QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

This Article Summary Full Text Full Text (PDF) Image Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Keegan, B. R. Articles by Yelon, D. Search for Related Content PubMed PubMed Citation Articles by Keegan, B. R. Articles by Yelon, D.

The elongation factors Pandora/Spt6 and Foggy/Spt5 promote transcription in the zebrafish embryo

Brian R. Keegan^1,*, Jessica L. Feldman^1,*, Diana H. Lee¹, David S. Koos², Robert K. Ho², Didier Y. R. Stainier³ and Deborah Yelon^1,

¹ Developmental Genetics Program and Department of Cell Biology, Skirball Institute of Biomolecular Medicine, New York University School of Medicine, New York, NY, USA
² Department of Molecular Biology, Princeton University, Princeton, NJ, USA
³ Department of Biochemistry and Biophysics and Programs in Developmental Biology, Genetics, and Human Genetics, University of California, San Francisco, San Francisco, CA, USA
^* These authors contributed equally to the work

View larger version (91K): [in a new window]   Fig. 1. Morphological and molecular defects are similar among a class of zebrafish mutants. (A-E) Lateral views of live embryos at 36 hours post-fertilization (hpf), anterior towards the left. All embryos are depicted at the same magnification. Compared with wild-type embryos (A), pan (B), sk8 (C) and s30 (D) mutants exhibit reduced pigmentation, short tails, small ears and pericardial edema. s30;pan double mutants (E) exhibit a more extreme phenotype, including a shorter tail and neural cell death. (F-U) Whole-mount in situ hybridization indicates expression of gata4 (F-I), nkx2.5 (J-M) or cmlc2 (N-U); dorsal views of embryos, anterior towards the top. At 13 hpf (eight-somite stage), expression of gata4 and nkx2.5 are comparable in wild-type embryos (F,J), pan mutants (G,K) and s30 mutants (H,L); however, expression is reduced in s30;pan double mutants (I,M). At 16.5 hpf (15-somite stage), wild-type embryos (N) exhibit robust expression of cmlc2 in differentiating myocardiocytes, but pan mutants (O), s30 mutants (P) and s30;pan double mutants (Q) all lack cmlc2 expression. By 26 hpf, wild-type embryos (R) form a cmlc2-expressing heart tube, and pan mutants (S) generate a small and variable number of disorganized myocardiocytes (Yelon et al., 1999), but cmlc2 remains undetectable in s30 mutants (T) and s30;pan double mutants (U).

View larger version (28K): [in a new window]   Fig. 2. The panm313 mutation causes mis-splicing of spt6 mRNA and truncation of Spt6 protein. (A) spt6 cDNA from pan mutants at 30 hpf contains a 92 nucleotide insertion. PCR primers that flank an 808 bp fragment of wild-type spt6 cDNA amplify a 900 bp fragment from pan mutant cDNA. This larger fragment represents the major splice isoform of spt6 in pan mutants, although we have infrequently detected trace amounts of normally spliced cDNA that could represent a low level of maintained maternal mRNA or a low level of normal splicing of zygotic mRNA. (B) Comparison of wild-type and pan mutant spt6 cDNA. Labeled regions in the wild-type ORF are proposed to encode discrete protein domains (see D below). The 92 nucleotide insertion in pan mutant cDNA is indicated by a loop, and the location of the in-frame stop codon is marked with an asterisk. (C) Genomic sequence of exon-intron boundaries flanking the mutated intron. The insertion in pan mutant cDNA corresponds to a 92 bp intron. pan genomic DNA contains a point mutation (boxed) in the second nucleotide of this intron. This mutation prevents correct splicing at this junction in pan mutants. The in-frame stop codon is underlined and marked with an asterisk. (D) Predicted protein structure for wild-type Spt6 protein (1726 amino acids) and truncated pan mutant Spt6 protein (829 amino acids). The acidic region, S1 RNA binding domain and SH2 domain are indicated. The insertion of intronic sequence in pan mutant cDNA is predicted to result in the addition of nine mis-sense amino acids before reaching a premature in-frame stop codon. Therefore, the pan mutant Spt6 protein would lack both the S1 RNA binding domain and SH2 domain.

View larger version (31K): [in a new window]   Fig. 3. The s30 and sk8 mutations disrupt the foggy/spt5 locus. (A) The s30 mutation deletes spt5. The first and second lanes demonstrate that a 145 bp fragment of the 5' end of spt5 genomic DNA can not be amplified by PCR from s30 genomic DNA. The third and fourth lanes demonstrate a similar result for a 172 bp fragment of the 3' end of spt5 genomic DNA. Similar results were obtained for all regions of spt5 genomic DNA (data not shown). A control fragment (225 bp) can be amplified from all samples. (B) spt5 cDNA from sk8 mutants is missing 31 bp. PCR primers that flank a 310 bp fragment of wild-type spt5 cDNA amplify a 279 bp fragment from sk8 mutant cDNA. This smaller fragment represents the major splice isoform of spt5 in sk8 mutants at 24 hpf, although we have infrequently detected trace amounts of normally spliced cDNA that could represent a low level of maintained maternal mRNA or a low level of normally spliced mutant mRNA. (C) Comparison of wild-type and sk8 mutant spt5 cDNA. Labeled regions are proposed to encode discrete protein features (see E). The location of the missing nucleotides is indicated following nucleotide 520 in sk8 mutant cDNA. (D) Structure of spt5 genomic DNA indicates that the missing bases correspond to a 31 bp exon, flanked by a 88 bp intron and a 103 bp intron. In wild-type embryos, this region is spliced normally and the 31 bp exon is located between nucleotides 520 and 552 in spt5 cDNA. In sk8 mutants, incorrect splicing results in the omission of the exon and both introns from spt5 cDNA. sk8 genomic DNA contains a point mutation (boxed) in the first nucleotide of the intron following the 31 bp exon. (E) Predicted protein structure for wild-type Spt5 protein (1084 amino acids) and truncated sk8 mutant Spt5 protein (184 amino acids). The acidic region, KOW motifs, and hexapeptide repeats are indicated (Guo et al., 2000). The omission of an exon in sk8 mutant cDNA creates a frameshift that is predicted to result in the addition of 10 mis-sense amino acids after residue 174 before reaching a premature in-frame stop codon. Therefore, the sk8 mutant Spt5 protein would lack all KOW motifs and hexapeptide repeats.

View larger version (55K): [in a new window]   Fig. 4. spt6 and spt5 mRNA rescue the defects in pan and fogs30 mutants, respectively. Uninjected embryos (A,B,E,F,I,J,M,N) are compared with embryos that were injected with synthetic mRNA, either spt6 (C,D,G,H) or spt5 (K,L,O,P). All embryonic genotypes were confirmed following photography (see Materials and Methods). Genotypes are: wild-type embryos (A,C,E,G,I,K,M,O); pan mutants (B,D,F,H); and fogs30 mutants (J,L,N,P). (A-D,I-L) Lateral views of live embryos at 36 hpf, anterior towards the left. Injection of wild-type mRNA rescues pigmentation, tail length, ear formation and cardiac function in mutants, and does not affect wild-type siblings. Rescued embryos survive for 4-5 days post-fertilization. (E-H,M-P) Whole-mount in situ hybridization indicates expression of cmlc2 at 16.5-17.5 hpf; dorsal views of embryos, anterior towards the top. Injection of wild-type mRNA rescues the timely and robust expression of cmlc2 expression in mutants and does not affect expression in wild-type siblings.

View larger version (74K): [in a new window]   Fig. 5. spt6 and spt5 are expressed maternally and zygotically throughout the zebrafish embryo. Whole-mount in situ hybridization, comparing the expression patterns of spt6 (A-C) and spt5 (D-F). Wild-type embryos at 2.75 hpf (512-cell) (A) or 2.25 hpf (128-cell) (D) demonstrate that both spt6 (A) and spt5 (D) are expressed throughout the blastoderm before the initiation of zygotic transcription and are therefore maternally supplied. (A,D) Lateral views, with the animal pole at the top. Wild-type embryos at 16.5 hpf (B) or 10 hpf (E) demonstrate that spt6 (B) and spt5 (E) are expressed throughout the embryo after the initiation of zygotic transcription. Inset in E depicts a fogs30 mutant embryo at 10 hpf. spt5 mRNA is barely detectable in fogs30 mutants at this stage, in keeping with the deletion of spt5 in fogs30 mutant genomic DNA. (B,E) Lateral views, anterior towards the top. Wild-type embryos at 24 hpf (C) or 26 hpf (F) demonstrate that spt6 (C) and spt5 (F) are still broadly expressed, with highest levels in the embryonic brain. (C,F) Lateral views, anterior towards the left.

View larger version (50K): [in a new window]   Fig. 6. Spt6 and Spt5 are required for an efficient heat shock response. (A-L) Lateral views of live embryos at 20 hpf (22-somite stage), anterior towards the left. Green fluorescence indicates production of GFP after a 1 hour heat shock and a 0.5 hour recovery period. Autofluorescence (yellow-green) is also visible in embryonic yolks. All photographs were taken with the same exposure time, and embryonic genotypes were confirmed after photography (see Materials and Methods). Genotypes are wild-type embryos (A,E,I); pan mutants (B,F,J); fogs30 mutants (C,G,K); and fogs30;pan double mutants (D,H,L). All embryos are heterozygous for the hsp70-egfp transgene (Halloran et al., 2000). Embryos maintained at normal temperature do not produce detectable GFP (A-D). After heat shock, pan and fog mutants produce less GFP than wild-type embryos, and double mutants do not produce any detectable GFP (E-H, with higher magnification views of the tail in I-L). Similar results were obtained when heat shock was administered from 15-16 hpf, before mutant phenotypes are morphologically apparent (data not shown). (M) Spt6 and Spt5 are required for efficient induction of hsp70 expression during a 1 hour heat shock. Graph of induction of endogenous hsp70 expression, detected by real-time RT-PCR. (See Materials and Methods for details of technique and data processing.) Wild-type embryos (black) induce significantly higher levels of hsp70 mRNA with more rapid kinetics than pan mutants (green) or fogs30 mutants (blue). hsp70 induction is further inhibited in fogs30;pan double mutants (red). Each data point represents the average degree of induction of hsp70 expression at a particular time point, relative to the low, but detectable, levels of hsp70 expression at time zero. Standard deviation from the mean is indicated by error bars. To account for small variances in RNA extraction and cDNA synthesis, levels of hsp70 expression were normalized relative to levels of stable ß-actin expression (see Materials and Methods).