) receptors, and 6,7-dinitroquinoxaline-2,3-dione (NBQX), a blocker of AMPA and kainate receptors, to assess the number of neurons and the number of neurons with long neurites in cultures of homozygous weaver granule neurons; (4) two-electrode voltage clamp recordings to study AMPA glutamate receptor expression in Xenopus oocytes after injection of mRNA isolated from cerebella of normal and weaver P5-6, postnatal day 10 (P10) and postnatal day 23 (P23) mice; and (5) ethanol, which at low 1-10 mM concentrations had been shown previously to rescue homozygous weaver granule neurons in culture [Liesi et al., J. Neurosci. Res. 48 (1997) 571-579], to examine its effect on modulation of AMPA receptors expressed from mRNA. By RT-PCR, the mRNA coding for AMPA receptor subunits GluR1-4 were detected from +/+ and wv/wv cerebella, and by immunocytochemistry, GluR1, GluR2/3 and GluR4 were observed to be expressed in cultured +/+ and wv/wv granule cells. CNQX at 10 microM or NBQX at 10 microM significantly increased the number of surviving neurons and the number with long neurites as compared to wv/wv controls. In addition, CNQX was significantly more effective than NBQX. In oocytes injected with mRNA from P10 normal or weaver cerebella, the amplitudes of the responses to kainate were about equal. In contrast, the amplitudes of the kainate-activated currents in oocytes injected with weaver P23 mRNA were about twice as large as the currents observed in oocytes injected with mRNA from normal P23 cerebella, and both were larger than kainate-activated currents observed after injection of P10 normal and weaver mRNA. Kainate-activated AMPA receptor currents in oocytes injected with mRNA from P10 and P23 normal and homozygous weaver cerebella were inhibited by ethanol. There were no significant differences in the inhibition produced by ethanol on currents from P10 or P23 normal and wv/wv mRNA. Thus, P23 weaver cerebellar mRNA expressed more kainate-activated current in oocytes than P23 normal cerebellar mRNA; both normal and weaver cerebellar granule neurons express mRNA coding for functional AMPA receptors that are susceptible to ethanol inhibition.

Schedule of NMDA receptor subunit expression and functional channel formation in the course of in vitro-induced neurogenesis

NE-7C2 neuroectodermal cells derived from forebrain vesicles of p53-deficient mouse embryos (E9) produce neurons and astrocytes in vitro if induced by all-trans retinoic acid. The reproducible morphological stages of neurogenesis were correlated with the expression of various NMDA receptor subunits. RT-PCR studies revealed that GluRepsilon1 and GluRepsilon4 subunit mRNAs were transcribed by both non-induced and neuronally differentiated cells. GluRepsilon3 subunit mRNAs were not synthesized by NE-7C2 cells and increased numbers of messages from the GluRepsilon2 gene were detected only after neural network formation. The presence of the GluRzeta1 protein was detected throughout neural induction, whereas retinoic acid-induced neuron formation elevated the amount of exon 21 (C1)- and exon 22 (C2)-containing GluRzeta1 mRNAs and resulted in the appearance of exon 5 (N1)-containing transcripts. NMDA-elicited Ca(2+)-signals were detected only in cells displaying neuronal morphology, but preceding the appearance of synapsin-I immunoreactivity. Our findings demonstrated that, in spite of the presence of subunits necessary for channel formation, functional channels were formed by NE-7C2 cells no sooner than the time of neurite maturation. The data show that the cell line provides a suitable model to analyse the mechanisms involved in NMDA receptor gene expression before the appearance of synaptic communication.

Involvement of GIRK2 in postnatal development of the weaver cerebellum

We demonstrate that the homozygous weaver granule neurons cultured on a laminin substratum fail to express inwardly rectifying potassium currents, including a functional G-protein coupled inwardly rectifying potassium (GIRK)2 potassium channel. By contrast, both normal and weaver Purkinje cells express inwardly rectifying potassium currents, and normal granule cells exhibit inwardly rectifying potassium currents inducible with GTP-gamma-S. In protein extracts of the vermal postnatal day (P)5-9 weaver cerebellum, the GIRK2 protein could not be detected by Western analysis, although the GIRK2 protein was detectable in extracts of the normal vermis. Northern analysis indicated that during early postnatal cerebellar development, the GIRK2 mRNA is expressed at extremely low levels being detectable at P18-23 in the normal but not yet in the homozygous weaver cerebellum. Using reverse transcriptase-polymerase chain reaction (RT-PCR), the GIRK2 mRNA was detected in both normal and weaver cerebella, but quantitative PCR confirmed that the weaver cerebellum expressed the GIRK2 gene at significantly lower levels as compared to the normal cerebellum (P = 0.01, paired t-test). Sequencing indicated that the weaver GIRK2 channel gene had the point mutation proposed to be responsible for the weaver phenotype. Rescue of both survival and neurite outgrowth of the cultured vermal weaver granule neurons by verapamil (Liesi and Wright, 1996; Liesi et al., 1999) induced expression of immunocytochemically detectable levels of the GIRK2 protein. Sequencing revealed that the GIRK2 mRNA of the rescued weaver granule neurons remained the mutated variant of the GIRK2 channel gene. Our results indicate that expression of the mutated GIRK2 protein and/or mRNA in the weaver granule neurons may be an indicator of rescue rather than death of the weaver granule neurons. That the weaver granule neurons expressed no functional GIRK2 receptors during a time period of neuronal death and migration failure suggests that the point mutation in the H5 membrane spanning region of the GIRK2 gene may associate with, but not be responsible for the weaver phenotype.

Roles of NMDA receptor activity and nitric oxide production in brain development

The concept that neural activity is important for brain maturation has focused much research interest on the developmental role of the NMDA receptor, a key mediator of experience-dependent synaptic plasticity. However, a mechanism able to link spatial and temporal parameters of synaptic activity during development emerged as a necessary condition to explain how axons segregate into a common brain region and make specific synapses on neuronal sub-populations. To comply with this developmental constraint, it was proposed that nitric oxide (NO), or other substances having similar chemical and biological characteristics, could act as short-lived, activity-dependent spatial signals, able to stabilize active synapses by diffusing through a local volume of tissue. The present article addresses this issue, by reviewing the experimental evidence for a correlated role of the activity of the NMDA receptor and the production of NO in key steps of neural development. Evidence for such a functional coupling emerges not only concerning synaptogenesis and formation of neural maps, for which it was originally proposed, but also for some earlier phases of neurogenesis, such as neural cell proliferation and migration. Regarding synaptogenesis and neural map formation in some cases, there is so far no conclusive experimental evidence for a coupled functional role of NMDA receptor activation and NO production. Some technical problems related to the use of inhibitors of NO formation and of gene knockout animals are discussed. It is also suggested that other substances, known to act as spatial signals in adult synaptic plasticity, could have a role in developmental plasticity. Concerning the crucial developmental phase of neuronal survival or elimination through programmed cell death, the well-documented survival role related to NMDA receptor activation also starts to find evidence for a concomitant requirement of downstream NO production. On the basis of the reviewed literature, some of the major controversial issues are addressed and, in some cases, suggestions for possible future experiments are proposed.