The Coevolution of Insect Muscle TpnT and TpnI Gene Isoforms
http://www.100md.com
分子生物学进展 2005年第11期
Departamento de Bioquímica, Instituto de Investigaciones Biomédicas "Alberto Sols" UAM-CSIC, Facultad de Medicina, Universidad Autónoma de Madrid, C/Arzobispo Morcillo 4, 28029 Madrid, Spain
E-mail: roberto.marco@uam.es.
Abstract
In bilaterians, the main regulator of muscle contraction is the troponin (Tpn) complex, comprising three closely interacting subunits (C, T, and I). To understand how evolutionary forces drive molecular change in protein complexes, we have compared the gene structures and expression patterns of Tpn genes in insects. In this class, while TpnC is encoded by multiple genes, TpnT and TpnI are encoded by single genes. Their isoform expression pattern is highly conserved within the Drosophilidae, and single orthologous genes were identified in the sequenced genomes of Drosophila pseudoobscura, Anopheles gambiae, and Apis mellifera. Apis expression patterns also support the equivalence of their exon organization throughout holometabolous insects. All TpnT genes include a previously unidentified indirect flight muscle (IFM)–specific exon (10A) that has evolved an expression pattern similar to that of exon 9 in TpnI. Thus, expression patterns, sequence evolution trends, and structural data indicate that Tpn genes and their isoforms have coevolved, building species- and muscle-specific troponin complexes. Furthermore, a clear case can be made for independent evolution of the IFM-specific isoforms containing alanine/proline-rich sequences. Dipteran genomes contain one tropomyosin gene that encodes one or two high–molecular weight isoforms (TmH) incorporating APPAEGA-rich sequences, specifically expressed in IFM. Corresponding exons do not exist in the Apis tropomyosin gene, but equivalent sequences occur in a high–molecular weight Apis IFM-specific TpnI isoform (TnH). Overall, our approach to comparatively analyze supramolecular complexes reveals coevolutionary trends not only in gene families but in isoforms generated by alternative splicing.
Key Words: troponin ? tropomyosin ? coevolution ? Insecta ? Drosophila ? Anopheles ? Apis ? phylogenetics
Introduction
One of the most intriguing aspects of molecular evolution is the multiplicity of isoforms produced by genes (Roberts and Smith 2002), the functional significance of which still largely escapes our understanding. Gene diversification has clearly relied on two mechanisms, either entire genes (or even chromosome fragments) are duplicated generating a gene family, or specific exons are duplicated along with the capacity for generating alternatively spliced transcripts. Many studies have compared the evolutionary advantages of each process (Lynch and Conery 2000; Ohta 2000; Lynch 2002). Exon-duplication events seem to be preferred if the result increases the variability of only small domains in a protein (Abi Rached, McDermott, and Pontarotti 1999).
Particularly relevant are the cases where groups of proteins interact to form high supramolecular complex structures, especially if the physiological effects of isoforms are well characterized. Our group has been studying these gene-evolution processes in troponin, a thin-filament complex, by focusing on one of the most diversified branches of life, the insects. The troponin complex comprises three subunits, namely troponin C (TpnC), troponin I (TpnI), and troponin T (TpnT), and mediates the thin-filament response to calcium when striated muscular contraction is initiated. In mammalian systems, extensive studies (Filatov et al. 1999; Gordon, Homsher, and Regnier 2000; Gordon, Regnier, and Homsher 2001) have addressed the structure-function relationships between the Tpn complex subunits. As a consequence of Ca2+ binding to TpnC, this polypeptide undergoes a conformational change, altering its interaction with TpnI, releasing the inhibitory TpnI-actin binding, and allowing movements of the troponin-trpomyosin complex on the F-actin that increases myosin binding to actin and thus promoting contraction.
The gene numbers for each troponin constituent vary among the metazoa. While only one copy each of the TpnT and TpnI genes occur in insects, up to three copies are found in vertebrates. There are only two TpnC genes in mammals (encoding cardiac and striated isoforms), but five to six troponin C genes have been found in holometabolous insects (Herranz, Mateos, and Marco 2005). Nevertheless, each troponin subunit exists in multiple isoforms. The single Drosophila troponin I gene has a total of 13 exons, and up to 10 different TpnI isoforms are produced by alternative splicing processes of some exons (Beall and Fyrberg 1991; Barbas et al. 1993). Variability regions are localized at specific points across the protein encoded by the alternative exons 3 and 9 and by four mutually exclusive exon 6's. The TpnT gene in both Drosophila melanogaster and Drosophila virilis was described as containing a set of 11 exons and up to four alternatively splicing isoforms, generated by the inclusion or exclusion of exons 3, 4, and 5 (Benoist et al. 1998). No splicing variants affecting the Ct half of the protein were described.
Muscle structure in protostomes is much more diverse than in deuterostomes. For example, holometabolous insects have two different musculatures, the larval one formed during embryogenesis, and the adult one formed during the pupal metamorphosis, when the muscles responsible for flight develop in the adult thorax. These muscles are the indirect flight muscles (IFMs), responsible for wing beating, and the tergal depressor of the trochanter (TDT) muscle, responsible for the jump at take-off of the fly. Stretch activation is an important property of IFM in many insects. Its direct consequence is that their oscillatory contractions are asynchronous, meaning that the nervous impulses received by the fibers are more irregular and less frequent than the contractions themselves. Specific protein variants are expressed in these muscles, sometimes encoded by different genes, as for example actin (Lovato et al. 2001) or troponin C (Qiu et al. 2003; Herranz et al. 2004), but more usually as the result of differential splicing of transcripts from a single gene. In addition, other muscle-specific components, such as arthrin and troponin H, are thin-filament proteins exclusively expressed in the IFMs. Arthrin is an ubiquitinated actin independently acquired in several insect orders (Schmitz et al. 2003). Troponin H (Bullard et al. 1988), a high–molecular weight component of Lethocerus flight muscles, was identified as similar to the heavy tropomyosin isoforms expressed in Drosophila IFM (Karlik and Fyrberg 1986; Hanke and Storti 1988). Troponin H and heavy tropomyosin (TmH; Mateos et al., unpublished data) contain alanine/proline-rich extensions in the Ct of the proteins.
In the present work, we describe how TpnI and TpnT isoforms have coevolved in insects. Our purpose in undertaking a systematic study of the insect troponin genes is to understand the coevolutionary mechanisms that led to functional divergence of isoform variation generated by the differential expression of the troponin complex genes. The specific repertoire of troponin isoforms expressed in distinct muscle types or stages of development has been established. In particular, we have analyzed the TpnI and TpnT genes of four Drosophilidae species (melanogaster, subobscura, pseudoobscura, and virilis), the evolution of which diverged during the last 60 Myr, as well as in Anopheles gambiae and Apis mellifera that diverged from the Drosophilidae around 250 and 300 MYA, respectively (Powell 1997; Dudley 2000; Gaunt and Miles 2002).
With these objectives, our studies have allowed us to identify previously undescribed features of the troponins T and I in insects. We have identified a new mutually excluding exon (10A) for the troponin T gene that produces an IFM/TDT differentially expressed isoform, adding a new variability region near the 3' end of the gene. We have refined the information on the TpnI expression patterns identifying TpnI exon 9 as more specific than originally described, and as the functional partner of the TpnT exon 10A isoform in the IFM. In contrast to the situation in Diptera, in Apis the alanine/proline-rich extensions do not occur attached to the Tm gene but are encoded in the TpnI gene, as three APPAEGA-rich 3' exons, H1, H2, and H3.
Materials and Methods
Insect Stocks
The Oregon R strain was used as wild-type source of D. melanogaster material. Drosophila subobscura was obtained from Rosa de Frutos (Valencia, Spain) and D. virilis was from Manuel Calleja (Madrid, Spain). These species diverged from D. melanogaster around 12 and 50 MYA (Powell 1997; Gaunt and Miles 2002). Apis mellifera samples were obtained in the Summer 2003 from Industrias Alonso located in El Vellon (Madrid). We have also included the data obtained from D. pseudoobscura (Human Genome Sequencing Project at Baylor College of Medicine Blast server) and the mosquito A. gambiae (Holt et al. 2002) genomes.
Nucleic Acid Extraction
Tissues from D. subobscura and D. virilis were obtained at different developmental stages as well as adult body parts (head, thorax, and abdomen). Dissections were done in cold acetone at –70°C. RNA extractions were made using TRIZOL Reagent (Invitrogen, Paisley, UK). The same procedure was carried out with A. mellifera (two different larval and two pupal stages, as well as recently emerged and mature honeybees, that were dissected to separate heads, thoraces, and abdomens). Specific muscles, IFM or TDT, were recovered from animals pretreated for at least 1 week in dehydrating acetone solution at –20°C, which facilitates the fiber isolation. Genomic DNA was extracted using a Tris-HCl 10 mM pH 7 homogenizing solution containing 0.5% sodium dodecyl sulfate and NaCl 60 mM, followed by a standard phenol-based purification (Sambrook, Fritsch, and Maniatis 1989).
Reverse Transcriptase–Polymerase Chain Reaction
We used total RNA (2 μg) and oligo-dT (1 μg) in order to normalize the amount of cDNA used as template in the polymerase chain reactions (PCRs). Specific probes were designed for each gene amplification, based on known sequences from D. melanogaster or A. mellifera. PCRs were carried out for 30–35 amplification cycles (94°C/30 s, 55–60°C/45 s, and 72°C/45 s) with thermostable polymerase (DyNAzime, Fynnzymes) in a Gene Amp 2700 System thermocycler of PE Applied Biosystems (Foster City, Calif.). For genomic fragment amplifications, hybridization and elongation times were extended up to 1 min and TpnI PCR experiments were performed in the presence of 2.5% dimethyl sulfoxide.
Rapid Amplification cDNA Extension
We have determined the 3' untranslated region (UTR) of the different TpnI transcripts in A. mellifera by using the Ambion FirstChoice RLM-RACE Kit, with a TpnI exon 8 probe plus the 3' outer and inner kit probes.
Cloning and Sequencing
Standard commercial protocols were followed for DNA extraction (Concert Rapid Gel Extraction, GIBCO), ligations (pGEM-T, Promega, Madison, Wisc.), Escherichia coli transformations (DH5 strain), and plasmid purification (Wizard Plus SV minipreps, Promega) sometimes followed by a phenol-based concentration step (Sambrook, Fritsch, and Maniatis 1989). Standard probes included in pGEM-T easy plasmid (SP6 and T7) were used for sequencing reactions in an automatic sequencer. All the detected reverse transcriptase (RT)–PCR products were sequenced in this way using internal oligonucleotides when necessary.
Bioinformatics Tools
Sequences were obtained and preliminarily compared by National Center for Biotechnology Information Blast (Altschul et al. 1997). Sequence information was processed using GeneJockey II software. Phylogenetic trees were obtained from sequence comparisons and alignments based on ClustalW with BIOEDIT (Hall 1999) software using "Fitch-Margoliash and least-squares distance methods" algorithm or remotely with Phylodendrom version 0.8d, beta 1999, and ClustalW version 1.75 (neighbor-joining method with 1,000 bootstraps [Thompson, Higgins, and Gibson 1994]). Tree output was rendered with TreeView version 1.6.6 (Page 1996). Branches with bootstrap values below 70% have been collapsed. The 2D structure predictions were obtained from the ExPASy web server using the PSIPRED method (Jones 1999; McGuffin, Bryson, and Jones 2000).
Accession Numbers
Sequences obtained or annotated by us have been submitted to the corresponding section of the GenBank with the following accession numbers: D. melanogaster and D. virilis TpnT gene, AY439172 and AY439178–9; TpnT and TpnI Drosophilidae transcripts, AY439173–7, AY439180–4, and BK001654–5; and A. gambiae and A. mellifera TpnT and TpnI genes, BK005279–82. This information has also been prepared in a Supplementary Material Table A (Supplementary material online) that also includes the accession numbers of the rest of the sequences used in this work. All gene annotations contain a detailed description of all the identified transcripts including their sequences and expression patterns.
Results
Conservation of Orthologous TpnI and TpnT Gene Structures in Insects
We have used the previously reported D. melanogaster TpnT and TpnI gene sequences (Barbas et al. 1993; Benoist et al. 1998) to find putative orthologues in the Anopheles and Apis genomes. The structural organization of these genes has been preserved to a high degree in insects (fig. 1). In general, we observe an increase in gene size, mainly due to intron length expansion from the smallest genome—178 Mb in D. melanogaster (Adams et al. 2000)—to those of Anopheles and Apis that are almost double in size (Holt et al. 2002). Chromosomal location is not a broadly preserved feature in insects, neither for TpnC genes (Herranz, Mateos, and Marco 2005) nor for those of TpnI or TpnT. The TpnT and TpnI genes are both located on the D. melanogaster X chromosome, but the Anopheles TpnT gene is located cytologically on chromosome arm 2R, while TpnI remains to be localized (Mongin et al. 2004).
FIG. 1.— Troponin T (top) and troponin I (bottom) gene structures in holometabolous insects. Genes are labeled with the species binomial two-letter abbreviation. Noncoding (white), constitutive (gray), alternative (black) or mutually exclusive (diagonal stripes), and the PAANGKA- or APPAEGA-containing (horizontal stripes) alternative exons are indicated. Light shaded exon 6's in TpnI have not yet been detected in RT-PCR experiments. Exon numbers in Anopheles and Apis genes have been assigned by homology to those of their Drosophila counterparts. Genomic DNA is represented by a continuous line that is therefore missing in gene sequences derived from cDNA data. Gene organization is well preserved overall, and gene size correlates with the organism genome size. The main difference affects alanine/proline-rich sequences including the TpnI exon 3 encoding the PAANGKA sequences in the Drosophilidae, but not in Anopheles or Apis. Several Apis TpnI gene exons encode an APPAEGA-containing sequence in their 3' region.
We have cloned and sequenced the transcripts of these genes in the drosophilid species and Apis, paying special attention to the alternatively spliced exons. In the case of troponin T, the differentially spliced Nt exons are present in all the analyzed species, although in Apis the TpnT gene shows an additional alternatively spliced exon, named 5'. The Apis equivalent to the dipteran exon 6 is divided into five constitutive exons (a, b, c, d, and e). Sequencing of transcripts identified a new variable exon 10, that had been overlooked in previous studies of D. melanogaster and D. virilis sequenced TpnT transcripts (Benoist et al. 1998). Comparison of the TpnT genomic sequences in insects showed that two different exons, named 10A and 10B, are present in all holometabolous species. Both are 79-nt long, encoding 26 aa. Another difference in the TpnT gene structure is found in exon 11, which encodes a long polyglutamic tail with variable size in all studied protostome species (Benoist et al. 1998).
The constitutive exons and exon 9 of TpnI gene are present in all analyzed insects (fig. 1), but exon 3 is not, to the extent that the Drosophila orthologous exons 2 and 4 are fused in a unique exon in Anopheles, but not in Apis where a smaller exon 3 is found the main variability of Troponin I arises from mutually exclusive exon 6's. In the case of Anopheles, we detected three exon 6 variants by sequence homology, while four were detected in Apis. Interestingly, additional exons, that we have designated H1, H2, and H3, were found downstream of exon 10 in the Apis TpnI gene, encoding the APPAEGA-repetitive sequences (see below).
Sequence Variations of the Alternatively Spliced Exons in TpnT and TpnI Genes
Amino acid sequences from TpnT and TpnI constitutive exons are conserved almost perfectly in insects. As shown in figure 2A, even the sequences of the alternatively spliced exons (3, 4, and 5) from TpnT are well conserved among the Drosophilidae. Exons 3 and 4 are also conserved in A. gambiae and A. mellifera while a higher degree of variation is found in exon 5, reflecting the evolutionary distance between these insects. In this region, the Apis TpnT gene has an additional alternatively spliced exon. Exons 3 and 4 are clearly conserved, but exons 5, 5', and the constitutive exon 6a have very different sequences, being more similar to sequences in the orthopteran Periplaneta (Wolf 1999) and the odonate Libellula TpnT exons (Fitzhugh and Marden 1997; Marden et al. 1999, 2001) despite the much larger evolutionary distance of these species from holometabolous insects.
FIG. 2.— Alternatively spliced exons in TpnT and TpnI genes. (A) Alignment of the 5' region of the troponin T transcripts in insects. Amino acid conservation (?), silent variation () (cDNA), equivalent variation (), and nonequivalent variation () in relation to the Drosophila melanogaster sequence are shown. These sequences, labeled with the two-letter species abbreviations and the corresponding exon number (E10A Dm means exon 10A in D. melanogaster for instance), are conserved in the four drosophilid species (only silent variation appears) and also in the Anopheles and Apis exons except exon 5. The Libellula (AF133521 [GenBank] ) and Periplaneta (AF133520 [GenBank] ) TpnT sequences (obtained from GenBank) retain the same properties although the alternative exon sequences show higher levels of variation. (B) Alignment and phylogenetic tree of the alternative TpnT exon 10 sequences in insects. Branches with bootstrap values below 70% have been collapsed in this cDNA-based tree. Most changes affect the number of phosphorylable residues, indicated at the right of the alignment, with the exception of the conserved final threonine marked with an asterisk. (C) Alignment and phylogenetic tree of the alternative TpnI exon 9 and 10 sequences in insects. The Haemaphysalis longicornis TpnI sequence (AB051079 [GenBank] ), in which exon 9 has not been described, was used as out-group.
The sequences of the mutually exclusive TpnT exons 10A and 10B (fig. 2B) are also well preserved among Drosophilidae. Larger differences are found in Anopheles and Apis. Nonequivalent variations in exon 10A are equally distributed throughout the exon, but less so in exon 10B. Interestingly, a higher number of threonine residues appear in insect exon 10A than in exon 10B, but only the last threonine of exon 10A is conserved in all the insects analyzed. Drosophila TpnT isoforms are readily phosphorylated in vivo (Domingo et al. 1998), precisely in the exon 10A–encoded sequence (Nongthomba and Sparrow, personal communication).
With regard to the TpnI genes, constitutive exons are almost strictly conserved and the sequence of alternative exons 9 and 10 (fig. 2C) reflect a situation similar to that of TpnT exons 10A and 10B. Even though we know that the alternatively spliced sequences are short, phylogenetic trees based on the comparison among the sequences encoded by these exons can be constructed. Although this information is not sufficient to establish the evolutionary origin of these exons, it can be used to detect how they have been changing in different insect groups when the data of all exons are analyzed together. In both Tpn genes, one of the duplicated exons (TpnT exon 10A and TpnI exon 9) appears to be less conserved than the other (TpnT exon 10B and TpnI exon 10), showing similar evolutionary patterns in each of the three insect groups analyzed. It is therefore likely that the duplication of these TpnT and TpnI alternative exons may have occurred at the origin of the holometabolous-type of development. In accordance with this idea, only a single TpnT exon 10 with intermediate sequence features has been found in the hemimetabolous Periplaneta americana and Libellula pulchella. Furthermore, only a single exon 9/10 has been found in the arachnid Haemaphysalis longicornis TpnI gene (You et al. 2001). Interestingly, TpnI exon 9 in the Drosophilidae does not contain a complete stop codon, which is formed by the splicing of its last nucleotide with the two first bases of exon 10 (Barbas et al. 1993). This feature is not observed in Anopheles or Apis, where stop codons and 3' UTRs are found both in exons 9 and 10.
The TpnI mutually exclusive exon 6 sequences have also been analyzed. In a consensus phylogenetic tree based on exon 6 nucleotide sequences (fig. 3A), the low bootstrap values at the base of the insect tree suggest an ancient origin for these duplication events, but the tree supports a clear differentiation of exon 6a's from exon 6b's. The 6b exons are variable enough to be phylogenetically discriminative. The 6a exons are so different among themselves that the Apis paralogous exons 6a1 and 6a2 are as related to each other as to their putative drosophilid orthologous exons (see the polytomy in the 6a exon branch). The presence of a single exon 6a in Anopheles and the loss of the exon 6a2 canonical donor-splicing signal in the Drosophilidae are also noteworthy. In a sequence alignment (fig. 3B), it can be seen that each Anopheles and Apis exon 6 is as closely related to each other as it is to those from the Drosophilidae. Finally, the Drosophilidae TpnI exon 3 (fig. 4A) are essentially alanine/proline-rich sequences. All show the PAANGKA motif which appears repeated several times at different positions in the D. melanogaster, D. subobscura, and D. virilis extensions. This domain is absent in Anopheles and in Apis TpnIs, where a very short constitutive exon 3 is found.
FIG. 3.— Phylogenetic tree and alignment of the differentially spliced TpnI exon 6's. (A) Branches with bootstrap values below 70% have been collapsed in this cDNA-based tree. Exons 6b1 and 6b2 are clearly separated, but an exon 6a phylogeny cannot be extracted from the tree. (B) Alignment of the translated exon 6 sequences (34 aa translated from the last nucleotide of exon 5 plus the exon 6 nucleotides), flanked by the splicing sequences when available. The sequences are designated by the two-letter species abbreviation and the corresponding exon number (Dm E6b1 for instance). The 6a exons are the more variable ones, having even lost the canonical splicing donor sequence (bold and italics letters in the figure) in the drosophilid 6a2 exons.
FIG. 4.— Alignment of the IFM-specific troponin I and tropomyosin alanine/proline-rich extensions. (A) Although the TpnI exon 3 sequences are not perfectly conserved even in the Drosophilidae, the heptad PAANGKA (shaded box) is conserved. It is repeated in some species surrounded by an alanine/proline-rich region. This sequence is absent in Anopheles and Apis because their TpnI genes lack an exon 3 with these properties. (B) The TpnH extension is encoded by a single Tm1 gene exon in Anopheles, while in the Apis TpnI gene it involves up to three gene 3' exons (separated in the three rows of the alignment), all of them encoding an alanine/proline-rich sequence with the APPAEGA motif (shaded box). An equivalent sequence has been found in the Lethocerus TpnI gene (AJ621044 [GenBank] ) and is included as the final sequence in the alignment.
Three new exons have been located downstream from exon 10 in the Apis TpnI gene. These exons can be spliced together to produce an Apis-specific TpnI isoform containing a Ct extension encoding a repetitive APPAEGA sequence, through a new GT-splicing donor site in exon 10, located 25 bp before the stop codon. This extension is clearly similar to the alanine/proline-rich extension of the large–molecular weight IFM-specific tropomyosin in the Diptera, both in sequence and protein length (fig. 4B). Interestingly, the Apis tropomyosin Tm1 gene lacks this type of sequence (Mateos et al., unpublished results). A similar repetitive alanine/proline extension is found at the 3' end of a TpnI gene transcript from the hemipteran Lethocerus (Qiu et al. 2003 and sequence AJ621044).
Expression Pattern of TpnI and TpnT Genes Are Conserved in Drosophilidae
The expression profiles of TpnI and TpnT isoforms in D. subobscura and D. virilis were detected using RT-PCR techniques (unpublished data). All the transcripts previously identified in D. melanogaster (Barbas et al. 1993; Benoist et al. 1998) were detected in the second instar larvae and late pupae stages of these drosophilids. Combinations of the troponin T alternative exons 3, 4, and 5 appeared in larvae or in adult abdominal muscle transcripts in patterns characteristic for each muscle. The major transcripts of the adult thoracic muscles do not contain any of these alternative exons. Troponin I exon 9 inclusion is found exclusively in transcripts of the adult musculature, being accompanied by exon 3 in the major thoracic transcript. The four exon 6 types were detected at varying levels in all the tissues studied, those containing exon 6b1 being the more highly expressed in IFM (Barbas et al. 1993).
The discovery of the new TpnT exon 10A led to a study of its expression pattern. A single probe for constitutive exon 6 and four specific probes for exons 10A or 10B (fig. 5A) were used for RT-PCR. In embryos, only the 10B exon-containing transcript was detected, while in whole adult RNAs, transcripts containing either 10A or 10B appear.
FIG. 5.— Differences in the expression profile of TpnT and TpnI transcripts in the Drosophila melanogaster musculature. Agarose 1.2% gel separations of RT-PCRs. Each transcript detected by PCR as a band is marked with an arrow labeled with the name of the amplified gene and its exon composition. (A) Expression pattern of mutually exclusive exon 10's detected in adult or embryonic RNA extractions with four exon 10A- or 10B-specific primers (10A1, 10A2, 10B1, and 10B2). (B) Patterns of expression in adult body parts (head, IFM, TDT, and abdomen). Using probes for TpnT exon 6 and 10A (10A2 probe) or 10B (10B1 probe), the exon 10 expression pattern is shown. Exon 10B is expressed in all muscles in the adult except in the IFM. Exon 10A is the only one expressed in the IFM, but it is also expressed in the TDT muscles. (C) Probes for the 3' region of the TpnI gene were used in RT-PCRs (exons 8–9 and 8–10). Exon 9 uses the first two residues of exon 10 to generate a stop codon, so the exon 10 probe detects it as a lower band, the transcripts lacking exon 9, and as a higher band, the transcripts containing exon 9. So exon 9 is expressed mainly in IFM and only weakly in TDT muscles. (D) Using exon 2 and exon 9 probes for TpnI, it can be seen that exon 3 only appears heavily expressed in thoracic muscles including always exon 9. RT-negative control lanes are not shown.
Detection of exon 10A was particularly strong in the thorax, suggesting a specific role of the sequence encoded by this exon in the flight-related musculature (data not shown). Exon 10A- and 10B-specific probes detected, by RT-PCR, the classic 10B exon in the head, abdomen, and dissected TDT fibers (fig. 5B). Exon 10A expression was found almost exclusively in the RT-PCR product from RNA extracted from IFM and TDT fibers although some residual expression was also detected in abdomens. Consequently, IFM TpnT contains sequences encoded only by exon 10A, while TDT muscles contain a mixture of transcripts with exon 10A or 10B. The same thoracic enrichment of exon 10A transcripts was obtained in D. subobscura and D. virilis.
TpnI alternatively spliced exons were detected using several probe combinations. Exon 9 is present in TDT and IFM transcripts (fig. 5C), but while in IFMs it is the only one expressed, in TDT it is almost completely replaced by the exon 10–containing transcripts. Using probes for the complete coding region (exon 2 to exon 9/10) in the adult body parts showed that the alternatively spliced exon 3 is expressed in TDT/IFM, but is only present in transcripts that also incorporate exon 9 (fig. 5D).
Expression Profile of the TpnT and TpnI Genes in A. mellifera
The same procedure has been carried out with the honeybee to discover if the same patterns of isoform abundance occur. The expression profile of the TpnT transcripts during A. mellifera development (fig. 6A) shows that larval muscles contain a mixture of transcripts with alternative exons 3, 4, and 5' or exons 3 and 4. The IFMs contain an isoform encoded by a transcript lacking all the 5' region alternatively spliced exons. Other adult muscles contain a mixture of transcripts containing the exons 3, 4, 5, and 5' or 3, 4, and 5. In relation to the 3' half of the genes, exon 10B–containing transcripts were found in all muscles except in IFM, where it is only marginally detected (dorsoventral indirect flight muscle) or not at all (dorsolateral indirect flight muscle). Exon 10A–containing transcripts are adult thorax-specific, similar to the drosophilid results (see above).
FIG. 6.— Expression profile of TpnI and TpnT transcripts in Apis mellifera. Each transcript detected by PCR as a band is marked with an arrow labeled with the name of the amplified gene and its exon composition. Adjacent empty lanes to each PCR lane are negative controls (mRNA without reverse transcription). (A) Troponin T expression profile during development (larval L1 and L2 and pupal P1 and P2 stages) and in the thorax muscles (dorsoventral indirect flight muscles [DV-IFM], dorsolateral indirect flight muscles [DL-IFM], legs, and other thoracic muscles) of Apis. Apis TpnT shows a similar expression pattern to that described in Drosophila. (B) The TpnI expression pattern in the same sample is shown. The alternative exons, 9 and 10, both containing a proper stop codon and terminator signals, are expressed in different levels in all muscles and stages. Using probes from exon 4 and exon H1 of the Apis TpnI gene, we have detected an IFM-specific expression pattern for this TpnH isoform in hymenopterans. The four exon 6's identified are expressed in the TpnI transcripts (detected by inner PCR using internal exon 6 type-specific probes in a second PCR using the P2 fraction band as substrate DNA), except in the IFM-specific transcript where only 6b exons are detected. (C) The 3' rapid amplification cDNA extension (RACE) experiments were performed with thorax, IFM, and larval samples to detect the transcription stops in the Apis TpnI gene. Transcripts representing eight different stop sites have been proportionally represented on the left, with their size indicated and different arrows styles (depending on the transcripts last exon, dotted line for exon 9, continuous line for exon 10, and broken line for exon H's) signaling the corresponding band in the RACE electrophoresis gels. Three of them (one transcript for each group) are the major ones used in the thoracic muscles (their lengths are shown inside circles), but the transcript containing the TpnH exons is absent in larval muscles. RACE-negative control lanes are shown (– lanes).
The TpnI expression pattern detected using RT-PCR is shown in figure 6B. Probes for the region between exons 4 and 9 or exons 4 and 10 revealed that transcripts containing exon 9 or exon 10 are detected in the different developmental stages and in all muscles of the adult Apis musculature. Despite inherent limitations to these PCR experiments, exon 9–containing transcripts seem to appear later than those containing exon 10. Using probes for each specific exon 6 and exon 8 in an inner PCR of the previous amplified products, the four exon 6's in both exon 9– and exon 10–containing transcripts were detected. Although quantitative PCR was not carried out, the results indicate that maximum expression levels of exons 6a and 10 occur during larval stages, while those of 6b and 9 exons occur in adults.
Finally, RT-PCR using probes designed to amplify the region between exon 4 and exon H1 of the Apis TpnI gene showed, as expected, that a heavy Apis TpnI (troponin H) is expressed exclusively in the IFM (fig. 6B). This band was cloned and tested for its exon 6 content, using an exon 8 probe in combination with a specific probe for each one of the four exon 6. Sequencing of 14 clones confirmed that 7 clones contained exon 6b1 and the other 7 clones contained exon 6b2. In addition, these TpnH transcripts include a shorter spliced version of exon 10 that could not be detected in earlier RT-PCR assays using the exon 10 probe. In order to locate the 3' ends of the TpnI transcripts, we have performed rapid amplification cDNA extension experiments with three representative Apis RNA fractions (fig. 6C). We found up to eight different transcript ends for the TpnI gene, but only three major ones occur in mRNA from the thoracic muscles and IFM samples. Two of them are 3' to exon 9 and exon 10, produce the standard TpnI transcripts, and are also used in the larval muscles. The other end incorporates the initial part of exon 10 but extends into exons H1, H2, and H3, producing a TpnH isoform with a long alanine/proline-rich extension. We have also detected a transcript in which exon H1 does not use its acceptor-splicing site and this would produce a TnH isoform with a shorter alanine/proline-rich extension.
Discussion
It is generally accepted that muscle tissue arose early on in the evolution of bilateria. Muscle structure and its mechanism of function have been retained relatively unchanged since that time, although individual muscle tissues are fine-tuned to reflect their physiological functions. Because the myofibril comprises a series of supramolecular complexes, studying the evolution of an individual protein must be carried out in the context of the complex in which it is a part. We have initiated this analysis by focusing on one of the better-characterized substructures, the thin filament, starting with its main regulatory switch, the troponin complex. In this work, a comparative approach has been used to expand the description of how troponin genes have evolved in insects to reach their final tissue-specific isoform repertoires.
Similarly Evolved Troponin Complex Isoforms Share a Similar Expression Pattern
Comparing gene sequences in the four drosophilid species has made us aware of a new set of isoforms in both troponin I and T genes and we have established their specific expression in the IFM. Alternative splicing of TpnT exon 10A increases variability at the 3' end of these transcripts. Interestingly, the existence of two variable regions at both the Nt and Ct of the proteins causes the distribution of variability in insect TpnT and TpnI alternatively spliced regions to resemble more closely the situation of the vertebrate genes.
Table 1 highlights the coordinate expression of different transcripts of the three Tpn subunits in various drosophilid muscle types. In particular, it exposes the expression pattern coincidences between the TpnI exon 6's and the TpnC genes, and between TpnI exon 9/10 and TpnT exon 10A/B, also observed in Apis. All these expression partners seem to have evolved in a similar way as a result of their interactions in the thin-filament structure. Insect TpnC genes also show a complex evolutionary pattern, involving both divergence and convergence events (Herranz, Mateos, and Marco 2005). For instance, Type I TpnC genes, expressing larval hypodermic isoforms, have evolved independently in holometabolous insects, leading to two recently acquired isoforms in the drosophilid species, only one isoform in Anopheles, and two distantly related isoforms in Apis. TpnI 6a exons, the transcripts of which are mainly expressed in larval hypodermic muscles, may have evolved similarly. There is just one in Anopheles and two 6a exons in the Drosophilidae and Apis that could have been obtained independently in both insect orders. A different case of coupled evolution affects the TpnI exon 9 and TpnT exon 10A pair so that after appearing, probably at the beginning of the holometabolous insect branching, they became coexpressed in the very specialized IFM. In fact, this pair of exons shows higher sequence variability than its counterpart, the TnI exon 10 and TnT exon 10B pair. This higher variability in IFM-expressed isoforms can be related with the strong selection forces under these muscles.
Table 1 Summary of the Revised Drosophilid TpnT, TpnI, and TpnC Gene Expression Patterns
Alternatively Spliced Exon Coevolution Can Be Linked to the Tpn Complex Structure
Following these ideas, our aim was to explore how far a functional basis of this coexpression/coevolutionary trend could be established. Comparing the secondary structure predictions of the insect and human orthologous troponin proteins led to conclusions that, although the sequence conservation between deuterostomes and protostomes troponins is not very high, the structural predictions are conserved. For example, as can be seen in figure 7, the secondary structure predictions of Drosophila IFM TpnI isoform and the human cardiac TpnI are remarkably conserved. This conservation includes the interaction sites of TpnI with TpnC, Tm, TpnT, and actin as previously described in mammals (Filatov et al. 1999; Gordon, Homsher, and Regnier 2000; Gordon, Regnier, and Homsher 2001) and recently was confirmed in a crystal structure of half of a cardiac troponin complex (Takeda et al. 2003). In particular, the insect TpnI alternative exon 6b2 sequence is equivalent to that of the cardiac TpnI exon interacting with the TpnC Nt sequence or actin, depending on the calcium concentration. All these informations, together with structural data from vertebrates and from the insect Lethocerus (Wendt, Guenebaut, and Leonard 1997; Wendt and Leonard 1999), lead to a model for the troponin complex interactions in Drosophila that takes into account possible effects of isoform variability. Complex integrity may be stabilized by the interaction of the TpnI and TpnT variable Ct regions (exon 9/10 and exon 10A/10B, respectively) and TpnI Nt region with the globular Ct domain of TpnC. TpnT interacts with Tm mainly through its central domains, but its Ct polyglutamic tail and its Nt variability region (also polyglutamic rich) probably contribute to stabilize this interaction.
FIG. 7.— Comparison between the secondary structure predictions of Drosophila IFM and human cardiac TpnI isoforms. Predictions were obtained and represented using the PSIPRED software (McGuffin, Bryson, and Jones 2000). The explanation of symbols appears in the inset. Conserved features of the sequences appear boxed showing the vertebrate-known interactions among thin-filament components. The Drosophila TpnI alternative exons 6b2 and 9 are also indicated. For clarity, the long random-coiled predicted Drosophila exon 3 sequence, absent in other organisms, has not been included in the figure. The regions of interactions with other components of the vertebrate thin filament appear inside broken line contours. The conservation of the secondary structure is also remarkable in the rest of the Tpn isoforms predictions (data not shown).
The expression patterns (table 1), sequence evolutionary trends (figs. 2 and 3), and the secondary structure predictions (fig. 7) taken together indicate that the alternatively spliced exons in TpnI and TpnT genes have evolved in a concerted way, a result consistent with the interactions of their products in the troponin complex. However, this has occurred independently in the different insect orders, at least for TpnC genes (Herranz, Mateos, and Marco 2005) and for TmH/TpnH (Mateos et al., in press). The relationship between the variable regions of the troponin complex proteins and their putative interactions with coexpressed isoforms in different muscles or developmental stages are represented in figure 8 as an evolutionary/functional network for the different insect groups. The interactions among the different isoforms of the troponin-tropomyosin complex components produced by alternative splicing (TpnI and TpnT) or differential gene expression are indicated in the figure. The overall conservation of the relationships of TpnI with TpnC and TpnT can be observed among the insects studied. The main exception lies in the switching of the TmH for TnH in Apis, as discussed below. Furthermore, in the drosophilids, a TpnI exon 3 incorporating a PAANGKA heptad repeat is also only expressed in Drosophila IFM because Anopheles and Apis lack this exon. The reasons for this variability remain to be clarified.
FIG. 8.— Coevolution of the tissue-specific variable regions of the troponins. These schemes combine the available information on different gene and/or isoform expression patterns, evolutionary rates, and putative structural interactions in the different insect groups studied. Although each species show differences in how they generate their isoform repertoire, isoforms organize themselves in similar interaction networks. Gene names are in italics. Thorax-expressed isoforms appear in bold. Direct molecular interactions are indicated with broken lines. A question mark indicates an indirect or merely stabilizing interaction.
Troponin H Alanine/Proline-Rich Extensions Absent in Apis Tm1 Gene Are Located in the TpnI Gene
APPAEGA heptad repetitions are encoded in the IFM-specific exons (16 and 17) of the Drosophila tropomyosin Tm1 gene that produces two heavy tropomyosins (TmH 33 and 34) with long Ct extensions. In the Anopheles Tm1 gene, only one exon containing this kind of sequence has been located. Interestingly, in Apis the tropomyosin genes lack these long Ct extensions (Mateos et al., unpublished data) but a similar region is present in the TpnI gene where it is encoded by exons H1, H2, and H3 spliced together to produce a similarly sized, APPAEGA-rich extension (figs. 1 and 4). We have also found a transcript containing only the first 30% of the alanine/proline-rich extension in the IFM. The presence of these TpnH extended isoforms in the Apis TpnI gene does not affect the standard repertoire of TpnIs, which are similarly processed and expressed as in dipterans.
Asynchrony as a Trigger for Independent Coevolution of the Troponins
The main variability in the isoforms' expression and sequences occurs in the asynchronous IFM. Two ideas are generally accepted in relation to these muscles. First, no single biochemical feature is known that completely correlates with the stretch-activation phenomenon that is also observed in skeletal and cardiac muscles of vertebrates (Steiger 1977). The phenomenon is much stronger and persistent in insect IFM (Linari et al. 2004). Second, the IFM asynchrony has evolved independently several times during insect evolution (Cullen 1974; Pringle 1981; Dudley 2000; Josephson, Malamud, and Stokes 2000). Molecular data are no exception. The interspecies-independent but intraspecies-adapted evolution of both the TpnC genes (Herranz, Mateos, and Marco 2005) and the TpnT and TpnI gene–splicing variants could have played important roles in how the IFMs have achieved their special properties. The possible functional relevance of these alanine/proline-rich extension sequences to the evolution of the stretch-activation phenomenon has been discussed elsewhere (Mateos et al., unpublished data).
The Evolution of Supramolecular Complexes and the Functional Role of Isoform Patterns
Previously published work has addressed the issue of isoform variability and function in relation to its evolutionary conservation (e.g., Robson et al. 2000; Shagin et al. 2004), but, to our knowledge, a comprehensive analysis of the type performed in this article has not been attempted yet. Together, the higher variability of muscle structures in protostomes and the increasing availability of genome information across taxa open the way to understanding the independent coevolution of genes whose proteins are involved in supramolecular complexes. Our approach reveals coevolutionary trends in components of the complexes sharing tissue expression patterns. The advent of whole-genome sequences from further insect species will help in extending and refining our model for the evolution of the troponin complex. Some splicing isoforms or even complete genes that are weakly expressed or expressed in a very restricted pattern of tissues can be overlooked during a conventional study. What are the functional consequences of the expression of minor isoforms in a supramolecular complex? It could just provide a background repertoire of isoforms to improve and be used in remodeling evolutionary processes but its conservation across different species suggests a more direct role. In the case of insect thin filaments, the tropomyosin-troponin complex has been coevolving to respond to the functional necessities of the asynchronous flight musculature in different orders. Some alanine/proline-rich motifs have appeared and have become associated with different polypeptides in the complex. Different alternatively spliced exons encoding parts of different subunits of the complex, some of them also independently evolved, correlate with the type and function of the muscle. The sequencing effort in different organisms offers new opportunities to test, among other things, gene annotation accuracy, but more importantly the correlation of genotype changes and the functional phenotypic features that have evolved in particular groups of organisms.
Supplementary Material
Table A (GenBank accession numbers) and the unrooted trees in a PHYLIP format (showing bootstrapping values for TreeView application) used in this work are available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.org/).
Acknowledgements
We are especially grateful to R. Cripps, M. Manzanares, and J. Sparrow for correcting the English and for critical reading of the manuscript as well as two anonymous referees who have greatly helped in improving the quality of the manuscript. This research was supported by grants from the Spanish Government PB96-0069, ESP1999-0379-C02 and PNE-008/2001-C, BMC2001-1454.
References
Abi Rached, L., M. F. McDermott, and P. Pontarotti. 1999. The MHC big bang. Immunol. Rev. 167:33–44.
Adams, M. D., S. E. Celniker, R. A. Holt et al. (191 co-authors). 2000. The genome sequence of Drosophila melanogaster. Science 287(5461):2185–2195.
Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25(17):3389–3402.
Barbas, J. A., J. Galceran, L. Torroja, A. Prado, and A. Ferrus. 1993. Abnormal muscle development in the heldup3 mutant of Drosophila melanogaster is caused by a splicing defect affecting selected troponin I isoforms. Mol. Cell. Biol. 13(3):1433–1439.
Beall, C. J., and E. Fyrberg. 1991. Muscle abnormalities in Drosophila melanogaster heldup mutants are caused by missing or aberrant troponin-I isoforms. J. Cell Biol. 114(5):941–951.
Benoist, P., J. A. Mas, R. Marco, and M. Cervera. 1998. Differential muscle-type expression of the Drosophila troponin T gene. A 3-base pair microexon is involved in visceral and adult hypodermic muscle specification. J. Biol. Chem. 273(13):7538–7546.
Bullard, B., K. Leonard, A. Larkins, G. Butcher, C. Karlik, and E. Fyrberg. 1988. Troponin of asynchronous flight muscle. J. Mol. Biol. 204(3):621–637.
Cullen, M. J. 1974. The distribution of asynchronous muscle in insects with special reference to the Hemiptera: an electron microscope study. J. Entomol. 49:17–41.
Domingo, A., J. Gonzalez-Jurado, M. Maroto, C. Diaz, J. Vinos, C. Carrasco, M. Cervera, and R. Marco. 1998. Troponin-T is a calcium-binding protein in insect muscle: in vivo phosphorylation, muscle-specific isoforms and developmental profile in Drosophila melanogaster. J Muscle Res. Cell Motil. 19(4):393–403.
Dudley, R. 2000. The biomechanics of insect flight: form, function, evolution. Princeton University Press, Princeton, N. J.
Filatov, V. L., A. G. Katrukha, T. V. Bulargina, and N. B. Gusev. 1999. Troponin: structure, properties, and mechanism of functioning. Biochemistry (Mosc.) 64(9):969–985.
Fitzhugh, G., and J. Marden. 1997. Maturational changes in troponin T expression, Ca2+-sensitivity and twitch contraction kinetics in dragonfly flight muscle. J. Exp. Biol. 200(Pt 10):1473–1482.
Gaunt, M. W., and M. A. Miles. 2002. An insect molecular clock dates the origin of the insects and accords with palaeontological and biogeographic landmarks. Mol. Biol. Evol. 19(5):748–761.
Gordon, A. M., E. Homsher, and M. Regnier. 2000. Regulation of contraction in striated muscle. Physiol. Rev. 80(2):853–924.
Gordon, A. M., M. Regnier, and E. Homsher. 2001. Skeletal and cardiac muscle contractile activation: tropomyosin "rocks and rolls". News Physiol. Sci. 16:49–55.
Hall, T. A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41:95–98.
Hanke, P. D., and R. V. Storti. 1988. The Drosophila melanogaster tropomyosin II gene produces multiple proteins by use of alternative tissue-specific promoters and alternative splicing. Mol. Cell. Biol. 8(9):3591–3602.
Herranz, R., C. Díaz-Castillo, T. G. Nguyen, T. L. Lovato, R. M. Cripps, and R. Marco. 2004. Expression pattern characterization of the whole Troponin C gene repertoire during Drosophila development. Gene Expr. Patterns 4(2):183–190.
Herranz, R., J. Mateos, and R. Marco. 2005. Diversification and independent evolution of TpnC genes in insects. J. Mol. Evol. 60(1):31–44.
Holt, R. A., G. M. Subramanian, A. Halpern et al. (121 co-authors). 2002. The genome sequence of the malaria mosquito Anopheles gambiae. Science 298(5591):129–149.
Jones, D. T. 1999. Protein secondary structure prediction based on position-specific scoring matrices. J. Mol. Biol. 292(2):195–202.
Josephson, R. K., J. G. Malamud, and D. R. Stokes. 2000. Asynchronous muscle: a primer. J. Exp. Biol. 203(Pt 18):2713–2722.
Karlik, C. C., and E. A. Fyrberg. 1986. Two Drosophila melanogaster tropomyosin genes: structural and functional aspects. Mol. Cell. Biol. 6(6):1965–1973.
Linari, M., M. K. Reedy, M. C. Reedy, V. Lombardi, and G. Piazzesi. 2004. Ca-activation and stretch-activation in insect flight muscle. Biophys. J. 87(2):1101–1111.
Lovato, T. L., S. M. Meadows, P. W. Baker, J. C. Sparrow, and R. M. Cripps. 2001. Characterization of muscle actin genes in Drosophila virilis reveals significant molecular complexity in skeletal muscle types. Insect Mol. Biol. 10(4):333–340.
Lynch, M. 2002. Genomics. Gene duplication and evolution. Science 297(5583):945–947.
Lynch, M., and J. S. Conery. 2000. The evolutionary fate and consequences of duplicate genes. Science 290(5494):1151–1155.
Marden, J. H., G. H. Fitzhugh, M. Girgenrath, M. R. Wolf, and S. Girgenrath. 2001. Alternative splicing, muscle contraction and intraspecific variation: associations between troponin T transcripts, Ca(2+) sensitivity and the force and power output of dragonfly flight muscles during oscillatory contraction. J. Exp. Biol. 204(Pt 20):3457–3470.
Marden, J. H., G. H. Fitzhugh, M. R. Wolf, K. D. Arnold, and B. Rowan. 1999. Alternative splicing, muscle calcium sensitivity, and the modulation of dragonfly flight performance. Proc. Natl. Acad. Sci. USA 96(26):15304–15309.
McGuffin, L. J., K. Bryson, and D. T. Jones. 2000. The PSIPRED protein structure prediction server. Bioinformatics 16(4):404–405.
Mongin, E., C. Louis, R. A. Holt, E. Birney, and F. H. Collins. 2004. The Anopheles gambiae genome: an update. Trends Parasitol. 20(2):49–52.
Ohta, T. 2000. Evolution of gene families. Gene 259(1–2):45–52.
Page, R. D. 1996. TreeView: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci. 12(4):357–358.
Powell, J. R. 1997. Progress and prospects in evolutionary genetics: the Drosophila model. Oxford University Press, New York.
Pringle, J. W. 1981. The evolution of fibrillar muscle in insects. J. Exp. Biol. 94:1–14.
Qiu, F., A. Lakey, B. Agianian, A. Hutchings, G. W. Butcher, S. Labeit, K. Leonard, and B. Bullard. 2003. Troponin C in different insect muscle types: identification of two isoforms in Lethocerus, Drosophila and Anopheles that are specific to asynchronous flight muscle in the adult insect. Biochem. J. 371(Pt 3):811–821.
Roberts, G. C., and C. W. Smith. 2002. Alternative splicing: combinatorial output from the genome. Curr. Opin. Chem. Biol. 6(3):375–383.
Robson, P., G. M. Wright, J. H. Youson, and F. W. Keeley. 2000. The structure and organization of lamprin genes: multiple-copy genes with alternative splicing and convergent evolution with insect structural proteins. Mol. Biol. Evol. 17(11):1739–1752.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. 2nd edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Schmitz, S., C. J. Schankin, H. Prinz et al. (10 co-authors). 2003. Molecular evolutionary convergence of the flight muscle protein arthrin in Diptera and hemiptera. Mol. Biol. Evol. 20(12):2019–2033.
Shagin, D. A., E. V. Barsova, Y. G. Yanushevich et al. (13 co-authors). 2004. GFP-like proteins as ubiquitous metazoan superfamily: evolution of functional features and structural complexity. Mol. Biol. Evol. 21(5):841–850.
Steiger, G. J. 1977. Stretch-activation and tension transients in cardiac, skeletal and insect flight muscle. Pp. 221–268 in R. T. Tregear and J. W. Pringle, eds. Insect Flight Muscle. NorthHolland Press, Amsterdam. Insect Flight Muscle.
Takeda, S., A. Yamashita, K. Maeda, and Y. Maeda. 2003. Structure of the core domain of human cardiac troponin in the Ca(2+)-saturated form. Nature 424(6944):35–41.
Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22(22):4673–4680.
Wendt, T., V. Guenebaut, and K. R. Leonard. 1997. Structure of the Lethocerus troponin–tropomyosin complex as determined by electron microscopy. J. Struct. Biol. 118(1):1–8.
Wendt, T., and K. Leonard. 1999. Structure of the insect troponin complex. J. Mol. Biol. 285(4):1845–1856.
Wolf, M. R. 1999. Molecular and functional characterization of troponin T in Libellula pulchella and Periplaneta americana. Doctoral dissertation, Pennsylvania State University, University Park, Pa.
You, M., X. Xuan, N. Tsuji, T. Kamio, I. Igarashi, H. Nagasawa, T. Mikami, and K. Fujisaki. 2001. Molecular characterization of a troponin I-like protein from the hard tick Haemaphysalis longicornis. Insect Biochem. Mol. Biol. 32(1):67–73.(Raúl Herranz, Jesús Mateo)
E-mail: roberto.marco@uam.es.
Abstract
In bilaterians, the main regulator of muscle contraction is the troponin (Tpn) complex, comprising three closely interacting subunits (C, T, and I). To understand how evolutionary forces drive molecular change in protein complexes, we have compared the gene structures and expression patterns of Tpn genes in insects. In this class, while TpnC is encoded by multiple genes, TpnT and TpnI are encoded by single genes. Their isoform expression pattern is highly conserved within the Drosophilidae, and single orthologous genes were identified in the sequenced genomes of Drosophila pseudoobscura, Anopheles gambiae, and Apis mellifera. Apis expression patterns also support the equivalence of their exon organization throughout holometabolous insects. All TpnT genes include a previously unidentified indirect flight muscle (IFM)–specific exon (10A) that has evolved an expression pattern similar to that of exon 9 in TpnI. Thus, expression patterns, sequence evolution trends, and structural data indicate that Tpn genes and their isoforms have coevolved, building species- and muscle-specific troponin complexes. Furthermore, a clear case can be made for independent evolution of the IFM-specific isoforms containing alanine/proline-rich sequences. Dipteran genomes contain one tropomyosin gene that encodes one or two high–molecular weight isoforms (TmH) incorporating APPAEGA-rich sequences, specifically expressed in IFM. Corresponding exons do not exist in the Apis tropomyosin gene, but equivalent sequences occur in a high–molecular weight Apis IFM-specific TpnI isoform (TnH). Overall, our approach to comparatively analyze supramolecular complexes reveals coevolutionary trends not only in gene families but in isoforms generated by alternative splicing.
Key Words: troponin ? tropomyosin ? coevolution ? Insecta ? Drosophila ? Anopheles ? Apis ? phylogenetics
Introduction
One of the most intriguing aspects of molecular evolution is the multiplicity of isoforms produced by genes (Roberts and Smith 2002), the functional significance of which still largely escapes our understanding. Gene diversification has clearly relied on two mechanisms, either entire genes (or even chromosome fragments) are duplicated generating a gene family, or specific exons are duplicated along with the capacity for generating alternatively spliced transcripts. Many studies have compared the evolutionary advantages of each process (Lynch and Conery 2000; Ohta 2000; Lynch 2002). Exon-duplication events seem to be preferred if the result increases the variability of only small domains in a protein (Abi Rached, McDermott, and Pontarotti 1999).
Particularly relevant are the cases where groups of proteins interact to form high supramolecular complex structures, especially if the physiological effects of isoforms are well characterized. Our group has been studying these gene-evolution processes in troponin, a thin-filament complex, by focusing on one of the most diversified branches of life, the insects. The troponin complex comprises three subunits, namely troponin C (TpnC), troponin I (TpnI), and troponin T (TpnT), and mediates the thin-filament response to calcium when striated muscular contraction is initiated. In mammalian systems, extensive studies (Filatov et al. 1999; Gordon, Homsher, and Regnier 2000; Gordon, Regnier, and Homsher 2001) have addressed the structure-function relationships between the Tpn complex subunits. As a consequence of Ca2+ binding to TpnC, this polypeptide undergoes a conformational change, altering its interaction with TpnI, releasing the inhibitory TpnI-actin binding, and allowing movements of the troponin-trpomyosin complex on the F-actin that increases myosin binding to actin and thus promoting contraction.
The gene numbers for each troponin constituent vary among the metazoa. While only one copy each of the TpnT and TpnI genes occur in insects, up to three copies are found in vertebrates. There are only two TpnC genes in mammals (encoding cardiac and striated isoforms), but five to six troponin C genes have been found in holometabolous insects (Herranz, Mateos, and Marco 2005). Nevertheless, each troponin subunit exists in multiple isoforms. The single Drosophila troponin I gene has a total of 13 exons, and up to 10 different TpnI isoforms are produced by alternative splicing processes of some exons (Beall and Fyrberg 1991; Barbas et al. 1993). Variability regions are localized at specific points across the protein encoded by the alternative exons 3 and 9 and by four mutually exclusive exon 6's. The TpnT gene in both Drosophila melanogaster and Drosophila virilis was described as containing a set of 11 exons and up to four alternatively splicing isoforms, generated by the inclusion or exclusion of exons 3, 4, and 5 (Benoist et al. 1998). No splicing variants affecting the Ct half of the protein were described.
Muscle structure in protostomes is much more diverse than in deuterostomes. For example, holometabolous insects have two different musculatures, the larval one formed during embryogenesis, and the adult one formed during the pupal metamorphosis, when the muscles responsible for flight develop in the adult thorax. These muscles are the indirect flight muscles (IFMs), responsible for wing beating, and the tergal depressor of the trochanter (TDT) muscle, responsible for the jump at take-off of the fly. Stretch activation is an important property of IFM in many insects. Its direct consequence is that their oscillatory contractions are asynchronous, meaning that the nervous impulses received by the fibers are more irregular and less frequent than the contractions themselves. Specific protein variants are expressed in these muscles, sometimes encoded by different genes, as for example actin (Lovato et al. 2001) or troponin C (Qiu et al. 2003; Herranz et al. 2004), but more usually as the result of differential splicing of transcripts from a single gene. In addition, other muscle-specific components, such as arthrin and troponin H, are thin-filament proteins exclusively expressed in the IFMs. Arthrin is an ubiquitinated actin independently acquired in several insect orders (Schmitz et al. 2003). Troponin H (Bullard et al. 1988), a high–molecular weight component of Lethocerus flight muscles, was identified as similar to the heavy tropomyosin isoforms expressed in Drosophila IFM (Karlik and Fyrberg 1986; Hanke and Storti 1988). Troponin H and heavy tropomyosin (TmH; Mateos et al., unpublished data) contain alanine/proline-rich extensions in the Ct of the proteins.
In the present work, we describe how TpnI and TpnT isoforms have coevolved in insects. Our purpose in undertaking a systematic study of the insect troponin genes is to understand the coevolutionary mechanisms that led to functional divergence of isoform variation generated by the differential expression of the troponin complex genes. The specific repertoire of troponin isoforms expressed in distinct muscle types or stages of development has been established. In particular, we have analyzed the TpnI and TpnT genes of four Drosophilidae species (melanogaster, subobscura, pseudoobscura, and virilis), the evolution of which diverged during the last 60 Myr, as well as in Anopheles gambiae and Apis mellifera that diverged from the Drosophilidae around 250 and 300 MYA, respectively (Powell 1997; Dudley 2000; Gaunt and Miles 2002).
With these objectives, our studies have allowed us to identify previously undescribed features of the troponins T and I in insects. We have identified a new mutually excluding exon (10A) for the troponin T gene that produces an IFM/TDT differentially expressed isoform, adding a new variability region near the 3' end of the gene. We have refined the information on the TpnI expression patterns identifying TpnI exon 9 as more specific than originally described, and as the functional partner of the TpnT exon 10A isoform in the IFM. In contrast to the situation in Diptera, in Apis the alanine/proline-rich extensions do not occur attached to the Tm gene but are encoded in the TpnI gene, as three APPAEGA-rich 3' exons, H1, H2, and H3.
Materials and Methods
Insect Stocks
The Oregon R strain was used as wild-type source of D. melanogaster material. Drosophila subobscura was obtained from Rosa de Frutos (Valencia, Spain) and D. virilis was from Manuel Calleja (Madrid, Spain). These species diverged from D. melanogaster around 12 and 50 MYA (Powell 1997; Gaunt and Miles 2002). Apis mellifera samples were obtained in the Summer 2003 from Industrias Alonso located in El Vellon (Madrid). We have also included the data obtained from D. pseudoobscura (Human Genome Sequencing Project at Baylor College of Medicine Blast server) and the mosquito A. gambiae (Holt et al. 2002) genomes.
Nucleic Acid Extraction
Tissues from D. subobscura and D. virilis were obtained at different developmental stages as well as adult body parts (head, thorax, and abdomen). Dissections were done in cold acetone at –70°C. RNA extractions were made using TRIZOL Reagent (Invitrogen, Paisley, UK). The same procedure was carried out with A. mellifera (two different larval and two pupal stages, as well as recently emerged and mature honeybees, that were dissected to separate heads, thoraces, and abdomens). Specific muscles, IFM or TDT, were recovered from animals pretreated for at least 1 week in dehydrating acetone solution at –20°C, which facilitates the fiber isolation. Genomic DNA was extracted using a Tris-HCl 10 mM pH 7 homogenizing solution containing 0.5% sodium dodecyl sulfate and NaCl 60 mM, followed by a standard phenol-based purification (Sambrook, Fritsch, and Maniatis 1989).
Reverse Transcriptase–Polymerase Chain Reaction
We used total RNA (2 μg) and oligo-dT (1 μg) in order to normalize the amount of cDNA used as template in the polymerase chain reactions (PCRs). Specific probes were designed for each gene amplification, based on known sequences from D. melanogaster or A. mellifera. PCRs were carried out for 30–35 amplification cycles (94°C/30 s, 55–60°C/45 s, and 72°C/45 s) with thermostable polymerase (DyNAzime, Fynnzymes) in a Gene Amp 2700 System thermocycler of PE Applied Biosystems (Foster City, Calif.). For genomic fragment amplifications, hybridization and elongation times were extended up to 1 min and TpnI PCR experiments were performed in the presence of 2.5% dimethyl sulfoxide.
Rapid Amplification cDNA Extension
We have determined the 3' untranslated region (UTR) of the different TpnI transcripts in A. mellifera by using the Ambion FirstChoice RLM-RACE Kit, with a TpnI exon 8 probe plus the 3' outer and inner kit probes.
Cloning and Sequencing
Standard commercial protocols were followed for DNA extraction (Concert Rapid Gel Extraction, GIBCO), ligations (pGEM-T, Promega, Madison, Wisc.), Escherichia coli transformations (DH5 strain), and plasmid purification (Wizard Plus SV minipreps, Promega) sometimes followed by a phenol-based concentration step (Sambrook, Fritsch, and Maniatis 1989). Standard probes included in pGEM-T easy plasmid (SP6 and T7) were used for sequencing reactions in an automatic sequencer. All the detected reverse transcriptase (RT)–PCR products were sequenced in this way using internal oligonucleotides when necessary.
Bioinformatics Tools
Sequences were obtained and preliminarily compared by National Center for Biotechnology Information Blast (Altschul et al. 1997). Sequence information was processed using GeneJockey II software. Phylogenetic trees were obtained from sequence comparisons and alignments based on ClustalW with BIOEDIT (Hall 1999) software using "Fitch-Margoliash and least-squares distance methods" algorithm or remotely with Phylodendrom version 0.8d, beta 1999, and ClustalW version 1.75 (neighbor-joining method with 1,000 bootstraps [Thompson, Higgins, and Gibson 1994]). Tree output was rendered with TreeView version 1.6.6 (Page 1996). Branches with bootstrap values below 70% have been collapsed. The 2D structure predictions were obtained from the ExPASy web server using the PSIPRED method (Jones 1999; McGuffin, Bryson, and Jones 2000).
Accession Numbers
Sequences obtained or annotated by us have been submitted to the corresponding section of the GenBank with the following accession numbers: D. melanogaster and D. virilis TpnT gene, AY439172 and AY439178–9; TpnT and TpnI Drosophilidae transcripts, AY439173–7, AY439180–4, and BK001654–5; and A. gambiae and A. mellifera TpnT and TpnI genes, BK005279–82. This information has also been prepared in a Supplementary Material Table A (Supplementary material online) that also includes the accession numbers of the rest of the sequences used in this work. All gene annotations contain a detailed description of all the identified transcripts including their sequences and expression patterns.
Results
Conservation of Orthologous TpnI and TpnT Gene Structures in Insects
We have used the previously reported D. melanogaster TpnT and TpnI gene sequences (Barbas et al. 1993; Benoist et al. 1998) to find putative orthologues in the Anopheles and Apis genomes. The structural organization of these genes has been preserved to a high degree in insects (fig. 1). In general, we observe an increase in gene size, mainly due to intron length expansion from the smallest genome—178 Mb in D. melanogaster (Adams et al. 2000)—to those of Anopheles and Apis that are almost double in size (Holt et al. 2002). Chromosomal location is not a broadly preserved feature in insects, neither for TpnC genes (Herranz, Mateos, and Marco 2005) nor for those of TpnI or TpnT. The TpnT and TpnI genes are both located on the D. melanogaster X chromosome, but the Anopheles TpnT gene is located cytologically on chromosome arm 2R, while TpnI remains to be localized (Mongin et al. 2004).
FIG. 1.— Troponin T (top) and troponin I (bottom) gene structures in holometabolous insects. Genes are labeled with the species binomial two-letter abbreviation. Noncoding (white), constitutive (gray), alternative (black) or mutually exclusive (diagonal stripes), and the PAANGKA- or APPAEGA-containing (horizontal stripes) alternative exons are indicated. Light shaded exon 6's in TpnI have not yet been detected in RT-PCR experiments. Exon numbers in Anopheles and Apis genes have been assigned by homology to those of their Drosophila counterparts. Genomic DNA is represented by a continuous line that is therefore missing in gene sequences derived from cDNA data. Gene organization is well preserved overall, and gene size correlates with the organism genome size. The main difference affects alanine/proline-rich sequences including the TpnI exon 3 encoding the PAANGKA sequences in the Drosophilidae, but not in Anopheles or Apis. Several Apis TpnI gene exons encode an APPAEGA-containing sequence in their 3' region.
We have cloned and sequenced the transcripts of these genes in the drosophilid species and Apis, paying special attention to the alternatively spliced exons. In the case of troponin T, the differentially spliced Nt exons are present in all the analyzed species, although in Apis the TpnT gene shows an additional alternatively spliced exon, named 5'. The Apis equivalent to the dipteran exon 6 is divided into five constitutive exons (a, b, c, d, and e). Sequencing of transcripts identified a new variable exon 10, that had been overlooked in previous studies of D. melanogaster and D. virilis sequenced TpnT transcripts (Benoist et al. 1998). Comparison of the TpnT genomic sequences in insects showed that two different exons, named 10A and 10B, are present in all holometabolous species. Both are 79-nt long, encoding 26 aa. Another difference in the TpnT gene structure is found in exon 11, which encodes a long polyglutamic tail with variable size in all studied protostome species (Benoist et al. 1998).
The constitutive exons and exon 9 of TpnI gene are present in all analyzed insects (fig. 1), but exon 3 is not, to the extent that the Drosophila orthologous exons 2 and 4 are fused in a unique exon in Anopheles, but not in Apis where a smaller exon 3 is found the main variability of Troponin I arises from mutually exclusive exon 6's. In the case of Anopheles, we detected three exon 6 variants by sequence homology, while four were detected in Apis. Interestingly, additional exons, that we have designated H1, H2, and H3, were found downstream of exon 10 in the Apis TpnI gene, encoding the APPAEGA-repetitive sequences (see below).
Sequence Variations of the Alternatively Spliced Exons in TpnT and TpnI Genes
Amino acid sequences from TpnT and TpnI constitutive exons are conserved almost perfectly in insects. As shown in figure 2A, even the sequences of the alternatively spliced exons (3, 4, and 5) from TpnT are well conserved among the Drosophilidae. Exons 3 and 4 are also conserved in A. gambiae and A. mellifera while a higher degree of variation is found in exon 5, reflecting the evolutionary distance between these insects. In this region, the Apis TpnT gene has an additional alternatively spliced exon. Exons 3 and 4 are clearly conserved, but exons 5, 5', and the constitutive exon 6a have very different sequences, being more similar to sequences in the orthopteran Periplaneta (Wolf 1999) and the odonate Libellula TpnT exons (Fitzhugh and Marden 1997; Marden et al. 1999, 2001) despite the much larger evolutionary distance of these species from holometabolous insects.
FIG. 2.— Alternatively spliced exons in TpnT and TpnI genes. (A) Alignment of the 5' region of the troponin T transcripts in insects. Amino acid conservation (?), silent variation () (cDNA), equivalent variation (), and nonequivalent variation () in relation to the Drosophila melanogaster sequence are shown. These sequences, labeled with the two-letter species abbreviations and the corresponding exon number (E10A Dm means exon 10A in D. melanogaster for instance), are conserved in the four drosophilid species (only silent variation appears) and also in the Anopheles and Apis exons except exon 5. The Libellula (AF133521 [GenBank] ) and Periplaneta (AF133520 [GenBank] ) TpnT sequences (obtained from GenBank) retain the same properties although the alternative exon sequences show higher levels of variation. (B) Alignment and phylogenetic tree of the alternative TpnT exon 10 sequences in insects. Branches with bootstrap values below 70% have been collapsed in this cDNA-based tree. Most changes affect the number of phosphorylable residues, indicated at the right of the alignment, with the exception of the conserved final threonine marked with an asterisk. (C) Alignment and phylogenetic tree of the alternative TpnI exon 9 and 10 sequences in insects. The Haemaphysalis longicornis TpnI sequence (AB051079 [GenBank] ), in which exon 9 has not been described, was used as out-group.
The sequences of the mutually exclusive TpnT exons 10A and 10B (fig. 2B) are also well preserved among Drosophilidae. Larger differences are found in Anopheles and Apis. Nonequivalent variations in exon 10A are equally distributed throughout the exon, but less so in exon 10B. Interestingly, a higher number of threonine residues appear in insect exon 10A than in exon 10B, but only the last threonine of exon 10A is conserved in all the insects analyzed. Drosophila TpnT isoforms are readily phosphorylated in vivo (Domingo et al. 1998), precisely in the exon 10A–encoded sequence (Nongthomba and Sparrow, personal communication).
With regard to the TpnI genes, constitutive exons are almost strictly conserved and the sequence of alternative exons 9 and 10 (fig. 2C) reflect a situation similar to that of TpnT exons 10A and 10B. Even though we know that the alternatively spliced sequences are short, phylogenetic trees based on the comparison among the sequences encoded by these exons can be constructed. Although this information is not sufficient to establish the evolutionary origin of these exons, it can be used to detect how they have been changing in different insect groups when the data of all exons are analyzed together. In both Tpn genes, one of the duplicated exons (TpnT exon 10A and TpnI exon 9) appears to be less conserved than the other (TpnT exon 10B and TpnI exon 10), showing similar evolutionary patterns in each of the three insect groups analyzed. It is therefore likely that the duplication of these TpnT and TpnI alternative exons may have occurred at the origin of the holometabolous-type of development. In accordance with this idea, only a single TpnT exon 10 with intermediate sequence features has been found in the hemimetabolous Periplaneta americana and Libellula pulchella. Furthermore, only a single exon 9/10 has been found in the arachnid Haemaphysalis longicornis TpnI gene (You et al. 2001). Interestingly, TpnI exon 9 in the Drosophilidae does not contain a complete stop codon, which is formed by the splicing of its last nucleotide with the two first bases of exon 10 (Barbas et al. 1993). This feature is not observed in Anopheles or Apis, where stop codons and 3' UTRs are found both in exons 9 and 10.
The TpnI mutually exclusive exon 6 sequences have also been analyzed. In a consensus phylogenetic tree based on exon 6 nucleotide sequences (fig. 3A), the low bootstrap values at the base of the insect tree suggest an ancient origin for these duplication events, but the tree supports a clear differentiation of exon 6a's from exon 6b's. The 6b exons are variable enough to be phylogenetically discriminative. The 6a exons are so different among themselves that the Apis paralogous exons 6a1 and 6a2 are as related to each other as to their putative drosophilid orthologous exons (see the polytomy in the 6a exon branch). The presence of a single exon 6a in Anopheles and the loss of the exon 6a2 canonical donor-splicing signal in the Drosophilidae are also noteworthy. In a sequence alignment (fig. 3B), it can be seen that each Anopheles and Apis exon 6 is as closely related to each other as it is to those from the Drosophilidae. Finally, the Drosophilidae TpnI exon 3 (fig. 4A) are essentially alanine/proline-rich sequences. All show the PAANGKA motif which appears repeated several times at different positions in the D. melanogaster, D. subobscura, and D. virilis extensions. This domain is absent in Anopheles and in Apis TpnIs, where a very short constitutive exon 3 is found.
FIG. 3.— Phylogenetic tree and alignment of the differentially spliced TpnI exon 6's. (A) Branches with bootstrap values below 70% have been collapsed in this cDNA-based tree. Exons 6b1 and 6b2 are clearly separated, but an exon 6a phylogeny cannot be extracted from the tree. (B) Alignment of the translated exon 6 sequences (34 aa translated from the last nucleotide of exon 5 plus the exon 6 nucleotides), flanked by the splicing sequences when available. The sequences are designated by the two-letter species abbreviation and the corresponding exon number (Dm E6b1 for instance). The 6a exons are the more variable ones, having even lost the canonical splicing donor sequence (bold and italics letters in the figure) in the drosophilid 6a2 exons.
FIG. 4.— Alignment of the IFM-specific troponin I and tropomyosin alanine/proline-rich extensions. (A) Although the TpnI exon 3 sequences are not perfectly conserved even in the Drosophilidae, the heptad PAANGKA (shaded box) is conserved. It is repeated in some species surrounded by an alanine/proline-rich region. This sequence is absent in Anopheles and Apis because their TpnI genes lack an exon 3 with these properties. (B) The TpnH extension is encoded by a single Tm1 gene exon in Anopheles, while in the Apis TpnI gene it involves up to three gene 3' exons (separated in the three rows of the alignment), all of them encoding an alanine/proline-rich sequence with the APPAEGA motif (shaded box). An equivalent sequence has been found in the Lethocerus TpnI gene (AJ621044 [GenBank] ) and is included as the final sequence in the alignment.
Three new exons have been located downstream from exon 10 in the Apis TpnI gene. These exons can be spliced together to produce an Apis-specific TpnI isoform containing a Ct extension encoding a repetitive APPAEGA sequence, through a new GT-splicing donor site in exon 10, located 25 bp before the stop codon. This extension is clearly similar to the alanine/proline-rich extension of the large–molecular weight IFM-specific tropomyosin in the Diptera, both in sequence and protein length (fig. 4B). Interestingly, the Apis tropomyosin Tm1 gene lacks this type of sequence (Mateos et al., unpublished results). A similar repetitive alanine/proline extension is found at the 3' end of a TpnI gene transcript from the hemipteran Lethocerus (Qiu et al. 2003 and sequence AJ621044).
Expression Pattern of TpnI and TpnT Genes Are Conserved in Drosophilidae
The expression profiles of TpnI and TpnT isoforms in D. subobscura and D. virilis were detected using RT-PCR techniques (unpublished data). All the transcripts previously identified in D. melanogaster (Barbas et al. 1993; Benoist et al. 1998) were detected in the second instar larvae and late pupae stages of these drosophilids. Combinations of the troponin T alternative exons 3, 4, and 5 appeared in larvae or in adult abdominal muscle transcripts in patterns characteristic for each muscle. The major transcripts of the adult thoracic muscles do not contain any of these alternative exons. Troponin I exon 9 inclusion is found exclusively in transcripts of the adult musculature, being accompanied by exon 3 in the major thoracic transcript. The four exon 6 types were detected at varying levels in all the tissues studied, those containing exon 6b1 being the more highly expressed in IFM (Barbas et al. 1993).
The discovery of the new TpnT exon 10A led to a study of its expression pattern. A single probe for constitutive exon 6 and four specific probes for exons 10A or 10B (fig. 5A) were used for RT-PCR. In embryos, only the 10B exon-containing transcript was detected, while in whole adult RNAs, transcripts containing either 10A or 10B appear.
FIG. 5.— Differences in the expression profile of TpnT and TpnI transcripts in the Drosophila melanogaster musculature. Agarose 1.2% gel separations of RT-PCRs. Each transcript detected by PCR as a band is marked with an arrow labeled with the name of the amplified gene and its exon composition. (A) Expression pattern of mutually exclusive exon 10's detected in adult or embryonic RNA extractions with four exon 10A- or 10B-specific primers (10A1, 10A2, 10B1, and 10B2). (B) Patterns of expression in adult body parts (head, IFM, TDT, and abdomen). Using probes for TpnT exon 6 and 10A (10A2 probe) or 10B (10B1 probe), the exon 10 expression pattern is shown. Exon 10B is expressed in all muscles in the adult except in the IFM. Exon 10A is the only one expressed in the IFM, but it is also expressed in the TDT muscles. (C) Probes for the 3' region of the TpnI gene were used in RT-PCRs (exons 8–9 and 8–10). Exon 9 uses the first two residues of exon 10 to generate a stop codon, so the exon 10 probe detects it as a lower band, the transcripts lacking exon 9, and as a higher band, the transcripts containing exon 9. So exon 9 is expressed mainly in IFM and only weakly in TDT muscles. (D) Using exon 2 and exon 9 probes for TpnI, it can be seen that exon 3 only appears heavily expressed in thoracic muscles including always exon 9. RT-negative control lanes are not shown.
Detection of exon 10A was particularly strong in the thorax, suggesting a specific role of the sequence encoded by this exon in the flight-related musculature (data not shown). Exon 10A- and 10B-specific probes detected, by RT-PCR, the classic 10B exon in the head, abdomen, and dissected TDT fibers (fig. 5B). Exon 10A expression was found almost exclusively in the RT-PCR product from RNA extracted from IFM and TDT fibers although some residual expression was also detected in abdomens. Consequently, IFM TpnT contains sequences encoded only by exon 10A, while TDT muscles contain a mixture of transcripts with exon 10A or 10B. The same thoracic enrichment of exon 10A transcripts was obtained in D. subobscura and D. virilis.
TpnI alternatively spliced exons were detected using several probe combinations. Exon 9 is present in TDT and IFM transcripts (fig. 5C), but while in IFMs it is the only one expressed, in TDT it is almost completely replaced by the exon 10–containing transcripts. Using probes for the complete coding region (exon 2 to exon 9/10) in the adult body parts showed that the alternatively spliced exon 3 is expressed in TDT/IFM, but is only present in transcripts that also incorporate exon 9 (fig. 5D).
Expression Profile of the TpnT and TpnI Genes in A. mellifera
The same procedure has been carried out with the honeybee to discover if the same patterns of isoform abundance occur. The expression profile of the TpnT transcripts during A. mellifera development (fig. 6A) shows that larval muscles contain a mixture of transcripts with alternative exons 3, 4, and 5' or exons 3 and 4. The IFMs contain an isoform encoded by a transcript lacking all the 5' region alternatively spliced exons. Other adult muscles contain a mixture of transcripts containing the exons 3, 4, 5, and 5' or 3, 4, and 5. In relation to the 3' half of the genes, exon 10B–containing transcripts were found in all muscles except in IFM, where it is only marginally detected (dorsoventral indirect flight muscle) or not at all (dorsolateral indirect flight muscle). Exon 10A–containing transcripts are adult thorax-specific, similar to the drosophilid results (see above).
FIG. 6.— Expression profile of TpnI and TpnT transcripts in Apis mellifera. Each transcript detected by PCR as a band is marked with an arrow labeled with the name of the amplified gene and its exon composition. Adjacent empty lanes to each PCR lane are negative controls (mRNA without reverse transcription). (A) Troponin T expression profile during development (larval L1 and L2 and pupal P1 and P2 stages) and in the thorax muscles (dorsoventral indirect flight muscles [DV-IFM], dorsolateral indirect flight muscles [DL-IFM], legs, and other thoracic muscles) of Apis. Apis TpnT shows a similar expression pattern to that described in Drosophila. (B) The TpnI expression pattern in the same sample is shown. The alternative exons, 9 and 10, both containing a proper stop codon and terminator signals, are expressed in different levels in all muscles and stages. Using probes from exon 4 and exon H1 of the Apis TpnI gene, we have detected an IFM-specific expression pattern for this TpnH isoform in hymenopterans. The four exon 6's identified are expressed in the TpnI transcripts (detected by inner PCR using internal exon 6 type-specific probes in a second PCR using the P2 fraction band as substrate DNA), except in the IFM-specific transcript where only 6b exons are detected. (C) The 3' rapid amplification cDNA extension (RACE) experiments were performed with thorax, IFM, and larval samples to detect the transcription stops in the Apis TpnI gene. Transcripts representing eight different stop sites have been proportionally represented on the left, with their size indicated and different arrows styles (depending on the transcripts last exon, dotted line for exon 9, continuous line for exon 10, and broken line for exon H's) signaling the corresponding band in the RACE electrophoresis gels. Three of them (one transcript for each group) are the major ones used in the thoracic muscles (their lengths are shown inside circles), but the transcript containing the TpnH exons is absent in larval muscles. RACE-negative control lanes are shown (– lanes).
The TpnI expression pattern detected using RT-PCR is shown in figure 6B. Probes for the region between exons 4 and 9 or exons 4 and 10 revealed that transcripts containing exon 9 or exon 10 are detected in the different developmental stages and in all muscles of the adult Apis musculature. Despite inherent limitations to these PCR experiments, exon 9–containing transcripts seem to appear later than those containing exon 10. Using probes for each specific exon 6 and exon 8 in an inner PCR of the previous amplified products, the four exon 6's in both exon 9– and exon 10–containing transcripts were detected. Although quantitative PCR was not carried out, the results indicate that maximum expression levels of exons 6a and 10 occur during larval stages, while those of 6b and 9 exons occur in adults.
Finally, RT-PCR using probes designed to amplify the region between exon 4 and exon H1 of the Apis TpnI gene showed, as expected, that a heavy Apis TpnI (troponin H) is expressed exclusively in the IFM (fig. 6B). This band was cloned and tested for its exon 6 content, using an exon 8 probe in combination with a specific probe for each one of the four exon 6. Sequencing of 14 clones confirmed that 7 clones contained exon 6b1 and the other 7 clones contained exon 6b2. In addition, these TpnH transcripts include a shorter spliced version of exon 10 that could not be detected in earlier RT-PCR assays using the exon 10 probe. In order to locate the 3' ends of the TpnI transcripts, we have performed rapid amplification cDNA extension experiments with three representative Apis RNA fractions (fig. 6C). We found up to eight different transcript ends for the TpnI gene, but only three major ones occur in mRNA from the thoracic muscles and IFM samples. Two of them are 3' to exon 9 and exon 10, produce the standard TpnI transcripts, and are also used in the larval muscles. The other end incorporates the initial part of exon 10 but extends into exons H1, H2, and H3, producing a TpnH isoform with a long alanine/proline-rich extension. We have also detected a transcript in which exon H1 does not use its acceptor-splicing site and this would produce a TnH isoform with a shorter alanine/proline-rich extension.
Discussion
It is generally accepted that muscle tissue arose early on in the evolution of bilateria. Muscle structure and its mechanism of function have been retained relatively unchanged since that time, although individual muscle tissues are fine-tuned to reflect their physiological functions. Because the myofibril comprises a series of supramolecular complexes, studying the evolution of an individual protein must be carried out in the context of the complex in which it is a part. We have initiated this analysis by focusing on one of the better-characterized substructures, the thin filament, starting with its main regulatory switch, the troponin complex. In this work, a comparative approach has been used to expand the description of how troponin genes have evolved in insects to reach their final tissue-specific isoform repertoires.
Similarly Evolved Troponin Complex Isoforms Share a Similar Expression Pattern
Comparing gene sequences in the four drosophilid species has made us aware of a new set of isoforms in both troponin I and T genes and we have established their specific expression in the IFM. Alternative splicing of TpnT exon 10A increases variability at the 3' end of these transcripts. Interestingly, the existence of two variable regions at both the Nt and Ct of the proteins causes the distribution of variability in insect TpnT and TpnI alternatively spliced regions to resemble more closely the situation of the vertebrate genes.
Table 1 highlights the coordinate expression of different transcripts of the three Tpn subunits in various drosophilid muscle types. In particular, it exposes the expression pattern coincidences between the TpnI exon 6's and the TpnC genes, and between TpnI exon 9/10 and TpnT exon 10A/B, also observed in Apis. All these expression partners seem to have evolved in a similar way as a result of their interactions in the thin-filament structure. Insect TpnC genes also show a complex evolutionary pattern, involving both divergence and convergence events (Herranz, Mateos, and Marco 2005). For instance, Type I TpnC genes, expressing larval hypodermic isoforms, have evolved independently in holometabolous insects, leading to two recently acquired isoforms in the drosophilid species, only one isoform in Anopheles, and two distantly related isoforms in Apis. TpnI 6a exons, the transcripts of which are mainly expressed in larval hypodermic muscles, may have evolved similarly. There is just one in Anopheles and two 6a exons in the Drosophilidae and Apis that could have been obtained independently in both insect orders. A different case of coupled evolution affects the TpnI exon 9 and TpnT exon 10A pair so that after appearing, probably at the beginning of the holometabolous insect branching, they became coexpressed in the very specialized IFM. In fact, this pair of exons shows higher sequence variability than its counterpart, the TnI exon 10 and TnT exon 10B pair. This higher variability in IFM-expressed isoforms can be related with the strong selection forces under these muscles.
Table 1 Summary of the Revised Drosophilid TpnT, TpnI, and TpnC Gene Expression Patterns
Alternatively Spliced Exon Coevolution Can Be Linked to the Tpn Complex Structure
Following these ideas, our aim was to explore how far a functional basis of this coexpression/coevolutionary trend could be established. Comparing the secondary structure predictions of the insect and human orthologous troponin proteins led to conclusions that, although the sequence conservation between deuterostomes and protostomes troponins is not very high, the structural predictions are conserved. For example, as can be seen in figure 7, the secondary structure predictions of Drosophila IFM TpnI isoform and the human cardiac TpnI are remarkably conserved. This conservation includes the interaction sites of TpnI with TpnC, Tm, TpnT, and actin as previously described in mammals (Filatov et al. 1999; Gordon, Homsher, and Regnier 2000; Gordon, Regnier, and Homsher 2001) and recently was confirmed in a crystal structure of half of a cardiac troponin complex (Takeda et al. 2003). In particular, the insect TpnI alternative exon 6b2 sequence is equivalent to that of the cardiac TpnI exon interacting with the TpnC Nt sequence or actin, depending on the calcium concentration. All these informations, together with structural data from vertebrates and from the insect Lethocerus (Wendt, Guenebaut, and Leonard 1997; Wendt and Leonard 1999), lead to a model for the troponin complex interactions in Drosophila that takes into account possible effects of isoform variability. Complex integrity may be stabilized by the interaction of the TpnI and TpnT variable Ct regions (exon 9/10 and exon 10A/10B, respectively) and TpnI Nt region with the globular Ct domain of TpnC. TpnT interacts with Tm mainly through its central domains, but its Ct polyglutamic tail and its Nt variability region (also polyglutamic rich) probably contribute to stabilize this interaction.
FIG. 7.— Comparison between the secondary structure predictions of Drosophila IFM and human cardiac TpnI isoforms. Predictions were obtained and represented using the PSIPRED software (McGuffin, Bryson, and Jones 2000). The explanation of symbols appears in the inset. Conserved features of the sequences appear boxed showing the vertebrate-known interactions among thin-filament components. The Drosophila TpnI alternative exons 6b2 and 9 are also indicated. For clarity, the long random-coiled predicted Drosophila exon 3 sequence, absent in other organisms, has not been included in the figure. The regions of interactions with other components of the vertebrate thin filament appear inside broken line contours. The conservation of the secondary structure is also remarkable in the rest of the Tpn isoforms predictions (data not shown).
The expression patterns (table 1), sequence evolutionary trends (figs. 2 and 3), and the secondary structure predictions (fig. 7) taken together indicate that the alternatively spliced exons in TpnI and TpnT genes have evolved in a concerted way, a result consistent with the interactions of their products in the troponin complex. However, this has occurred independently in the different insect orders, at least for TpnC genes (Herranz, Mateos, and Marco 2005) and for TmH/TpnH (Mateos et al., in press). The relationship between the variable regions of the troponin complex proteins and their putative interactions with coexpressed isoforms in different muscles or developmental stages are represented in figure 8 as an evolutionary/functional network for the different insect groups. The interactions among the different isoforms of the troponin-tropomyosin complex components produced by alternative splicing (TpnI and TpnT) or differential gene expression are indicated in the figure. The overall conservation of the relationships of TpnI with TpnC and TpnT can be observed among the insects studied. The main exception lies in the switching of the TmH for TnH in Apis, as discussed below. Furthermore, in the drosophilids, a TpnI exon 3 incorporating a PAANGKA heptad repeat is also only expressed in Drosophila IFM because Anopheles and Apis lack this exon. The reasons for this variability remain to be clarified.
FIG. 8.— Coevolution of the tissue-specific variable regions of the troponins. These schemes combine the available information on different gene and/or isoform expression patterns, evolutionary rates, and putative structural interactions in the different insect groups studied. Although each species show differences in how they generate their isoform repertoire, isoforms organize themselves in similar interaction networks. Gene names are in italics. Thorax-expressed isoforms appear in bold. Direct molecular interactions are indicated with broken lines. A question mark indicates an indirect or merely stabilizing interaction.
Troponin H Alanine/Proline-Rich Extensions Absent in Apis Tm1 Gene Are Located in the TpnI Gene
APPAEGA heptad repetitions are encoded in the IFM-specific exons (16 and 17) of the Drosophila tropomyosin Tm1 gene that produces two heavy tropomyosins (TmH 33 and 34) with long Ct extensions. In the Anopheles Tm1 gene, only one exon containing this kind of sequence has been located. Interestingly, in Apis the tropomyosin genes lack these long Ct extensions (Mateos et al., unpublished data) but a similar region is present in the TpnI gene where it is encoded by exons H1, H2, and H3 spliced together to produce a similarly sized, APPAEGA-rich extension (figs. 1 and 4). We have also found a transcript containing only the first 30% of the alanine/proline-rich extension in the IFM. The presence of these TpnH extended isoforms in the Apis TpnI gene does not affect the standard repertoire of TpnIs, which are similarly processed and expressed as in dipterans.
Asynchrony as a Trigger for Independent Coevolution of the Troponins
The main variability in the isoforms' expression and sequences occurs in the asynchronous IFM. Two ideas are generally accepted in relation to these muscles. First, no single biochemical feature is known that completely correlates with the stretch-activation phenomenon that is also observed in skeletal and cardiac muscles of vertebrates (Steiger 1977). The phenomenon is much stronger and persistent in insect IFM (Linari et al. 2004). Second, the IFM asynchrony has evolved independently several times during insect evolution (Cullen 1974; Pringle 1981; Dudley 2000; Josephson, Malamud, and Stokes 2000). Molecular data are no exception. The interspecies-independent but intraspecies-adapted evolution of both the TpnC genes (Herranz, Mateos, and Marco 2005) and the TpnT and TpnI gene–splicing variants could have played important roles in how the IFMs have achieved their special properties. The possible functional relevance of these alanine/proline-rich extension sequences to the evolution of the stretch-activation phenomenon has been discussed elsewhere (Mateos et al., unpublished data).
The Evolution of Supramolecular Complexes and the Functional Role of Isoform Patterns
Previously published work has addressed the issue of isoform variability and function in relation to its evolutionary conservation (e.g., Robson et al. 2000; Shagin et al. 2004), but, to our knowledge, a comprehensive analysis of the type performed in this article has not been attempted yet. Together, the higher variability of muscle structures in protostomes and the increasing availability of genome information across taxa open the way to understanding the independent coevolution of genes whose proteins are involved in supramolecular complexes. Our approach reveals coevolutionary trends in components of the complexes sharing tissue expression patterns. The advent of whole-genome sequences from further insect species will help in extending and refining our model for the evolution of the troponin complex. Some splicing isoforms or even complete genes that are weakly expressed or expressed in a very restricted pattern of tissues can be overlooked during a conventional study. What are the functional consequences of the expression of minor isoforms in a supramolecular complex? It could just provide a background repertoire of isoforms to improve and be used in remodeling evolutionary processes but its conservation across different species suggests a more direct role. In the case of insect thin filaments, the tropomyosin-troponin complex has been coevolving to respond to the functional necessities of the asynchronous flight musculature in different orders. Some alanine/proline-rich motifs have appeared and have become associated with different polypeptides in the complex. Different alternatively spliced exons encoding parts of different subunits of the complex, some of them also independently evolved, correlate with the type and function of the muscle. The sequencing effort in different organisms offers new opportunities to test, among other things, gene annotation accuracy, but more importantly the correlation of genotype changes and the functional phenotypic features that have evolved in particular groups of organisms.
Supplementary Material
Table A (GenBank accession numbers) and the unrooted trees in a PHYLIP format (showing bootstrapping values for TreeView application) used in this work are available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.org/).
Acknowledgements
We are especially grateful to R. Cripps, M. Manzanares, and J. Sparrow for correcting the English and for critical reading of the manuscript as well as two anonymous referees who have greatly helped in improving the quality of the manuscript. This research was supported by grants from the Spanish Government PB96-0069, ESP1999-0379-C02 and PNE-008/2001-C, BMC2001-1454.
References
Abi Rached, L., M. F. McDermott, and P. Pontarotti. 1999. The MHC big bang. Immunol. Rev. 167:33–44.
Adams, M. D., S. E. Celniker, R. A. Holt et al. (191 co-authors). 2000. The genome sequence of Drosophila melanogaster. Science 287(5461):2185–2195.
Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25(17):3389–3402.
Barbas, J. A., J. Galceran, L. Torroja, A. Prado, and A. Ferrus. 1993. Abnormal muscle development in the heldup3 mutant of Drosophila melanogaster is caused by a splicing defect affecting selected troponin I isoforms. Mol. Cell. Biol. 13(3):1433–1439.
Beall, C. J., and E. Fyrberg. 1991. Muscle abnormalities in Drosophila melanogaster heldup mutants are caused by missing or aberrant troponin-I isoforms. J. Cell Biol. 114(5):941–951.
Benoist, P., J. A. Mas, R. Marco, and M. Cervera. 1998. Differential muscle-type expression of the Drosophila troponin T gene. A 3-base pair microexon is involved in visceral and adult hypodermic muscle specification. J. Biol. Chem. 273(13):7538–7546.
Bullard, B., K. Leonard, A. Larkins, G. Butcher, C. Karlik, and E. Fyrberg. 1988. Troponin of asynchronous flight muscle. J. Mol. Biol. 204(3):621–637.
Cullen, M. J. 1974. The distribution of asynchronous muscle in insects with special reference to the Hemiptera: an electron microscope study. J. Entomol. 49:17–41.
Domingo, A., J. Gonzalez-Jurado, M. Maroto, C. Diaz, J. Vinos, C. Carrasco, M. Cervera, and R. Marco. 1998. Troponin-T is a calcium-binding protein in insect muscle: in vivo phosphorylation, muscle-specific isoforms and developmental profile in Drosophila melanogaster. J Muscle Res. Cell Motil. 19(4):393–403.
Dudley, R. 2000. The biomechanics of insect flight: form, function, evolution. Princeton University Press, Princeton, N. J.
Filatov, V. L., A. G. Katrukha, T. V. Bulargina, and N. B. Gusev. 1999. Troponin: structure, properties, and mechanism of functioning. Biochemistry (Mosc.) 64(9):969–985.
Fitzhugh, G., and J. Marden. 1997. Maturational changes in troponin T expression, Ca2+-sensitivity and twitch contraction kinetics in dragonfly flight muscle. J. Exp. Biol. 200(Pt 10):1473–1482.
Gaunt, M. W., and M. A. Miles. 2002. An insect molecular clock dates the origin of the insects and accords with palaeontological and biogeographic landmarks. Mol. Biol. Evol. 19(5):748–761.
Gordon, A. M., E. Homsher, and M. Regnier. 2000. Regulation of contraction in striated muscle. Physiol. Rev. 80(2):853–924.
Gordon, A. M., M. Regnier, and E. Homsher. 2001. Skeletal and cardiac muscle contractile activation: tropomyosin "rocks and rolls". News Physiol. Sci. 16:49–55.
Hall, T. A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41:95–98.
Hanke, P. D., and R. V. Storti. 1988. The Drosophila melanogaster tropomyosin II gene produces multiple proteins by use of alternative tissue-specific promoters and alternative splicing. Mol. Cell. Biol. 8(9):3591–3602.
Herranz, R., C. Díaz-Castillo, T. G. Nguyen, T. L. Lovato, R. M. Cripps, and R. Marco. 2004. Expression pattern characterization of the whole Troponin C gene repertoire during Drosophila development. Gene Expr. Patterns 4(2):183–190.
Herranz, R., J. Mateos, and R. Marco. 2005. Diversification and independent evolution of TpnC genes in insects. J. Mol. Evol. 60(1):31–44.
Holt, R. A., G. M. Subramanian, A. Halpern et al. (121 co-authors). 2002. The genome sequence of the malaria mosquito Anopheles gambiae. Science 298(5591):129–149.
Jones, D. T. 1999. Protein secondary structure prediction based on position-specific scoring matrices. J. Mol. Biol. 292(2):195–202.
Josephson, R. K., J. G. Malamud, and D. R. Stokes. 2000. Asynchronous muscle: a primer. J. Exp. Biol. 203(Pt 18):2713–2722.
Karlik, C. C., and E. A. Fyrberg. 1986. Two Drosophila melanogaster tropomyosin genes: structural and functional aspects. Mol. Cell. Biol. 6(6):1965–1973.
Linari, M., M. K. Reedy, M. C. Reedy, V. Lombardi, and G. Piazzesi. 2004. Ca-activation and stretch-activation in insect flight muscle. Biophys. J. 87(2):1101–1111.
Lovato, T. L., S. M. Meadows, P. W. Baker, J. C. Sparrow, and R. M. Cripps. 2001. Characterization of muscle actin genes in Drosophila virilis reveals significant molecular complexity in skeletal muscle types. Insect Mol. Biol. 10(4):333–340.
Lynch, M. 2002. Genomics. Gene duplication and evolution. Science 297(5583):945–947.
Lynch, M., and J. S. Conery. 2000. The evolutionary fate and consequences of duplicate genes. Science 290(5494):1151–1155.
Marden, J. H., G. H. Fitzhugh, M. Girgenrath, M. R. Wolf, and S. Girgenrath. 2001. Alternative splicing, muscle contraction and intraspecific variation: associations between troponin T transcripts, Ca(2+) sensitivity and the force and power output of dragonfly flight muscles during oscillatory contraction. J. Exp. Biol. 204(Pt 20):3457–3470.
Marden, J. H., G. H. Fitzhugh, M. R. Wolf, K. D. Arnold, and B. Rowan. 1999. Alternative splicing, muscle calcium sensitivity, and the modulation of dragonfly flight performance. Proc. Natl. Acad. Sci. USA 96(26):15304–15309.
McGuffin, L. J., K. Bryson, and D. T. Jones. 2000. The PSIPRED protein structure prediction server. Bioinformatics 16(4):404–405.
Mongin, E., C. Louis, R. A. Holt, E. Birney, and F. H. Collins. 2004. The Anopheles gambiae genome: an update. Trends Parasitol. 20(2):49–52.
Ohta, T. 2000. Evolution of gene families. Gene 259(1–2):45–52.
Page, R. D. 1996. TreeView: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci. 12(4):357–358.
Powell, J. R. 1997. Progress and prospects in evolutionary genetics: the Drosophila model. Oxford University Press, New York.
Pringle, J. W. 1981. The evolution of fibrillar muscle in insects. J. Exp. Biol. 94:1–14.
Qiu, F., A. Lakey, B. Agianian, A. Hutchings, G. W. Butcher, S. Labeit, K. Leonard, and B. Bullard. 2003. Troponin C in different insect muscle types: identification of two isoforms in Lethocerus, Drosophila and Anopheles that are specific to asynchronous flight muscle in the adult insect. Biochem. J. 371(Pt 3):811–821.
Roberts, G. C., and C. W. Smith. 2002. Alternative splicing: combinatorial output from the genome. Curr. Opin. Chem. Biol. 6(3):375–383.
Robson, P., G. M. Wright, J. H. Youson, and F. W. Keeley. 2000. The structure and organization of lamprin genes: multiple-copy genes with alternative splicing and convergent evolution with insect structural proteins. Mol. Biol. Evol. 17(11):1739–1752.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. 2nd edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Schmitz, S., C. J. Schankin, H. Prinz et al. (10 co-authors). 2003. Molecular evolutionary convergence of the flight muscle protein arthrin in Diptera and hemiptera. Mol. Biol. Evol. 20(12):2019–2033.
Shagin, D. A., E. V. Barsova, Y. G. Yanushevich et al. (13 co-authors). 2004. GFP-like proteins as ubiquitous metazoan superfamily: evolution of functional features and structural complexity. Mol. Biol. Evol. 21(5):841–850.
Steiger, G. J. 1977. Stretch-activation and tension transients in cardiac, skeletal and insect flight muscle. Pp. 221–268 in R. T. Tregear and J. W. Pringle, eds. Insect Flight Muscle. NorthHolland Press, Amsterdam. Insect Flight Muscle.
Takeda, S., A. Yamashita, K. Maeda, and Y. Maeda. 2003. Structure of the core domain of human cardiac troponin in the Ca(2+)-saturated form. Nature 424(6944):35–41.
Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22(22):4673–4680.
Wendt, T., V. Guenebaut, and K. R. Leonard. 1997. Structure of the Lethocerus troponin–tropomyosin complex as determined by electron microscopy. J. Struct. Biol. 118(1):1–8.
Wendt, T., and K. Leonard. 1999. Structure of the insect troponin complex. J. Mol. Biol. 285(4):1845–1856.
Wolf, M. R. 1999. Molecular and functional characterization of troponin T in Libellula pulchella and Periplaneta americana. Doctoral dissertation, Pennsylvania State University, University Park, Pa.
You, M., X. Xuan, N. Tsuji, T. Kamio, I. Igarashi, H. Nagasawa, T. Mikami, and K. Fujisaki. 2001. Molecular characterization of a troponin I-like protein from the hard tick Haemaphysalis longicornis. Insect Biochem. Mol. Biol. 32(1):67–73.(Raúl Herranz, Jesús Mateo)