BR A IN RE S EA RCH 1 1 10 ( 20 0 6 ) 4 6 –54
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / b r a i n r e s
Research Report
Embryonic expression of zebrafish AMPA receptor genes: Zygotic gria2α expression initiates at the midblastula transition Wei-Hsiang Lina,b , Chan-Hwa Wua , Yu-Chia Chena , Wei-Yuan Chowa,b,⁎ a
Institute of Molecular and Cellular Biology, National Tsing-Hua University, 101, Sec 2, Kang Fu Road, Hsinchu, Taiwan 30043, Republic of China b Department of Life Science, National Tsing-Hua University, 101, Sec 2, Kang Fu Road, Hsinchu, Taiwan 30043, Republic of China
A R T I C LE I N FO
AB S T R A C T
Article history:
The AMPA-preferring receptors (AMPARs) mediate rapid excitatory synaptic transmission in
Accepted 19 June 2006
the central nervous system of vertebrates. Expression profiles of 8 AMPAR genes were
Available online 2 August 2006
studied by RT-PCR analyses to elucidate the properties of AMPARs during early zebrafish development. Transcripts of all AMPAR genes are detected at the time of fertilization,
Keywords:
suggesting maternal transcriptions of zebrafish AMPAR genes. The amounts of gria1 and
AMPA receptor
gria2 transcripts are several-fold higher than that of gria3 and gria4 between 10 and 72 hpf
Gene expression
(hour postfertilization). The edited gria2α transcript decreases during gastrulation period,
RNA editing
suggesting that zygotic expression of gria2α begins around the time of midblastula
C-terminal splice form
transition. Relative to the amount of β-actin, the amounts of AMPAR transcripts increase
Midblastula transition
significantly after the completion of neurulation. The amounts of gria2 transcripts exceed
Zebrafish embryogenesis
the total amounts of the remaining AMPAR transcripts after 36 hpf, suggesting increases in the representation of low Ca2+ permeable AMPARs during neuronal maturation. Many but
Abbreviations:
not all of the known mammalian protein–protein interaction motifs are preserved in the
AMPAR, AMPA (α-amino-3-hydroxy-
C-terminal domains (CTD) of zebrafish AMPARs. Before 16 hpf, the embryos express
5-methyl-4-isoxazole propionic
predominantly the alternative splice forms encoding longer CTD. Representations of the
acid)-preferring receptor
short CTD splice forms of gria2 and gria4α increase after 24 hpf, when neurulation is nearly
CTD, C-terminal domain
completed.
hpf, hour postfertilization
© 2006 Elsevier B.V. All rights reserved.
NMDA, N-methyl-D-aspartate
1.
Introduction
A majority of the fast excitatory synaptic transmission in the vertebrate central nervous system is mediated by the AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid)preferring receptor (AMPAR), which is one of the three pharmacologically defined families of glutamate-gated ion
channels. Vertebrate AMPARs are heterogeneous, in that they are assembled from the products of four genes, gria1–gria4 (also known as GluR1–GluR4 and GluRA–GluRD), in either heteromeric or homomeric forms (Bredt and Nicoll, 2003; Hollmann and Heinemann, 1994). Due to a teleost-specific gene duplication event, gria1–gria4 are duplicated in the zebrafish (Chen et al., 2001). The Gria2 subunit plays a dominant role in
⁎ Corresponding author. Institute of Molecular and Cellular Biology, National Tsing-Hua University, 101, Sec 2, Kang Fu Road, Hsinchu, Taiwan 30043, Republic of China. Fax: +86 886 3 5742687. E-mail address:
[email protected] (W.-Y. Chow). 0006-8993/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2006.06.054
BR A IN RE S E A RCH 1 1 10 ( 20 0 6 ) 4 6 –5 4
controlling the current–voltage relationship and Ca2+ permeability of heteromeric AMPAR channels (Sommer et al., 1991). Immunoprecipitation studies demonstrate that Gria2-containing complexes, associated either with Gria1 or Gria3, are major AMPAR populations present in the adult hippocampus (Wenthold et al., 1996). Electrophysiological studies also show that a majority of mammalian neurons express Gria2containing heteromeric channels (Geiger et al., 1995). The complexity of vertebrate AMPARs is further augmented by alternative splicing and RNA editing of the primary transcripts (Hollmann and Heinemann, 1994). Alternative splicing of two mutually exclusive exons, termed flop and flip, changes the sequence immediately preceding the fourth membrane segment of Gria1–4 subunits (Hollmann and Heinemann, 1994). In addition, mammalian gria2 and gria4 are alternatively spliced to encode two types of C-terminal domains (CTD) varying in lengths and sequences (Gallo et al., 1992; Köhler et al., 1994). The CTD of AMPAR subunits contain sites for protein phosphorylations and protein–protein interactions (Bredt and Nicoll, 2003; Malinow and Malenka, 2002). Interactions between AMPARs and cellular proteins regulate the trafficking of receptors in and out of the synaptic membrane, playing crucial roles in changing synaptic efficacies in long-term potentiation and long-term depression (Bredt and Nicoll, 2003; Malinow and Malenka, 2002). RNA editing of the gria2 transcript converts a genomic glutamine (Q, CAG) codon to an arginine (R, CGG) codon at the Q/R site, which controls the divalent ion permeability of AMPARs. The gria2 mRNA is fully edited throughout mouse development (Higuchi et al., 1993). Alternative splicing and RNA editing also operate on the teleost AMPAR transcripts (Kung et al., 1996; Kung et al., 2001; Wu et al., 1996). The zebrafish gria2α encodes a Q codon and its transcript, similar to the mammalian gria2, is modified by Q/R RNA editing, whereas the paralogous gria2β possesses a chromosomally encoded R codon at the Q/R site (Kung et al., 2001). The participation of ionotropic glutamate receptors in excitatory synaptic transmission, formation of higher brain functions and excitotoxicity have been well established (Hollmann and Heinemann, 1994; McDonald and Johnston, 1990). Biochemical and northern blot analyses of the developing central nervous system show that the expression of AMPAR genes precedes synaptogenesis in mammalian embryos (Åkesson et al., 2000; Durand and Zukin, 1993). Pharmacological studies reveal that AMPAR activation enhances the migration of mouse gonadotrophinreleasing hormone neurons (Simonian and Herbison, 2001). Electrophysiological and pharmacological studies demonstrate that AMPARs are expressed in the cortical progenitor cells before the final mitotic division and the activities of AMPAR influence the cell cycle of cortical progenitors (LoTurco et al., 1995; Luk and Sadikot, 2004). AMPAR agonists and antagonists, respectively, promote and decrease the survival of very young mammalian neurons at early culture stage (Drian et al., 2001). All of these results suggest that AMPARs participate in the development of the central nervous system. AMPAR-mediated miniature excitatory postsynaptic currents have been detected in the reticulospinal neuron regions of the 1.2- to 1.5-day postfertilization zebrafish embryos after main axon tracts have
47
been established (Ali et al., 2000). However, the onset of AMPAR gene expression and the molecular constituents of embryonic AMPARs remain largely unknown. In this study, we studied the temporal expression profiles of AMPAR genes during early zebrafish development. In contrast to the general understanding that AMPAR gene expressions are specific to the central nervous system, AMPAR transcripts were readily detectable in oocytes, cleavage stage and blastula embryos before the establishment of neuronal fate. The initiation of zygotic gria2α expression coincides at the timing of midblastula transition (Kane and Kimmel, 1993). The expression patterns of the CTD splice variants of AMPAR transcripts and the extent of Q/R RNA editing of gria2α transcript were also studied to elucidate possible molecular constituents and, consequently, the properties of AMPARs during early zebrafish development.
2.
Results
2.1. AMPAR transcripts are present before neuronal induction The developmental stages of zebrafish embryos were defined by hour postfertilization (hpf) and confirmed by morphological characteristics (Kimmel et al., 1995). Early neurogenesis is grossly divided into three overlapping stages: neural induction (blastula period), regionalization and neurulation (gastrula and segmentation periods) and axonogenesis (Strähle and Blader, 1994). The paralogous AMPAR genes of teleost are designated as α and β followed the gene names (Chen et al., 2001). In this report, a pair of paralogous zebrafish AMPAR genes is collectively referred without the postfix. For an example, gria2 means gria2α and gria2β. A pair of degenerate primers known to amplify all the zebrafish AMPAR transcripts was employed to perform an RT-PCR survey of AMPAR transcripts during zebrafish development (Chen et al., 2001). AMPAR transcripts were detected in all embryogenesis stages, including the 1-cell (0 hpf) and cleavage (2 hpf, data not shown) embryos before neural competence of the ectoderm is established (Fig. 1). Detectable amounts of AMPAR transcripts were also present in the blastula (4 hpf) and early gastrula embryos (data not shown). Because AMPAR transcripts were also
Fig. 1 – An RT-PCR survey of AMPAR transcripts during zebrafish development. An equal amount of cDNA was amplified by degenerate primers designed to amplify all gria1–4 (AMPAR subunits). Amplification of β-actin was run in parallel to standardize the input cDNA. Due to an insertion between the first and second membrane spanning segments of teleost Gria1 subunits, the amplicons of gria1 are longer than those of the gria2–4 (Wu et al., 1996).
48
BR A IN RE S EA RCH 1 1 10 ( 20 0 6 ) 4 6 –54
detected in oocytes, some if not all of the transcripts detected in these early-staged embryos may result from maternal expressions (data not shown). Because the amplification efficiency of each AMPAR transcript by the degenerate primer pair varied, quantitative RT-PCR analyses were performed with gene-specific primers to accurately reflect the expression level of each AMPAR transcript. The amounts of AMPAR transcripts expressed in each stage were normalized to the amount of β-actin (relative amount). The relative amounts of AMPAR transcripts detected at 16 hpf, when regionalization and neurulation were in progresses, were chosen as references for statistic analyses. All 8 AMPAR transcripts are present at 1-cell stage (0 hpf), although gria2β, gria3β and gria4α are expressed at much lower levels (less than 6 × 10− 7 per β-actin molecule) than the other 5 AMPAR gene transcripts (greater than 2 × 10− 5 per β-actin molecule; Fig. 2). The total amount of gria1 and gria2 transcripts approximately equals to that of gria3 and gria4 in the one-cell embryos. The relative amounts of gria1 and gria2α transcripts vary slightly between 0 and 10 hpf (Fig. 2A); however, that of gria3α and gria4β decrease drastically (Fig. 2B). AMPARs expressed after 4 hpf are mainly gria1 and gria2 (Fig. 2, the scales of Figs. 2A and B are different). A trend of
gradual increase expression of gria2α occurs between 4 and 16 hpf; however, relative to the amount detected at 16 hpf, the increase of gria2α expression becomes significant after 30 hpf, when neurulation is essentially completed and axonogenesis is well under way. The relative amount of gria2β transcript remains very low (
