Comparative Biochemistry and Physiology, Part B 158 (2011) 208–215

Contents lists available at ScienceDirect

Comparative Biochemistry and Physiology, Part B

journal homepage: www.elsevier.com/locate/cbpb

Molecular characterization of the MuRF in rainbow trout: Potential role in muscle degradation

Jiannan Wang a, Mohamed Salem a, Nan Qi a, P. Brett Kenney a, Caird E. Rexroad III b, Jianbo Yao a,⁎ a Laboratory of Animal Biotechnology and Genomics, Division of Animal and Nutritional Sciences, West Virginia University, Morgantown, WV 26506, USA b National Center for Cool and Cold Water Aquaculture, Kearneysville, WV 25430, USA article info abstract

Article history: Muscle growth is determined primarily by the balance between synthesis and degradation. When Received 16 July 2010 rates of protein synthesis are similar between individuals, protein degradation is critical in explaining Received in revised form 23 November 2010 differences in growth efficiency. Studies in mammals showed that muscle atrophy results from increased Accepted 23 November 2010 protein breakdown, and is associated with activation of the ubiquitin proteasome pathway, including Available online 8 December 2010 induction of the muscle-specific ubiquitin protein ligase, MuRF1. Animals lacking MuRF1 are resistant to muscle atrophy. In fish, little is known about the role of the proteasome/MuRF pathway in muscle Keywords: MuRF genes degradation. The objectives of this study were to: 1) clone and characterize MuRF genes in rainbow trout; and Ubiquitin proteasome pathway 2) determine expression of MuRF genes in association with starvation- and vitellogenesis-induced muscle Protein degradation atrophy in rainbow trout. We have identified full-length cDNA sequences for three MuRF genes (MuRF1, Muscle MuRF2, and MuRF3). These genes encode with typical MuRF structural domains, including a RING- Rainbow trout finger, a B-box and a Leucine-rich coiled-coil domain. RT-PCR analysis showed that MuRF genes are predominantly expressed in muscle and heart tissues. Real time PCR analysis revealed that expression of all MuRF genes is up-regulated during starvation and MuRF3 is up-regulated in vitellogenesis-associated muscle degradation. These results suggest that MuRF genes have an important role in fish muscle protein degradation. Further studies are warranted to assess the potential use of MuRF genes as tools to monitor fish muscle growth and degradation. © 2010 Elsevier Inc. All rights reserved.

1. Introduction Protein degradation in animal cells is a highly selective, regulated and energy-dependent process controlled by activities of proteolytic Muscle is the largest protein reservoir in fish, serving as a primary enzymes (Hershko et al., 2000). There are four known proteolytic reserve of amino acids that can be mobilized during fasting and systems involved in muscle proteolysis; the calcium-dependent, disease to provide a source of amino acids for hepatic gluconeogenesis calpain system; the lysosomal protease system (cathepsins); the and energy production (Kettelhut et al., 1988) as well as protein ubiquitin (Ub)–proteasome system; and the apoptosis protease synthesis. In aquaculture, muscle growth and fillet quality are system (caspase) (Salem et al., 2006b; Argilés et al., 2008). The important traits that impact profitability. Expression of molecular/ performance of these various systems has been widely discussed genetic markers can be used to evaluate muscle proteolysis and select (Millward, 1985; Kumamoto et al., 2000; Lecker et al., 2004) with for 1) increased muscle protein accretion during fish growth and 2) strong evidence that the ATP-dependent Ub–proteasome pathway is decreased postmortem protein degradation (less fillet softening). essential during the majority of protein degradation associated with Protein turnover is determined by rate of protein synthesis and muscle atrophy in mammals (Lecker et al., 1999; Jagoe and Goldberg, protein degradation. The rate of protein turnover is important in 2001). However, in fish, little is known about the proteolytic determining efficiency of animal growth and muscle catabolism strategies of muscle protein degradation. Some studies suggested (Young et al., 1975; Hawkins et al., 1989; Mommsen, 2004; Salem that the Ub–proteasome pathway is not involved in vitellogenesis- et al., 2006b). When rates of protein synthesis are similar between induced muscle atrophy (Salem et al., 2006a); other studies found individuals, protein degradation is critical in determining differences down-regulated expression of the proteasome pathway in fasting- in growth efficiency. Therefore, detailed knowledge of protein induced fish muscle degradation (Martin et al., 2002). A recent report degradation mechanisms in fish will benefit the aquaculture industry. showed that the polyubiquitination step of the Ub–proteasome route is regulated by starvation/refeeding status with little effect on 20 S proteasome activity (Seiliez et al., 2008). – ⁎ Corresponding author. Tel.: +1 304 293 1948; fax: +1 304 293 2232. The ubiquitin proteasome pathway of protein degradation E-mail address: [email protected] (J. Yao). involves tagging the substrate protein by covalent attachment of

1096-4959/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpb.2010.11.010 J. Wang et al. / Comparative Biochemistry and Physiology, Part B 158 (2011) 208–215 209 multiple Ub molecules, a 76-amino acid polypeptide (conjugation); (Pfam, http://www.sanger.ac.uk/Software/ Pfam/search.shtml) and and subsequent degradation of the tagged protein by the 26 S according to the work of Spencer and coworkers (Spencer et al., proteasome, resulting in short peptides of 7–9 amino acid residues. 2000a). Sequence analysis was done using the BioEdit software (http:// Ub–protein conjugation requires ATP, the Ub-activating enzyme (E1), www.mbio.ncsu.edu/BioEdit/BioEdit.html). Ub carrier proteins (E2 enzymes), and frequently, Ub–protein isopeptide ligases (E3 enzymes). The key enzyme responsible for 2.2. Fish and tissue collection targeting ubiquitination of specific substrate proteins is a Ub–protein ligase (E3) that catalyzes transfer of an activated form of Ub from a Samples used in quantitative measurement of MuRF expres- specificUb–carrier protein (E2) to a lysine residue on the substrate sion in spawning-induced muscle atrophy were collected as described (Obin et al., 1996). Recent results suggest that Atrogin-1 (muscle previously (Salem et al., 2006a). Mature fertile (diploid) and sterile atrophy F-box) and MuRF1 (muscle RING finger 1) are two Ub-protein (triploid) female rainbow trout (500 g) were collected from Flowing ligases (E3) that are key players in the regulation of ubiquitin– Springs Trout Farm (Delray, WV, USA) during the spawning season in proteasome-mediated muscle atrophy in mammals (Lecker, 2003; early October. Fish were cultured in identical raceways receiving water Cao et al., 2005). As evidence of the role of MuRF1 in this process, from a common spring at 13±3 °C. Fish were fed ad libitum (Zeiglar MuRF1−/− mice displayed a partial protection against denervation- Gold; Zeigler Bros., Gardeners, PA, USA) via demand feeders. No induced muscle wasting (Bodine et al., 2001). In addition under amino difference of food consumption was noticed between groups. Fish had acid depletion, MuRF1−/− mice had less muscle wasting maintaining access to feed when sampled. As confirmed by dissection, fertile fish normal de novo protein synthesis compared to wild-type mice were gravid with a gonado-somatic index (GSI=ovary weight/fish (Koyama et al., 2008). weight X100) of 15.8±0.3 (n=5). The GSI of sterile fish was 0.3±0.2 MuRF2 and MuRF3 were identified in yeast two-hybrid studies, as (n=5). White muscle samples (20 g) from five fish of each group were binding proteins of MuRF1 (Spencer et al., 2000b). MuRF2 and MuRF3 collected from the dorsal musculature and flash frozen in liquid were speculated to work together in muscle differentiation, via nitrogen and stored at −80 °C until RNA extraction. stabilizing and -based signaling mechanisms (Gregorio Samples used in quantitative analysis of MuRF gene expression in et al., 2005). MuRF2 was proposed to act as an adaptor between starvation-induced muscle atrophy were collected as described in a , , and titin during myofibrillogenesis. Studies previous study (Salem et al., 2005). Rainbow trout fingerlings (15– examining the function of MuRF2 showed importance for microtubule, 20 g) were brought from Bowden State Fish Hatchery, WV Division of intermediate filament and sarcomeric M-line maintenance in striated Natural Resources (Elkins, WV, USA) to the laboratory and acclima- muscle development (McElhinny et al., 2004). Mice lacking MuRF3 had tized for 30 days. Temperature (14.0 °C ±1), dissolved oxygen, pH, normal cardiac function but were susceptible to cardiac rupture after and other water quality parameters were held constant throughout acute myocardial infarction (Fielitz et al., 2007). A recent study showed the study. Fish were randomly divided into six experimental groups that MuRF1, but not MuRF2, plays important role in regulating cardiac (60 fish each). Each group was assigned to a net pen (60×60×60 cm). hypertrophy (Willis et al., 2007). In contrast, another study showed that To remove variation associated with potential differences in water mice lacking both MuRF1 and MuRF2 proteins develop cardiac and quality, all pens were kept in the same aquarium (1100 L). Three skeletal muscle hypertrophy, whereas animals lacking either MuRF1 or control groups were manually fed a commercial fish diet (Zeigler MuRF2 are normal (Witt et al., 2008). Bros.) at 1% of fish body weight, twice per day. Three experimental Muscle atrophy occurs in aging, denervation, disuse, muscular groups were subjected to a starvation regimen for 35 days. This dystrophies cystic fibrosis, AIDS, and after high dose treatment with starvation regimen was based on Tripathi et al. (Tripathi and Verma, glucocorticoids (Lecker et al., 2004; Mittal et al., 2010). In fish, muscle 2003) who showed that the maximum effect of fasting on rainbow deterioration occurs as a physiological response to the increased trout muscle protein content occurs after 35 days of starvation. At the energetic demands associated with egg growth and development and end of the experimental period, nine fish from each experimental in response to starvation (Mommsen, 2004; Salem et al., 2006b; Seiliez group were randomly sampled. Fish were euthanized and weighed, et al., 2008). A microarray gene expression analysis of atrophying and white muscle samples were collected. rainbow trout muscle showed up-regulated expression of a muscle- specific RING finger protein 3-like transcript (Salem et al., 2006a). The 2.3. Northern blot analysis primary objective of this study was to clone and characterize the MuRF gene family in rainbow trout and investigate its expression during Total RNA was isolated from muscle or heart samples using Trizol starvation, and egg growth and development leading up to spawning. reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer's protocol. Frozen tissue (0.2 g) was placed in 2 ml of Trizol reagent and 2. Materials and methods homogenized using a Polytron homogenizer. Messenger RNA was isolated from total RNA using the polyAtract mRNA isolation system 2.1. Identification of MuRF1, MuRF2 and MuRF3 cDNAs (Promega, Madison, WI, USA). Biotinylated oligo d(T) probe and streptavidin attached magnetic beads were used for this purpose. A Nucleotide and amino acid searches, against the 5-μg sample of mRNA was separated on a 1% denaturing agarose gel by GenBank database (http://www.ncbi.nlm.nih.gov/), were used to electrophoresis along with an RNA marker (Promega). Messenger RNA examine rainbow trout (Oncorhynchus mykiss) expressed sequence was subsequently transferred to a Hybond N+nylon membrane tags (ESTs) to identify MuRF-like clones. We identified three rainbow (Amersham Biosciences, Piscataway, NJ, USA). The mRNA was cross- trout cDNA clones (GenBank Accession No.: CA369980, CA342294, linked onto the membrane by UV rays, and then the membrane was pre- BX321103) that had high homology with mammalian MuRFs. Clones hybridized in DIG Easy Hyb solution (Roche Diagnostics, Indianapolis, were obtained from the United States Department of Agriculture, IN, USA) for 1 h and then hybridized in the same solution containing National Center for Cool and Cold Water Aquaculture (USDA-NCCCWA) DIG-labeled PCR probe (2 μL/mL in hybridization solution) overnight at (Kearneysville, WV, USA). All clones were completely sequenced and 68 °C. DIG-labeled PCR probes, used for Northern blot analysis, were aligned to generate full length sequences for rainbow trout MuRF1, synthesized using a DIG PCR labeling kit (Roche Diagnostics). DIG MuRF2 and MuRF3 cDNAs. CLUSTALW version 1.81 (http://www.cmbi. labeled DNA fragments of rainbow trout MuRF1 (429 bp), MuRF2 kun.nl/cgi-bin) algorithm was used for protein multiple alignments. The (570 bp) and MuRF3 (565 bp) were synthesized by PCR using the characteristic domains of three MuRF proteins were assigned by blasting corresponding cDNA clones as templates and gene-specific primers their sequences against the Protein Families Database of Alignment (Table 1). Following stringent washes (4×10 min with 2X SSC, 0.1% SDS 210 J. Wang et al. / Comparative Biochemistry and Physiology, Part B 158 (2011) 208–215

Table 1 PCR primers used in the study.

Gene name Primer sequence Method Amplicon size qPCR efficiency

MuRF1 Sense: CAAGAGCATCGAGGAGAACAG Standard RT-PCR 429 bp Antisense: TCCTCTGTCACCACATCATCA MuRF1 Sense: CTGATTAGTGGCAAGGAGCTG Real time PCR 101 bp 92.1% Antisense: GTAAGGTGCTCCATGTTCTCG MuRF2 Sense: AGGCACTAACATCTCCCACCT Standard RT-PCR 570 bp Antisense: AGCCAAGAAAAGGAGAAGACG MuRF2 Sense: TGGAGGAGTCAGAGATGGCTA Real time PCR 88 bp 97.8% Antisense: TCCAGGTGGGAGATGTTAGTG MuRF3 Sense: GATGGACTTTGGAATGGTCG Standard RT-PCR 565 bp Antisense: GCACGCAATCAATTCAAGCA MuRF3 Sense: ATGTCCATTGCAGGGACTCTA Real time PCR 143 bp 97.3% Antisense: AACTGGGGTAAGCCATTGTGT β-actin Sense: GCCGGCCGCGACCTCACAGACTAC Real time PCR 73 bp 100% Antisense: GGCCGTGGTGGTGAAGCTGTAAC at room temperature and 4×10 min with 0.1X SSC, 0.1% SDS at 68 °C), each sample, the amount of the target and internal reference genes the nylon membrane was incubated in the blocking solution for 30 min was determined from the appropriate standard curve. The amount of followed by additional incubation with a blocking solution that target gene was then divided by the amount of reference gene to contained a 1:10,000 dilution of alkaline phosphatase conjugated, obtain a normalized target value. The mean differences in expression anti-DIG antibody (Roche Diagnostics). The second incubation was for levels were determined by a t-test (SigmaStat version 3.11, Aspire 1 h at room temperature. After 2 washes in the washing buffer, the Software International, Leesburg, VA, USA). The β-actin gene was hybridized probe was detected with the chemilumisescent substrate, chosen as a control for normalization of the real-time PCR data based CSPD (Roche Diagnostics). Blocking and washing solutions were from a on our previous observation that expression of β-actin was not DIG wash and block buffer set (Roche Diagnostics). changed in spawning-induced muscle atrophy (Salem et al., 2006a).

2.4. RT-PCR 2.6. Phylogenetic analysis Total RNA from fish tissue samples was first treated with DNAse I A phylogenetic analysis of the rainbow trout MuRF genes was and then reverse transcribed to first-strand cDNA using oligo (dT) 18 performed using full MuRF protein sequences including Homo sapiens primer and Superscript III reverse transcriptase (Invitrogen). First- TRIM63 (19924163), H. sapiens ring finger protein 28 (30583585), strand cDNA from various tissues was used as a template for PCR H. sapiens TRIM-55 isoform 2 (34878821), Danio rerio hypothetical amplification of the three MuRF genes using gene-specific primers protein-LOC394070 (41055281), Tetraodon nigroviridis unnamed pro- (Table 1). The PCR reaction conditions (25-μL reaction mixture) tein product (47207992), D. rerio hypothetical protein LOC415223 included 5 min denaturation at 94 °C followed by 35 cycles of 94 °C (50344922), D.o rerio tripartite motif-containing 55 (50539776), Danio for 30 s, 59 °C for 45 s and 72 °C for 30 s, and a final extension at 72 °C rerio tripartite motif-containing 55b (90652813), Gallus gallus similar to for 10 min. Trout β-actin gene was used as internal control to verify RING finger protein 29 isoform 1 (118087028), G. gallus similar to ring RNA quality. finger protein 28 (118101656), Xenopus tropicalis TRIM 63 (118404880), X. laevis TRIM 54 (147905260), Salmo salar MuRF1-like 2.5. Quantitative real-time PCR assay (213511072), S. salar LOC394070 protein-like (213512264), D. rerio novel protein similar to TRIM54 (220941743), rainbow trout MuRF1- The mRNA expression of three MuRF genes in muscle samples (HM357611), MuRF2 (HM357612) and MuRF3 (HM357613). Recon- were measured using quantitative real-time PCR. Prior to reverse struction and analysis of phylogenetic relationships between rainbow transcription, total RNA was treated with DNase I (Ambion, Austin, TX, trout MuRF orthologues and paralogues sequences were performed USA) to eliminate genomic DNA contamination. Two μg of DNAse- using PhyML maximum likelihood based program (http://www. treated total RNA from each sample were converted to cDNA using phylogeny.fr/version2_cgi/index.cgi)(Dereeper et al., 2008). Approxi- Superscript III reverse transcriptase (Invitrogen). Negative control mate likelihood ratio test was used to construct the PhyML trees with cDNAs were prepared by reverse transcription reactions without program default settings. Values obtained from approximate likelihood adding the reverse transcriptase. ratio test are similar to standard 100 runs of bootstrapping. Real-time PCR primers (Table 1) for MuRF1, MuRF2 and MuRF3 genes and the endogenous control, β-actin gene, were designed based on the corresponding cDNA sequences (rainbow trout β-actin gene: 3. Results GenBank accession No.: AJ438158) using Primer3 software (http:// frodo. wi.mit.edu/cgi-bin/primer3/primer3_www.cgi). Quantitative 3.1. Identification and characterization of MuRF cDNA sequences PCR was performed in duplicate for each cDNA sample on a Bio-Rad iCycler iQ Real-Time PCR Detection System (Bio-Rad, Hercules, CA, By searching the rainbow trout EST database at the GenBank, USA). iQ SYBR Green Supermix (Bio-Rad) was used in a 25-μL reaction (http://www.ncbi.nlm.nih.gov/), we identified 3 rainbow trout MuRF- with cDNA derived from 0.2 μg of total RNA. Trout β-actin gene was like cDNA clones. The complete sequences of the cDNA clones were used as an internal control for normalization. Standard curves for the determined by primer walking (Core Sequencing Facility, University target and reference genes were constructed using 10-fold serial of Illinois at Urbana-Champaign, Urbana, IL, USA). The longest open dilutions of the corresponding plasmids. Melting curve analyses were reading frame of each sequence was predicted, and the amino acid done to confirm amplification of a single product. Standard curves sequences were aligned against known mammalian MuRF proteins for were run on the same plate with the samples. Threshold lines were annotation and characterization. Based on the amino acid homology, adjusted to intersect amplification lines in the linear portion of the clones CA369980, CA342294 and BX321103 were designated as amplification curve and cycles to threshold (Ct) were recorded. For rainbow trout MuRF1, MuRF2 and MuRF3, respectively. J. Wang et al. / Comparative Biochemistry and Physiology, Part B 158 (2011) 208–215 211

The complete nucleotide sequence of MuRF1 cDNA is 1791 bp in 3.2. Tissue distribution of rainbow trout MuRF genes length and contains an open reading frame encoding a protein of 359 amino acids (GenBank accession number: HM357611). The amino acid The tissue distribution patterns of MuRF1, MuRF2 and MuRF3 sequence shows 67% similarity with salmon MuRF1 (NP_001133124.1) transcripts were determined by RT-PCR analysis (Fig. 2). Rainbow and 54% with mouse MuRF1 (NP_001034137.2). The cDNA sequence for trout spleen, kidney, white muscle, liver, gill, heart, brain, skin, testis MuRF2 is 2193 bp long, and it codes for a protein of 490 amino acids and egg were selected for analysis of MuRF gene expression patterns. (HM357612). The deduced amino acid sequence is also closely related MuRF1 exhibited muscle specific expression because it was only seen to the known MuRF2 from other species as it shares 59% sequence in muscle and heart. MuRF2 is expressed primarily in muscle and identity with zebrafish (NP_001002358.1) and 66% with mouse MuRF2 heart, with lesser amounts in brain. MuRF3 mRNA is also predomi- (NP_001074750.1). The nucleotide sequence of MuRF3 consists of nantly expressed in muscle and heart, with greater amounts in heart 1785 bp, and it has an open reading frame coding for 350 amino acids (Fig. 2). (HM357613). The deduced amino acid sequence displayed high sequence homology with previously reported MuRF3 in other species, with 85% amino acid identity to zebrafish (CAX15444.1) and 53% to 3.3. Northern blot analysis mouse MuRF3 (NP_067422.1). Amino acid sequence analysis of MuRF1, MuRF2 and MuRF3 Northern blot analysis, using mRNA isolated from muscle or heart, revealed that all three MuRF proteins contain the characteristic RING- revealed a single transcript of ~1.9 kb for MuRF1, two transcripts of finger domain, B-box zinc finger domain and Leucine-rich coiled-coil ~2.0 and ~2.3 kb for MuRF2 and a single transcript of ~1.9 kb for domain (Fig. 1). Each domain contains the highly conserved sequence MuRF3 (Fig. 3). The size of each transcript is consistent with the motif that is essential for MuRF activity. The RING-finger domain is expected size of the corresponding cDNA (MuRF1, 1791 bp; MuRF2, located at the N-terminus, which is functional in protein–protein 2193 bp; MuRF3, 1785 bp). The additional band of 2.0 kb detected by interaction. Trout MuRF1 and MuRF2 share 49% protein sequence the MuRF2 probe may represent an alternative splice variant of MuRF2 identity (45% nucleotide identity), MuRF1 and MuRF3 share 67% or a transcript from a paralogous gene due to partially tetraploid trout protein sequence identity (64% nucleotide identity), and MuRF2 and genome (Allendorf and Thorgaard, 1984) or alternative use of MuRF3 share 57% protein sequence identity (47% nucleotide identity). polyadenylation signals.

Fig. 1. Multiple alignment of the deduced amino acid sequences of rainbow trout MuRF1, MuRF2 and MuRF3. Identical residues are shaded in black and conserved substitutions in grey. The MuRF functional domains are indicated: RING-finger domain (red box), B-box zinc finger domain (green box) and Leucine-rich coiled-coil domain (purple box). 212 J. Wang et al. / Comparative Biochemistry and Physiology, Part B 158 (2011) 208–215

3.5. Real time PCR quantification of MuRF gene expression in spawning-induced and starvation-induced muscle atrophies

To investigate mRNA expression of MuRF genes in spawning- induced muscle atrophy in rainbow trout females, fertile (2 N) and sterile (3 N) fish muscle samples were collected at the spawning season and subjected to quantitative real time PCR analysis. As shown in Fig. 5, numerical increase of expression for all three MuRF genes was observed in 2 N females, however, only MuRF3 expression was increased significantly in 2 N females (9.6-fold, Pb0.05). Expression of all three MuRF transcripts was up-regulated in starved fish muscle compared to normal-fed fish muscle. Abundance of mRNA was increased by 6.5-fold for MuRF1, 2.1-fold for MuRF2 and 3.5-fold for MuRF3 (Pb0.01) in fish that had food withdrawn for 35 days compared to fed controls (Fig. 6).

Fig. 2. RT-PCR analysis of tissue distribution of the rainbow trout MuRF genes. cDNA from multiple tissues were amplified by PCR using gene specific primers. Four sample sets were analyzed and data from one set is presented. All three MuRF genes were 4. Discussion abundantly expressed in muscle tissue. MuRF1 appears to be muscle specific as it was only seen in muscle and heart. MuRF2 is predominantly expressed in muscle and heart, In this study, we identified and characterized three full-length with lesser amounts in brain. MuRF3 mRNA is also predominantly expressed in muscle cDNA sequences for MuRF genes in rainbow trout. The cDNA and heart, with greater amounts in heart. sequences of these three genes contain complete open reading frames encoding MuRF proteins with characteristic N-terminal RING-finger, B-box zinc finger and Leucine-rich coiled-coil domains (RBCC). The 3.4. Phylogenetic analysis RING-finger domain is an unusual type of Cys-His zinc-binding motif, with a specialized type of Zn-finger domain containing 40 to 60 Fig. 4 presents the proposed phylogenetic tree of the MuRF genes. residues that bind two atoms of zinc (Freemont, 1993; Borden and This tree includes the deduced rainbow trout MuRF proteins; Homo Freemont, 1996; Saurin et al., 1996). The RING-finger domain is sapiens TRIM63, ring finger protein 28, TRIM-55 isoform 2; Gallus probably involved in mediating protein–protein interactions, and is gallus protein similar to RING finger protein 29 isoform 1 and protein identified in proteins with a wide range of functions such as signal similar to ring finger protein 28; Xenopus tropicalis TRIM 63; X. laevis transduction, gene transcription, differentiation, ubiquitination, mor- TRIM 54; Danio rerio hypothetical protein LOC394070, hypothetical phogenesis and microtubule stabilization (Spencer et al., 2000b). The protein LOC415223, TRIM55, TRIM55b and a novel protein similar to B-box zinc finger domain is also a protein–protein interaction domain TRIM54; Salmo salar MuRF1-like and LOC394070 protein-like; and (Borden, 1998). The leucine-rich coiled-coil domain mediates asso- pufferfish (Tetraodon nigroviridis) unnamed protein. All proteins used ciation with microtubules (Spencer et al., 2000b). Depending on the to construct the phylogenetic tree were found to show similarity to protein–protein interactions of the RING-finger domain, MuRF1 is a trout MuRF proteins by NCBI blast search. The topology of the bona fide RING finger-dependent, ubiquitin ligase that catalyzes a phylogenetic tree shows the evolutionary relationships between rate-limiting step in troponin I degradation. Troponin I is a MuRF1 rainbow trout MuRF orthologues and paralogues. The tree shows interaction partner, and MuRF1-troponin I interactions can be two main branches: 1) fish MuRF1/3 (Fig. 4 cluster A) and 2) fish detected in cardiomyocytes (Kedar et al., 2004). MuRF2 in addition to MuRF1-3 of other species (Fig. 4 clusters B1 & The MuRF genes are predominantly expressed in muscle and heart B2). The later branch is subdivided into the MuRF2 like cluster (B1) tissues (Fig. 2). Northern blot analyses showed a single transcript and MuRF1 like cluster (B2). Homologous sequences that differ corresponding to the cDNA size for MuRF1 and MuRF3. In addition to because they are found in different species appear to be orthologues. the transcript corresponding to the cDNA size for MuRF2, another The tree suggests a paralogous mode of evolution of trout MuRF1-3. transcript of 2.0 kb was detected by the MuRF2 probe. This transcript Fish MuRF1 and MuRF3 are closer to each other than to MuRF2. may represent an alternative splice variant of MuRF2 or a paralogous sequence. However, we were not able to find any splice variants of MuRF2 by BLAST search of the well characterized rainbow trout EST database (258,973 Sanger-based and ~1.3 million pyrosequencing ESTs, available through NCBI). As a member of the family Salmonidae, rainbow trout has descended from a tetraploid ancestor (Allendorf and Thorgaard, 1984). About 50% of protein-coding loci examined in salmonid species show duplicate gene expression (Bailey et al., 1978). Therefore, it is more likely that the two MuRF2 transcripts are produced from paralogous sequences. It would be hard to determine the origin of the additional transcript without a characterized rainbow trout genome sequence. The phylogenetic tree topology (Fig. 4) suggests the existence of a MuRF-like ancestral gene in all vertebrate superfamily members. Rainbow trout MuRF2 was found clustered with homologous protein from fish, amphibian, avian and mammalian species (Fig. 4). However, MuRF1 and MuRF3 were found on a separate branch of the tree with Fig. 3. Northern blot analysis of rainbow trout MuRF transcripts. Messenger RNA only fish members suggesting a significant paralogous divergence isolated from muscle/heart was separated on agarose gel and transferred to a nylon from other vertebrate MuRF-like proteins. The phylogenetic tree membrane before hybridizing to DIG labeled PCR probes for MuRF1, MuRF2 and MuRF3. fi fi A single transcript of 1.9 kb for MuRF1, two transcripts of 2.0 kb and 2.3 kb for MuRF2 identi es MuRF homologues from sh species (Danio rerio, Salmo salar and a single transcript of 1.9 kb for MuRF3 were detected. and Tetraodon nigroviridis) that warrant further studies. J. Wang et al. / Comparative Biochemistry and Physiology, Part B 158 (2011) 208–215 213

Fig. 4. A phylogram of the evolutionary relationships between rainbow trout MuRF orthologues and paralogues generated by using the PhyML maximum likelihood analysis. The tree shows two main branches: 1) fish MuRF1/3 (A) and 2) fish MuRF2 in addition to MuRF1-3 of other species (B1&2). The later is subdivided into the MuRF2 like cluster (B1) and MuRF1 like cluster (B2). Homologous sequences that differ because they are found in different species appear to be orthologues. The tree suggests a paralogous mode of evolution of trout MuRF1-3. Fish MuRF1 and MuRF3 are closer to each other than to MuRF2.

MuRF genes are described as ubiquitin E3 ligases, as such they are increased energetic demands placed on muscle during fish spawning highly likely to be involved in proteasome protein degradation and starvation (Mommsen, 2004; Salem et al., 2005; Rescan et al., (Lecker, 2003). However, Salem and coworkers showed that expres- 2007; Salem et al., 2007). These responses represent suitable models sion of the proteasome pathway genes is not associated with to study mechanisms of muscle degradation/regeneration in fish. vitellogenesis-induced atrophy in rainbow trout (Salem et al., During the spawning phase, white muscle acts as the major energy 2006a). Additionally, Martin and coworkers reported down-regulated source once lipids are depleted (Love, 1979; Mommsen et al., 1981; expression of proteasome genes in fasting-induced rainbow trout Ando et al., 1998; Mommsen, 2004). Significant decreases in body muscle degradation (Martin et al., 2001; Martin et al., 2002). mass and separable muscle were reported in sexually maturing Conversely, Insulin and IGF-I independently reduced the abundance rainbow trout and salmon. This process also involves a decline in net of the ubiquitin ligase mRNA in primary cultures of rainbow trout protein concentration (Martin et al., 1993 ; Salem et al., 2006b). myocytes (Cleveland and Weber, 2009, 2009). Therefore, MuRF genes Muscle acts as a major energy source to support vitellogenesis- may be involved in different mechanisms regulating muscle growth induced increases in energy demand. Our data showed significantly and degradation in fish that may not include the proteasome pathway. increased expression of MuRF3 and numerically increased expression of Further studies are needed to explain the role of MuRF genes in fish MuRF1 and MuRF2 in spawning 2 N females. These results are consistent muscle degradation. with our previous microarray gene expression analysis of atrophying Muscle atrophy refers to wasting of muscle tissue resulting from rainbow trout muscle which showed up-regulated expression of a conditions such as starvation or systemic diseases (cancer, diabetes, muscle-specificRINGfinger protein 3-like transcript (Salem et al., sepsis). Muscle atrophy also occurs as a physiological response to the 2006a). The increased MuRF gene expression is consistent with elevated

10 14 * * 12 8

10 6 8 * 6 4

4 * 2 Relative gene expression Relative gene expression 2

0 0

MuRF-1 Normal MuRF-2 Normal MuRF-3 Normal MuRF-1 Normal MuRF-2 Normal MuRF-3 Normal MuRF-1 Atrophying MuRF-2 Atrophying MuRF-3 Atrophying MuRF-1 Atrophying MuRF-2 Atrophying MuRF-3 Atrophying

Fig. 5. Effect of spawning-induced muscle atrophy on the expression of MuRF genes. Fig. 6. Effect of starvation-induced muscle atrophy on the expression of MuRF genes. Expression of the MuRF genes was analyzed by real-time PCR. Quantity of each MuRF Expression of the MuRF genes was analyzed by real-time PCR. Quantity of each MuRF mRNA was normalized to β-actin. The means of the normalized gene expression values mRNA was normalized to the trout β-actin gene. The means of the normalized gene were calculated and expressed as relative fold changes (n=5, mean±SEM). * Indicates expression values were calculated and expressed as relative fold changes (n=15, mean± significant difference (Pb0.05). SEM). * Indicates significant difference (Pb0.01). 214 J. Wang et al. / Comparative Biochemistry and Physiology, Part B 158 (2011) 208–215 protein degradation needed to supply carbon skeletons for intermediary Argilés, J.M., Lopez-Soriano, F.J., Busquets, S., 2008. Apoptosis signalling is essential and precedes protein degradation in wasting skeletal muscle during catabolic metabolism and thus ATP synthesis, suggesting a role of MuRF genes in conditions. Int J Biochem Cell Biol 40, 1674–1678. spawning-induced muscle atrophy and protein mobilization. The Bailey, G.S., Poulter, R.T., Stockwell, P.A., 1978. Gene duplication in tetraploid fish: discrepancy of MuRF gene expression (MuRF1/MuRF2 vs. MuRF3)in model for gene silencing at unlinked duplicated loci. Proc Natl Acad Sci USA 75, fi 5575–5579. spawning sh may indicate difference in relative importance and Bodine, S.C., Latres, E., Baumhueter, S., Lai, V.K., Nunez, L., Clarke, B.A., Poueymirou, W.T., functionality of the three MuRF isoforms. Panaro, F.J., Na, E., Dharmarajan, K., Pan, Z.Q., Valenzuela, D.M., DeChiara, T.M., Stitt, Starvation triggered up-regulated expression of all three MuRF T.N., Yancopoulos, G.D., Glass, D.J., 2001. Identification of ubiquitin ligases required – genes (Fig. 6). These data are consistent with recent studies that for skeletal muscle atrophy. Science 294, 1704 1708. Borden, K.L., 1998. RING fingers and B-boxes: zinc-binding protein-protein interaction showed dramatic increase in abundance of the muscle-specific domains. Biochem Cell Biol 76, 351–358. ubiquitin ligase atrogin-1 (MAFbx) mRNA in starved rainbow trout Borden, K.L., Freemont, P.S., 1996. The RING finger domain: a recent example of a – and salmon (Cleveland and Weber, 2009, 2009; Bower et al., 2010 ) sequence-structure family. Curr Opin Struct Biol 6, 395 401. Bower, N.I., Taylor, R.G., Johnston, I.A., 2009. Phasing of muscle gene expression with and inhibition of MuRF1 and MAFbx in responses to re-feeding (Bower fasting-induced recovery growth in Atlantic salmon. Front Zool 6, 18. doi:10.1186/ et al., 2009). White muscle responds to starvation with rapidly 1742-9994-6-18. declining protein synthesis rates (Pocrnjic et al., 1983; Fauconneau Bower, N.I., de la Serrana, D.G., Johnston, I.A., 2010. Characterisation and differential regulation of MAFbx/Atrogin-1 alpha and beta transcripts in skeletal muscle and Arnal, 1985; Houlihan et al., 1988). Sensitivity of white muscle of Atlantic salmon (Salmo salar). Biochem Biophys Res Commun 396, protein synthesis during starvation is important to overall growth; 265–271. any decrease in white muscle protein synthesis will have a substantial Cao, P.R., Kim, H.J., Lecker, S.H., 2005. Ubiquitin-protein ligases in muscle wasting. Int J Biochem Cell Biol 37, 2088–2097. effect on growth and metabolism (Houlihan et al., 1988). The Cleveland, B.M., Weber, G.M., 2009. Effects of insulin-like growth factor-I, insulin, and observed decrease in white muscle protein synthesis during starva- leucine on protein turnover and ubiquitin ligase expression in rainbow trout tion appears to be highly protein specific, with a 90% decrease in primary myocytes. Am J Physiol Regul Integr Comp Physiol 298, R341–R350. fi Cleveland, B.M., Weber, G.M., Blemings, K.P., Silverstein, J.T., 2009. Insulin-like growth myo brillar protein synthesis observed during starvation (Lowery factor-I and genetic effects on indices of protein degradation in response to feed and Somero, 1990). Considerably less is understood about the effects deprivation in rainbow trout (Oncorhynchus mykiss). Am J Physiol Regul Integr of starvation on protein degradation. Martin et al. demonstrated that Comp Physiol 297, R1332–R1342. activity of the proteasome pathway, a major vehicle of protein Dereeper, A., Guignon, V., Blanc, G., Audic, S., Buffet, S., Chevenet, F., Dufayard, J.F., Guindon, S., Lefort, V., Lescot, M., Claverie, J.M., Gascuel, O., 2008. Phylogeny.fr: robust degradation, is reduced after 2 weeks starvation in rainbow trout. In phylogenetic analysis for the non-specialist. Nucleic Acids Res 36, W465–W469. contrast, lysosomal and the calpain enzyme activities appears to Dobly, A., Martin, S.A., Blaney, S.C., Houlihan, D.F., 2004. Protein growth rate in rainbow increase during starvation (Martin et al., 2000; Salem et al., 2005). trout (Oncorhynchus mykiss) is negatively correlated to liver 20 S proteasome activity. Comp Biochem Physiol A Mol Integr Physiol 137, 75–85. Additionally, Dobly and coworkers reported a negative correlation Fauconneau, B., Arnal, M., 1985. In vivo protein synthesis in different tissues and the between trout liver, but not white muscle, 20 S proteasome activity whole body of rainbow trout (Salmo gairdnerii R.). Influence of environmental and specific growth rate (Dobly et al., 2004). Feed deprivation caused temperature. Comp Biochem Physiol A Comp Physiol 82, 179–187. Fielitz, J., van Rooij, E., Spencer, J.A., Shelton, J.M., Latif, S., van der Nagel, R., remarkable increase in abundance of MAFbx mRNA, without a Bezprozvannaya, S., de Windt, L., Richardson, J.A., Bassel-Duby, R., Olson, E.N., corresponding change in the proteasome transcripts (Cleveland and 2007. Loss of muscle-specific RING-finger 3 predisposes the heart to cardiac Weber, 2009, 2009). Together, these data suggest a significant role of rupture after myocardial infarction. Proc Natl Acad Sci USA 104, 4377–4382. fi fi Freemont, P.S., 1993. The RING nger. A novel protein sequence motif related to the the MuRF ubiquitin ligases in sh muscle degradation. However, how zinc finger. Ann NY Acad Sci 684, 174–192. the MuRF genes work within the proteolytic pathways is still not Gregorio, C.C., Perry, C.N., McElhinny, A.S., 2005. Functional properties of the titin/ understood, and a larger scale, follow-up investigation is warranted. connectin-associated proteins, the muscle-specific RING finger proteins (MURFs), fi in striated muscle. J Muscle Res Cell Motil 26, 389–400. In summary, we identi ed cDNA sequences for rainbow trout Hawkins, A.J.S., Widdows, J., Bayne, B.L., 1989. The relevance of whole body protein MuRF1, MuRF2 and MuRF3 genes. These genes are predominantly metabolism to measured costs of maintenance and growth in Mytilus edulis. Physiol expressed in skeletal (white) and cardiac muscle, and encode proteins Zool 62, 745–763. with characteristic MuRF structural domains. Expression of these Hershko, A., Ciechanover, A., Varshavsky, A., 2000. Basic Medical Research Award. The ubiquitin system. Nat Med 6, 1073–1081. genes in rainbow trout was up-regulated during starvation- and Houlihan, D.F., Agnisola, C., Lyndon, A.R., Gray, C., Hamilton, N.M., 1988. Protein spawning-induced muscle atrophy, suggesting important roles in synthesis in a fish heart: responses to increased power output. J Exp Biol 137, – muscle protein degradation. Future studies are necessary to 1) 565 587. Jagoe, R.T., Goldberg, A.L., 2001. What do we really know about the ubiquitin-proteasome determine the roles of MuRF genes in regulation of muscle pathway in muscle atrophy? Curr Opin Clin Nutr Metabol Care 4, 183–190. degradation in fish and 2) evaluate the potential use of MuRF genes Kedar, V., McDonough, H., Arya, R., Li, H.H., Rockman, H.A., Patterson, C., 2004. Muscle- as genetic markers for muscle growth and fillet firmness in trout, a specific RING finger 1 is a bona fide ubiquitin ligase that degrades cardiac troponin – fi I. Proc Natl Acad Sci USA 101, 18135 18140. signi cant teleost, aquatic food species. In addition, these markers Kettelhut, I.C., Wing, S.S., Goldberg, A.L., 1988. Endocrine regulation of protein may be used for conservation management of wild populations breakdown in skeletal muscle. Diabetes Metab Rev 4, 751–772. (Vignal et al., 2002). Koyama, S., Hata, S., Witt, C.C., Ono, Y., Lerche, S., Ojima, K., Chiba, T., Doi, N., Kitamura, F., Tanaka, K., Abe, K., Witt, S.H., Rybin, V., Gasch, A., Franz, T., Labeit, S., Sorimachi, H., 2008. Muscle RING-finger protein-1 (MuRF1) as a connector of muscle energy metabolism and protein synthesis. J Mol Biol 376, 1224–1236. Acknowledgements Kumamoto, T., Fujimoto, S., Ito, T., Horinouchi, H., Ueyama, H., Tsuda, T., 2000. Proteasome expression in the skeletal muscles of patients with muscular This project was supported by the National Research Initiative dystrophy. Acta Neuropathol 100, 595–602. Competitive Grant No. 2007-35205-17914 from the USDA Coopera- Lecker, S.H., 2003. Ubiquitin-protein ligases in muscle wasting: multiple parallel pathways? Curr Opin Clin Nutr Metabol Care 6, 271–275. tive State Research, Education, and Extension Service; and USDA-ARS Lecker, S.H., Solomon, V., Mitch, W.E., Goldberg, A.L., 1999. Muscle protein breakdown Cooperative Agreement No. 58-1930-5-537. It is published with the and the critical role of the ubiquitin-proteasome pathway in normal and disease – approval of the Director of the West Virginia Agriculture and Forestry states. J Nutr 129, 227S 237S. fi Lecker, S.H., Jagoe, R.T., Gilbert, A., Gomes, M., Baracos, V., Bailey, J., Price, S.R., Mitch, W.E., Experiment Station as scienti c paper No. 3086. Goldberg, A.L., 2004. Multiple types of skeletal muscle atrophy involve a common program of changes in gene expression. FASEB J 18, 39–51. Love, R.M., 1979. The post-mortem pH of cod and haddock muscle and its seasonal References variation. J Sci Food Agric 30, 433–438. Lowery, M.S., Somero, G.N., 1990. Starvation effects on protein synthesis in red and Allendorf, F.W., Thorgaard, G.H., 1984. In: BJ, T. (Ed.), Tetraploidy and the evolution of white muscle of the barred sand bass, Paralabras nebulifer. Physiol Zool 63, salmonid fishes. Evolutionary Genetics of Fishes Plenum Press, New York, pp. 1–53. 630–648. Ando, K., Ito, Y., Kumada, M., Fujita, T., 1998. Oxidative stress increases adrenomedullin Martin, N.B., Houlihan, D.F., Talbot, C., Palmer, R.M., 1993. Protein metabolism during mRNA levels in cultured rat vascular smooth muscle cells. Hypertens Res 21, sexual maturation in female Atlantic salmon (Salmo salar L). Fish Physiol Biochem 187–191. 12, 131–141. J. Wang et al. / Comparative Biochemistry and Physiology, Part B 158 (2011) 208–215 215

Martin, A.C., del Pozo, J.C., Iglesias, J., Rubio, V., Solano, R., de La Pena, A., Leyva, A., Paz- in muscle wasting during starvation. Comp Biochem Physiol B Biochem Mol Biol Ares, J., 2000. Influence of cytokinins on the expression of phosphate starvation 140, 63–71. responsive genes in Arabidopsis. Plant J 24, 559–567. Salem, M., Kenney, P.B., Rexroad 3rd, C.E., Yao, J., 2006a. Microarray gene expression Martin, S.A.M., Cash, P., Blaney, S., Houlihan, D.F., 2001. Proteome analysis of rainbow analysis in atrophying rainbow trout muscle: a unique nonmammalian muscle trout (Oncorhynchus mykiss) liver proteins during short term starvation. Fish degradation model. Physiol Genomics 28, 33–45. Physiol Biochem 24, 259–270. Salem, M., Kenney, P.B., Rexroad, C.E., Yao, J., 2006b. Molecular characterization of Martin, S.A., Blaney, S., Bowman, A.S., Houlihan, D.F., 2002. Ubiquitin-proteasome- muscle atrophy and proteolysis associated with spawning in rainbow trout. Comp dependent proteolysis in rainbow trout (Oncorhynchus mykiss): effect of food Biochem Physiol D Genomics Proteomics 1, 227–237. deprivation. Pflugers Arch 445, 257–266. Salem, M., Silverstein, J., Rexroad 3rd, C.E., Yao, J., 2007. Effect of starvation on global McElhinny, A.S., Perry, C.N., Witt, C.C., Labeit, S., Gregorio, C.C., 2004. Muscle-specific gene expression and proteolysis in rainbow trout (Oncorhynchus mykiss). BMC RING finger-2 (MURF2) is important for microtubule, intermediate filament and Genomics 8, 328. sarcomeric M-line maintenance in striated muscle development. J Cell Sci 117, Saurin, A.J., Borden, K.L., Boddy, M.N., Freemont, P.S., 1996. Does this have a familiar 3175–3188. RING? Trends Biochem Sci 21, 208–214. Millward, D.J., 1985. The physiological regulation of proteolysis in muscle. Biochem Soc Seiliez, I., Panserat, S., Skiba-Cassy, S., Fricot, A., Vachot, C., Kaushik, S., Tesseraud, S., Trans 13, 1023–1026. 2008. Feeding status regulates the polyubiquitination step of the ubiquitin- Mittal, A., Bhatnagar, S., Kumar, A., Lach-Trifilieff, E., Wauters, S., Li, H., Makonchuk, D.Y., proteasome-dependent proteolysis in rainbow trout (Oncorhynchus mykiss) Glass, D.J., 2010. The TWEAK-Fn14 system is a critical regulator of denervation- muscle. J Nutr 138, 487–491. induced skeletal muscle atrophy in mice. J Cell Biol 188, 833–849. Spencer, J.A., Eliazer, S., Ilaria Jr., R.L., Richardson, J.A., Olson, E.N., 2000a. Regulation of Mommsen, T.P., 2004. Salmon spawning migration and muscle protein metabolism: the microtubule dynamics and myogenic differentiation by MURF, a striated muscle August Krogh principle at work. Comp Biochem Physiol B Biochem Mol Biol 139, RING-finger protein. J Cell Biol 150, 771–784. 383–400. Spencer, J.A., Eliazer, S., Ilaria, R.L., Richardson, J.A., Olson, E.N., 2000b. Regulation of Mommsen, T.P., Ballantyne, J., Macdonald, D., Gosline, J., Hochachka, P.W., 1981. microtubule dynamics and myogenic differentiation by MURF, a striated muscle Analogues of red and white muscle in squid mantle. Proc Natl Acad Sci USA 78, RING-finger protein. J Cell Biol 150, 771–784. 3274–3278. Tripathi, G., Verma, P., 2003. Starvation-induced impairment of metabolism in a Obin, M.S., Jahngen-Hodge, J., Nowell, T., Taylor, A., 1996. Ubiquitinylation and freshwater catfish. Z Naturforsch C 58, 446–451. ubiquitin-dependent proteolysis in vertebrate photoreceptors (rod outer seg- Vignal, A., Milan, D., SanCristobal, M., Eggen, A., 2002. A review on SNP and other types ments). Evidence for ubiquitinylation of Gt and rhodopsin. J Biol Chem 271, of molecular markers and their use in animal genetics. Genet Sel Evol 34, 275–305. 14473–14484. Willis, M.S., Ike, C., Li, L., Wang, D.Z., Glass, D.J., Patterson, C., 2007. Muscle ring finger 1, Pocrnjic, Z., Mathews, R.W., Rappaport, S., Haschemeyer, A.E., 1983. Quantitative but not muscle ring finger 2, regulates cardiac hypertrophy in vivo. Circ Res 100, protein synthetic rates in various tissues of a temperate fish in vivo by the method 456–459. of phenylalanine swamping. Comp Biochem Physiol B 74, 735–738. Witt, C.C., Witt, S.H., Lerche, S., Labeit, D., Back, W., Labeit, S., 2008. Cooperative control Rescan, P.Y., Montfort, J., Rallière, C., Le Cam, A., Esquerré, D., Hugot, K., 2007. of striated muscle mass and metabolism by MuRF1 and MuRF2. EMBO J 27, Dynamic gene expression in fish muscle during recovery growth induced by a 350–360. fasting-refeeding schedule. BMC Genomics 8, 438. doi:10.1186/1471-2164-8-438. Young, V.R., Steffee, W.P., Pencharz, P.B., Winterer, J.C., Scrimshaw, N.S., 1975. Total Salem, M., Nath, J., Rexroad, C.E., Killefer, J., Yao, J., 2005. Identification and molecular human body protein synthesis in relation to protein requirements at various ages. characterization of the rainbow trout calpains (Capn1 and Capn2): their expression Nature 253, 92–194.