| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Biology, New Mexico State University, Las Cruces, New Mexico, USA
1Correspondence: Department of Biology, Foster Hall, New Mexico State University, Las Cruces, NM 88003, USA. E-mail: gunguez{at}nmsu.edu
ABSTRACT
Electrocytes, the current-producing cells of electric organs (EOs) in electric fish, are unique in that they derive from striated muscle and they possess biochemical characteristics of both muscle and non-muscle cells. In the freshwater teleost Sternopygus macrurus, electrocytes are multinucleated cells that do not contract yet retain expression of some proteins common to skeletal muscle cells. Given the role that transcriptional regulation plays in the activation of the myogenic program in vertebrates, we examined the expression patterns of several genes associated with multiple functions of skeletal muscle in mature electrocytes of S. macrurus. Our expression analyses detected transcripts for 
-actin, -acetylcholine (ACh) receptor (-AChR), desmin, muscle creatine kinase (MCK), myosin heavy chain (MHC) isoforms, titin, tropomyosin, and troponin-T genes in the EO. However, immunolabeling studies revealed that electrocytes do not contain MCK, MHCs, or tropomyosin or troponin-T proteins. These results underscore the contribution of gene regulatory mechanisms in the maintenance of the muscle-like phenotype of EO that may be transcriptional-independent. We also report the classification and frequency of distinct transcripts from a random selection of 420 clones from an EO cDNA library. This is the first characterization of expressed genes in an EO, and it is an important step toward identifying mechanisms that affect different muscle protein systems for the evolution of highly specialized noncontractile tissues. Evidence of post-transcriptional regulation in the maintenance of a partial muscle phenotype by electrogenic cells of S. macrurus.—Cuellar, H., Jung, K. A., and Unguez, G. A. Evidence of post-transcriptional regulation in the maintenance of a partial muscle phenotype by electrogenic cells of S. macrurus.
Key Words: electric fish • electric organ • muscle proteins • muscle gene regulation
ELECTRIC ORGANS (EOS) are a unique evolutionary adaptation of striated muscle specialized for the production of an electric field outside the body (1)
. EOs are found only in fish and they have evolved independently in at least six different groups (1
, 2)
. The current-producing cells of the myogenically derived EOs, i.e., electrocytes, are not contractile and are activated by a distinct population of electromotor neurons, which themselves are driven by medullary neurons (1
, 2)
. The electrogenic discharge by EOs plays an important role in behaviors like predation, defense, navigation, and social communication (1
, 2)
. Hence, the study of EOs and the other components of the electrogenic circuit have led to the identification of mechanisms underlying the formation of sensory and motor specializations (3
4
5
6
7)
, hormonal modulation of electrical signals (8
9
10)
, neuroethology of social communication (11
12
13)
, and physiology of neuronal and synaptic plasticity (14
, 15)
. EOs have also been tissues of choice for biochemical identification and isolation of various molecular components that make up postsynaptic junctions (16
17
18
19)
.
We have been studying EOs because they represent the most phylogenetically widespread occurrence of an extreme modification of the skeletal muscle phenotype in vertebrates. In all but one family of electric fish, EOs derive from a variety of skeletal musculature, including tail, axial, pectoral, and oculomotor muscles (1
, 2)
. In each case, mesodermal cells differentiate into myoblast-like cells that subsequently form multinucleated myotubes. In most fish, conversion of these myotubes into mature electrocytes results from the disintegration of myofilaments in parallel with a down-regulation of many sarcomeric proteins, a replacement of myofibrils with amorphous material, and extreme changes in cell size and shape (1
, 2)
. Fully differentiated electrocytes, however, do not down-regulate the muscle program completely. For instance, electrocytes in the gymnotiform Sternopygus macrurus continue to express proteins characteristic of skeletal muscle cells such as desmin, dystrophin, sarcomeric actin, -actinin, and nicotinic ACh receptors (-AChRs) (20
, 21)
. Further, mature electrocytes in S. macrurus express keratin, a protein not found in muscle fibers (21)
. The presence of keratin in electrocytes demonstrates that unique transcriptional regulatory pathways control the electrocyte phenotype. Identifying the molecular processes involved in maintaining a partial muscle program has significant implications in understanding the plasticity of striated muscle as it relates to tissue homeostasis and evolution of novel cell types.
In vertebrates, skeletal muscle development and the expression of many genes coding for skeletal muscle proteins are under transcriptional regulation (22
23
24)
. Some of these genes include those coding for desmin (25)
, myosin heavy chain (MHC) (26)
, -actin (27)
, AChRs (28)
, and muscle creatine kinase (MCK) (29)
. Given that activation of the vertebrate muscle program is under transcriptional control, and that in fully mature electrocytes some muscle genes are down-regulated while others continue to be expressed (20
21)
, our aim was to characterize different skeletal muscle genes in mature electrocytes of S. macrurus. We hypothesized that the EO phenotype might be coupled to changes in the expression of select muscle genes at the transcription level.
Our expression analyses detected transcripts for -actin, -AChR, desmin, muscle creatine kinase (MCK), myosin heavy chain (MHC) isoforms, and titin genes in the EO. However, immunolabeling studies revealed that electrocytes do not contain MCK, MHCs, or tropomyosin or troponin-T proteins. These results underscore the contribution of gene regulatory mechanisms that may be transcriptionally independent in the maintenance of the muscle-like phenotype of EO. We also report the classification and frequency of distinct transcripts from a random selection of 420 clones from an EO cDNA library from S. macrurus. Characterization of the transcripts identified reflect the common lineage of EO with skeletal muscle but are also consistent with the EO phenotype not being a product of a simple down-regulation of select muscle genes. This is the first characterization of expressed genes in an EO, and it is an important step toward identifying mechanisms that affect different muscle protein systems for the evolution of highly specialized noncontractile tissues.
MATERIALS AND METHODS
Animals
S. macrurus is a freshwater species of knife fish native to South America and was obtained commercially from Segrest Farms (Gibsonton, FL, USA). Adult fish, 20–35 cm in length, were housed individually in 56 to 75 liter aerated aquaria maintained at 25–28°C and were fed three times weekly.
Tissue isolation and preparation
Fish were anesthetized using 2-phenoxyethanol (1:1500 in tank water; Sigma, St. Louis, MO, USA), the distal third of the tail was amputated, and the ventral skeletal muscle and EO were excised under a dissecting microscope. Tissues were immediately immersed in RNAlaterTM (Ambion, Austin, TX, USA) and stored at –80°C until RNA extraction or quick-frozen in liquid nitrogen for Western blot analyses. In a second group of fish, the distal tail was amputated and frozen in isopentane cooled in liquid nitrogen for histological processing. Immediately after tail amputation, fish were returned to their tanks and monitored until they recovered fully from anesthesia. All procedures used in this study followed the American Physiological Society Animal Care Guidelines and were approved by the Animal Use Committee at New Mexico State University.
Western blots
Total protein was isolated from skeletal muscle (113 mg) and EO (160 mg) following the protocol of Talmadge et al. (30)
. Protein concentrations were determined using the Bradford assay with 2.0 mg/mL of BSA (Sigma) as the standard. Proteins (0.5–10 µg/lane) were separated by SDS-PAGE on 4–15% Tris-HCl gels (Bio-Rad, Hercules, CA, USA) and transferred to a polyvinylidene difluoride (PVDF) membrane using a Trans-Blot Semi-Dry Electrophoretic Transfer Cell (Bio-Rad). PVDF membranes were briefly submersed in 100% methanol and washed in ddH2O for 5 min. Membranes were washed with 0.1 M PBS, 0.1% Tween-20 (PBST, pH 7.4) twice for 5 min and incubated in blocking solution (3% nonfat dried milk, PBST) at room temperature (RT) for 2 h with shaking. Membranes were washed twice for 5 min and incubated in monoclonal antibodies against sarcomeric MHC (MF20 1:5000), slow MHC (N2.261, 1:1000), fast MHC (F59, 1:1000), -actin (JLA20, 1:100), desmin (D76, 1:50), -AChR (mAb35, 1:100), tropomyosin (CH1, 1:200), MCK (ab8364, 1:500), and troponin-T (T6277, 1:200) at RT for 2 h with shaking. Antibodies MF20, N2.261, JLA20, D76, mAb35, and CH1 were obtained from the Developmental Studies Hybridoma Bank (Iowa City, IA, USA). Antibodies against MCK and troponin-T were purchased from Abcam (Cambridge, MA, USA) and Sigma, respectively, and antibody (Ab) F59 was a gift from Dr. Frank Stockdale (Stanford University, Palo Alto, CA, USA). Membranes were washed twice for 5 min with PBST and incubated in an anti-mouse HRP-conjugated secondary Ab (1:8000; Bio-Rad) for 2 h with shaking. To detect the Ab-antigen complex, membranes were incubated in the Opti-4CN substrate (Bio-Rad) per manufacturer’s instructions. Membranes were stored at RT in the dark, and final images were scanned on an HP Scanjet 5470c (Hewlett Packard, Palo Alto, CA, USA) and rendered using Adobe Photoshop (version 7.0; Adobe Systems, Inc., San Jose, CA, USA).
Immunofluorescence
Transverse and longitudinal cryosections (20–30 µm thick) from adult tails were mounted on glass slides and air-dried prior to rehydration in 0.1 M PBS (pH 7.4) for 5 min and incubation in blocking solution (PBS, 2% BSA, and 5% horse serum) for 1 h. Tissue sections were rinsed with PBS twice for 5 min and subsequently incubated at RT overnight in monoclonal antibodies against MHC (1:100), desmin (1:10), titin (1:10), -AChR (1:10), MCK (1:100), and troponin-T (1:100). The antibodies used in these experiments were identical to those used for Western blot assays (see above). Tissue sections were then washed three times for 5 min in PBS/BSA and incubated with an Alexa-488 secondary Ab (anti-mouse, 1:200; Molecular Probes, Eugene, OR, USA). Endogenous filamentous actin was visualized using rhodamine-phalloidin (Molecular Probes). All procedures were carried out at RT. Images of immunolabeled tissue sections were captured on a Bio-Rad 1024 confocal microscope.
EO cDNA library characterization
The EO cDNA library from an adult S. macrurus was provided by Dr. Harold Zakon (University of Texas, Austin, TX, USA). The Lambda ZAP II vector (Stratagene, La Jolla, CA, USA) was used because of its ability to accommodate DNA inserts from 0 to 10 Kb in length. The EO cDNA library was generated with EcoR1 linkers, cloned into the EcoR1 site, and plated according to manufacturer’s instructions (Stratagene).
Cloning and subcloning of cDNAs
Clones were randomly selected and kept as recombinant bacteriophage. Among these, 420 phage were rescued to recombinant plasmid form as described by manufacturer (Stratagene). A single colony per plate was randomly chosen and was restreaked for purification. LB-ampicillin plates were incubated at 37°C overnight. Glycerol stocks (40% glycerol) were made for each bacterial clone and stored at –80°C until DNA sequencing verification.
Plasmid isolation, purification, and sequencing
Bacterial cells from each clone were grown in LB medium at 37°C with shaking overnight. cDNA plasmids were isolated and purified using the QIAprep Spin Miniprep Kit (Qiagen, Valencia, CA, USA). Polymerase chain reaction (PCR) (conditions: 24 cycles of 96°C for 10 s, 50°C for 5 s, and 60°C for 4 min) was performed using the T7 and T3 sequencing primers (Sigma-Genosys), and the PCR products purified using the Performa DTR gel filtration spin columns (Edge Biosystems, Gaithersburg, MD, USA). Amplified inserts were sequenced using an Applied Biosystems automated DNA sequencer (Model 3100). Sequences from 420 plasmids containing a single cDNA from the EO were compared to known sequences found in the National Center for Biotechnology Information (NCBI) database using the standard nucleotide program basic local alignment search tool (BLAST) (http://www.ncbi.nlm.nih.gov/basic local alignment search tool/). cDNAs that had a high similarity to known sequences (
70%) or had an expected value of 1e–10 or higher were selected for further analysis and categorized according to their biological function.
Total RNA isolation, cDNA synthesis, and PCR
Total cellular RNA was isolated from skeletal muscle and liver using a protocol adapted from the guanidinium thiocyanate method of Sambrook et al. (31)
, and from EO using the RNeasy kit (Qiagen) from different adult fish. The transcript composition of each tissue did not appear to differ by using either of these RNA isolation procedures given that RNA samples from muscle or EO isolated using both procedures yielded similar results (data not shown). Residual DNA was removed by treating tissue RNAs with Turbo DNA free (Ambion) and analyzed by spectrophotometry (optical density260/optical density280). On average, total RNA isolations yielded 1–3% of starting material from each of the different tissues. cDNAs for skeletal muscle, EO, and liver were synthesized from total RNA (250–300 ng) using the SuperScriptTM First-Strand Synthesis System for RT-polymerase chain reaction (RT-PCR) (Invitrogen).
Heterologous degenerate oligonucleotide primers were designed for -AChR based on GenBank vertebrate sequences using CODEHOP (32)
. On verification of the cloned sequences obtained from muscle and EO by sequencing 15 cDNA clones per tissue using an Applied Biosystems automated DNA sequencer (Model 3100), homologous primers specific to S. macrurus were generated from the on-line Primer3 software program (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi). Based on mRNA sequences identified from the EO cDNA library (see below), gene-specific homologous primers for -actin, -AChR, desmin, MCK, fast MHC, slow MHC, and titin were also designed using Primer3. To characterize the transcription profile of these muscle genes in muscle and EO, semiquantitative PCR amplification of the newly generated cDNAs (1 µg of cDNA) was performed for 10–35 cycles of 15 s denaturation (94°C), 30 s primer-annealing (60–65°C, Table 1
), and 30–60 s elongation (72°C) following an initial 2 min denaturation (94°C). A final elongation step was performed for 7 min (72°C). PCR products were run out on a 1–2% agarose gel stained with ethidium bromide (10 µl/100 ml). To monitor DNA contamination in all experiments, control reactions were performed without the cDNA template or reverse transcriptase enzyme. In addition, PCR experiments with liver cDNA were used as a negative control.
|
In situ hybridization
In situ hybridization was performed as described in Basyuk et al. (33)
and modified for optimal application using S. macrurus tissues.
Probe synthesis and tissue preparation
RT-PCR products from mature EO for fast MHC (372 bp) and desmin (820 bp) were cloned into the pCR® 2.1-TOPO® vector (Invitrogen) and pGEM®-T Easy vector (Promega, Madison, WI, USA), respectively, and amplified by PCR as described previously (see above). Digoxigenin (DIG)-labeled riboprobes for fast MHC and desmin were generated in the sense and antisense directions in an in vitro transcription reaction using the DIG RNA labeling kit (Roche, Germany). Cryosections (12–14 µm thick) from adult tails were collected on SuperFrost charged glass slides (VWR, Chester, PA, USA). Tissue sections were air-dried for 20 min prior to fixation in 4% paraformaldehyde in PBS (pH 9.5) for 60 min. Sections were then washed twice for 3 min with PBS and permeabilized with 50 µg/ml proteinase K in PBST (PBS, 0.1% Tween-20) for 10–15 min before washing three times for 5 min with PBS and soaking in 5x saline-sodium citrate (SSC), 50% formamide for 15 min. All fixation and permeabilization steps were performed at RT.
Hybridization and detection of bound riboprobes
DIG-labeled riboprobes for fast MHC and desmin were diluted to a final concentration of 600 ng/mL in hybridization solution (50% formamide, 5x SSC, 0.02% RNase-free BSA, 250 µg/mL bakers’ yeast RNA, 10% dextran), heated at 80°C for 1 min and iced. Each riboprobe (100–150 µl) was added to tissue sections on a glass slide and coverslipped. Riboprobe incubation was performed (5x SSC, 50% formamide) in a humidified chamber overnight at 55°C. Slides were transferred to a Coplin jar with 2x SSC, 50% formamide for 15 min at 55°C to remove the coverslips. Slides were transferred to a clean Coplin jar and washed in 2x SSC, 50% formamide at 55°C for 30 min. Slides were then washed twice for 30 min in 0.2x SSC, 50% formamide at 55°C, transferred to Coplin jars to soak in 0.2x SSC for 5 min, and then washed with Wash Buffer (0.1 M maleic acid, 0.15 M NaCl; pH 7.5) for 5 min at RT. The DIG Nucleic Acid Detection Kit (Roche) was used to detect endogenous fast MHC and desmin mRNAs that hybridized to their respective DIG-labeled probes.
RESULTS
Protein profile of the electric organ differs from that of skeletal muscle
Since all the antibodies used were generated against mammalian antigens, we performed Western blots to test their specificity in skeletal muscle and EO of S. macrurus. Antibodies against -actin, -AChR, desmin, MCK, the different isoforms of MHC, tropomyosin, and troponin-T bound to proteins in S. macrurus tissue homogenates (Fig. 1
). All antibodies labeled bands in S. macrurus skeletal muscle that corresponded in size to the mammalian muscle proteins that these antibodies specifically detect. For example, antibodies against -actin, -AChR, desmin, and tropomyosin labeled bands of the expected sizes in both muscle and EO, i.e., 45 kDa, 80 kDa, 50 kDa, and 36 kDa in size, respectively. A faintly labeled band of 
118 kDa was also obtained in muscle and EO with the anti-AChR Ab (data not shown) supporting the affinity of Ab mAb35 to different ß-subunits of the nicotinic receptor (34)
. Unlike the -actin, -AChR, and desmin antibodies, the tropomyosin Ab labeled a band in the EO homogenate (Fig. 1)
, but it did not immunolabel electrocytes in tissue cryosections (Fig. 2
). Differences in detection might be influenced by conformation and/or accessibility of the antigen under different tissue processing techniques.
|
|
Antibodies against MCK and troponin-T labeled bands of expected size –43 kDa and 38 kDa, respectively, but only in skeletal muscle (Fig. 1)
. In the muscle, the MCK Ab also labeled a band of 20 kDa that could correspond to a degradation product or an unknown protein (data not shown). The troponin-T Ab also labeled a protein in muscle of 42 kDa in size, which might correspond to one of three splice variants reported in limb skeletal muscle in other vertebrates (data not shown, 35, 36). The antibodies used to detect slow and fast, or multiple MHC isoforms labeled bands in skeletal muscle of 200 kDa in size (Fig. 1)
, corresponding to the known sizes of these sarcomeric proteins (37
, 38)
. No proteins were labeled in the EO homogenate with each of these 3 antibodies (Fig. 1)
.
At the cellular level, we found that mature electrocytes immunoreact with antibodies against desmin, -actin, and -AChR, but not with antibodies against tropomyosin, slow MHC, fast MHC, or an Ab against multiple sarcomeric MHCs (Fig. 2)
. We also extended the characterization of the protein composition of electrocytes by including their immunoreactivity to antibodies against MCK, troponin-T and titin, and the rhodamine-conjugated phalloidin, which labels filamentous actin (Fig. 3
). We found that electrocytes are not immunolabeled by either anti-MCK or antitroponin-T antibodies (Fig. 3)
. Titin was detected in electrocytes (Fig. 3)
. Further, although the presence of molecular -actin in electrocytes was confirmed (Fig. 2)
, endogenous filamentous actin by phalloidin labeling was not observed (Fig. 3)
. As expected, all mature skeletal muscle fibers were positively labeled with the above antibodies and phalloidin (Figs. 2
, 3)
. In sum, with the exception of tropomyosin, the cross-reactivity obtained with these antibodies in tissue sections was in agreement with the banding pattern obtained in Western blots, and the data revealed that MCK, MHCs, tropomyosin, and troponin-T are not detected in cells of the EO.
|
Transcription profile of select muscle genes in the mature electric organ
To test whether phenotypic differences between muscle and EO are due to a differential transcription of muscle genes, we used PCR experiments to characterize the expression pattern of -actin, -AChR, desmin, MCK, MHC isoforms, tropomyosin, and troponin-T. Partial cDNA fragments specific to -AChR (722 bp), -actin (619 bp), desmin (820 bp), fast MHC (372 bp), slow MHC (512, 650, and 870 bp), MCK (550 bp), and titin (350 bp) were isolated and cloned from S. macrurus skeletal muscle and EO. The partial fast MHC cDNA isolated from both tissues was most similar to IIa and IIb MHC mRNAs, and all three slow MHC bands detected in muscle and EO corresponded to slow MHC isoforms annotated in the NCBI database. Each of these muscle transcripts was readily detected in adult skeletal muscle and, surprisingly, in the EO as well (Fig. 4
). Further, the detectable levels of -AChR, desmin, MHCs, and MCK transcripts in the EO were at intensities similar to those found in skeletal muscle, whereas bands corresponding to -actin and titin transcripts in EO were of lower intensity than those in muscle (Fig. 4)
.
|
The presence of fast MHC and MCK transcripts in EO is particularly noteworthy because this organ lacks these proteins (Figs. 1
2
3)
. Further, the relatively lower levels of titin mRNA in EO compared to those in muscle were also intriguing because the protein levels were similar in these tissues (Fig. 3)
. Hence, to establish the levels of endogenous transcripts of these genes in EO and skeletal muscle, we performed semiquantitative PCR experiments. As shown in Fig. 5
, transcript levels of fast MHC and MCK in muscle and EO were comparable after 22 and 18 amplification cycles, respectively, prior to the reaction reaching saturation. In contrast, titin mRNA content in skeletal muscle was higher than in EO across different amplification cycles (Fig. 5)
. Transcripts for MCK, fast MHC, and titin were not detected in nonmyogenic tissue such as the liver (Fig. 5)
. The presence of desmin and fast MHC transcripts in muscle fibers and electrocytes was confirmed by in situ hybridization using DIG-labeled RNA riboprobes specific to these S. macrurus genes (data not shown). Together, these data show that the transcript profile for different functional muscle genes of the EO did not coincide with its corresponding protein profile, as all 7 transcripts (Fig. 4)
, but not their encoded proteins (Table 2
; Figs. 2
, 3
), were detected.
|
|
cDNA profile in S. macrurus electric organ
We randomly selected a sample of 420 cDNA clones from an EO cDNA library. Table 3
shows the composition of this random cDNA sample: 128 cDNAs (30%) lacked enough information to be classified, or had unknown functions, and 292 cDNAs (70%) were classified into categories on the basis of general functions of the genes. We found that 168 of the 420 cDNAs sequenced corresponded to distinct genes in the NCBI database and these were classified into 11 categories based on protein function (Table 4; supplemental file). We evaluated the EO cDNAs by characterizing their known or preferred tissue expression using the human protein reference (http://www.hprd.org), the Swiss-Prot Protein (http://ca.expasy.org/sprot), and Gene Cards (http://biostatpub2.mdanderson.org/genecards/index.shtml) on-line databases. Based on these on-line databases, of the 168 different transcripts identified from the EO, 22 (13%) are restricted to muscle, 15 (9%) are found in muscle and nervous tissues, 112 (67%) are detected in multiple tissues in addition to muscle and nervous tissues, and 19 (11%) are not found in muscle (Table 4).
|
Among the 168 identified mRNAs, some were more highly represented in our population of cDNAs sequenced (Table 3)
. For example, the most abundant category of transcripts identified was that of cytoskeletal genes (18%). Within this category titin (20 clones), desmin (13 clones), and fast MHC (12 clones) transcripts were identified. In fact, actin and slow MHC transcripts were also identified in our cDNA pool (Table 4). The second category with the highest frequency of transcripts represented in our EO cDNA population was metabolism with MCK (8 copies) and cytochrome c oxidase (8 copies) being the two highest-frequency clones identified (Table 3)
. Hence, our cDNA sequence data corroborates the results from our PCR and in situ hybridization studies showing the presence of transcripts in the mature EO for the muscle genes -actin, desmin, fast-MHC, slow-MHC, MCK, and titin (Tables 2
, 4, Figs. 4
5
). That some of the mRNAs (11%) are not associated with skeletal muscle, and that 30% of the randomly selected cDNAs sequenced had unknown functions is indicative of a unique cell phenotype that is not a product of a simple down-regulation of select muscle genes.
DISCUSSION
The manifestation of a partial muscle phenotype in the protein composition of the mature EO is an intriguing problem in our current understanding of the regulation of gene expression that brings about the skeletal muscle program. In this study, we present evidence suggestive of the involvement of transcriptional-independent processes in the maintenance of the muscle-like phenotype of the EO in S. macrurus. We also report the classification and frequency of distinct transcripts from a random selection of 420 clones from an EO cDNA library. The present characterization of skeletal muscle gene expression at the transcript and protein levels provides a new insight into our understanding of the plasticity of striated muscle fibers to form the noncontractile EO.
Down-regulation of select muscle proteins in S. macrurus tissues
We studied the expression patterns of nine common gene markers—-AChR, -actin, desmin, fast MHC, slow MHC, MCK, titin, tropomyosin, and troponin-T—of skeletal muscle using Western blot and immunofluorescence analyses. The Western blot data obtained with eight of nine antibodies using muscle homogenates coincided with those obtained using tail cryosections indicating that these antibodies are reliable indicators of these muscle proteins in S. macrurus. Increasing our availability of informative markers that can monitor the expression of muscle genes in this fish species will be useful in future studies that focus on the changes that occur in the skeletal muscle program during the formation of electrocytes from muscle fibers at the protein level.
Our present data have further defined the fully differentiated EO phenotype in S. macrurus. Specifically, our immunofluorescence and Western blots showed an absence of MCK, slow MHC, fast MHC, and troponin-T proteins in the EO. Creatine kinases (CK) catalyze the regeneration of ATP and there are differences in kinetic parameters between various CK isozymes (39
, 40)
. It is very likely that the energy demand in electrocytes cannot be fulfilled by MCK, and this must be substituted by a different CK isozyme. For example, different populations of cholinergic neurons with distinct morphologies and electrical behavior innervate skeletal muscle and EO (1
, 41)
. Somatomotoneurons innervate muscles in the tail that we presume control tail position and fire irregularly at frequencies ranging from 6 to 12 Hz, as it occurs in other teleosts (42, 43). In contrast, the larger electromotoneurons that innervate electrocytes are driven continuously at rates of 100–200 Hz in this species (41)
. Energy-consuming processes imposed by this electrical activation pattern very likely requires modifications in the metabolic profiles of electrocytes, with changes in metabolic enzymes in a tissue-specific manner. Despite differences in neural activity patterns, however, our data is consistent with previous studies (4
, 20
, 21)
that have demonstrated synaptic transmission in this electromotor synapse to involve nicotinic AChRs at the nerve junction endplates. In fact, it appears that proteins whose functions are associated with stabilizing a large number of these AChRs in muscle fibers (e.g., actin, utrophin, and dystrophin) are also retained in the fully differentiated electrocytes (20)
.
Previous evidence revealed a lack of sarcomeric structures in mature electrocytes (44)
. Given our present data, a lack of sarcomeres in electrocytes results from the absence of proteins associated with contraction, including MHC isoforms, tropomyosin, troponin-T, and filamentous actin. Note that, although electrocytes contain actin monomers, the absence of phalloidin label suggests that polymerization into filamentous actin molecules does not occur in electrocytes. Unlike MHCs, tropomysin, and troponin-T proteins, we found that the titin protein is not down-regulated in mature electrocytes and its presence, despite the absence of sarcomeric structures, was unexpected. However, although titin plays a key role in establishing and maintaining the integrity of the contractile sarcomere unit in muscle fibers, it is also found interweaved throughout the cytoskeletal network of nonmuscle cells (45, 46). Hence, titin may be retained in electrocytes to work in concert with other cytoskeletal proteins, i.e., desmin and keratin, to help sustain their structural morphology (44)
. Furthermore, titin has emerged as being involved in several complex protein-protein interaction networks that are involved in pathways controlling gene expression and protein turnover (41)
. In this regard, electrocytes represent an ideal model system to study the distinct cellular functions that titin fulfills in the absence of sarcomeric structures.
Evidence for transcription-independent regulation processes in muscle and EO
The fact that skeletal muscle development and the expression of many genes coding for skeletal muscle proteins are under transcriptional regulation (22
23
24)
prompted us to test the hypothesis that the partial muscle phenotype of the EO is coupled to the select transcription of muscle genes. However, our transcript analysis revealed that -actin, -AChR, desmin, MCK, MHC isoforms, and titin are transcribed in the EO, despite our inability to detect some of these gene products at the protein level. It is feasible that mechanisms that increase the degradation rate of a select population of muscle mRNAs account for the absence of muscle gene products in the EO. Although we cannot exclude the possibility of sample contamination by muscle during EO dissection, our findings were consistent across different fish (Figs. 4
, 5)
from which EO tissue farthest from skeletal muscle was taken deems this possibility unlikely. Further, we have recently demonstrated the expression of myogenic regulatory factors (MRFs) in the mature EO of S. macrurus (20; unpublished data). Because these MRFs are essential for the transcriptional activation of the muscle genes characterized in this study (24
25
26
27
28
29)
, data from the present study suggest the presence of functional MRF proteins in skeletal muscle and EO of S. macrurus. Future experiments that fully evaluate the protein activity of MRFs and the activation of their target genes in mature electrocytes will be critical to characterize the unique mechanism of muscle gene regulation that exists in the EO.
Examples of the direct contribution of transcriptional-independent events in the maintenance and plasticity of muscle genes are relatively scarce, but several reports have demonstrated the importance of these mechanisms in the control of muscle gene expression. For instance, targeting of select mRNAs for post-transcriptional regulation via their 3' untranslated regions has been well reported for various muscle transcripts including MHC (48
, 49)
and a number of genes encoding synaptic proteins in muscle cells (50
, 51)
. Cytosolic calcium levels have also been implicated in the modulation of sarcomeric protein levels without changing mRNA levels (52
, 53)
. Lastly, evidence suggests that some aspects that distinguish different muscle fiber types, i.e., slow and fast types, are post-transcriptionally regulated and this post-transcriptional regulation is dependent on the electrical activity patterns imposed on the muscle fiber (54)
. The latter result is of great relevance to our model system as the electrical activation pattern of the EO is very different than that of muscle (41
42
43)
. Interestingly, mature electrocytes re-express MHC and tropomyosin and form sarcormeres de novo after removal of their neuronal input (4)
. In sum, we interpret these data to suggest that differences between the skeletal muscle and EO phenotypes might be accounted for by differences in post-transcriptional muscle gene expression mechanisms in addition to transcription control.
First transcript profile analysis of an electric organ
Our analysis of the 420 randomly selected cDNA clones from an EO library included 168 different mRNAs of known functions. Interestingly, over 90% of these mRNAs have been found in a variety of tissues other than muscle and their encoded proteins are known to differ in structure, function, and embryological origin. This level of transcript diversity in the EO could reflect the contribution of housekeeping genes and certain cell types shared by different tissues, including endothelial cells, fibroblasts, other connective tissue cell types, etc. Also of interest was the relatively high proportion of cDNAs with unknown function (30%) in this randomly selected sample. This transcription profile of the S. macrurus EO reveals the uniqueness of this noncontractile electrogenic tissue that is derived from fully differentiated skeletal muscle fibers (44)
. Although 168 mRNAs can be viewed as a relatively small fraction of the total number of transcripts in the cell, this transcript profile represents a starting point for a more complete phenotypic analysis of electrocytes. Given that this is the first source of publicly available mRNA sequences for a fully differentiated EO, these data provide a resource for the generation of probes that can be used to study the changes in gene transcription that may occur during the conversion of skeletal muscle to EO in S. macrurus, as well as in other electric fish species.
ACKNOWLEDGMENTS
This research was supported by National Institute of Health Grants S06-GMO8136–27, RR16480–01 and GM061222–05 to G.A.U. We are grateful to Dr. Colleen Jonsson, Roberto Medina, and Emanuel Salgado for technical assistance in cDNA subcloning and sequencing. We also appreciate the assistance of Shannon Manuelito and Amanda Meenach in the use of the confocal microscope, and Dr. Mary O’Connell for critical comments in an earlier draft of this manuscript.
Received for publication May 15, 2006. Accepted for publication August 14, 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  J. A. Kim, C. Laney, J. Curry, and G. A. Unguez Expression of myogenic regulatory factors in the muscle-derived electric organ of Sternopygus macrurus J. Exp. Biol., July 1, 2008; 211(13): 2172 - 2184. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
