Abstract
The important intracellular oxygen-binding protein, myoglobin (Mb), is thought to be absent from oxidative muscle tissues of the family of hemoglobinless Antarctic icefishes, Channichthyidae. Within this family of fishes, which is endemic to the Southern Ocean surrounding Antarctica, there exist 15 known species and 11 genera. To date, we have examined eight species of icefish (representing seven genera) using immunoblot analyses. Results indicate that Mb is present in heart ventricles from five of these species of icefish. Mb is absent from heart auricle and oxidative skeletal muscle of all species. We have identified a 0.9-kb mRNA in Mb-expressing species that hybridizes with a Mb cDNA probe from the closely related red-blooded Antarctic nototheniid fish, Notothenia coriiceps. In confirmation that the 0.9-kb mRNA encodes Mb, we report the full-length Mb cDNA sequence of the ocellated icefish, Chionodraco rastrospinosus. Of the eight icefish species examined, three lack Mb polypeptide in heart ventricle, although one of these expresses the Mb mRNA. All species of icefish retain the Mb gene in their genomic DNA. Based on phylogeny of the icefishes, loss of Mb expression has occurred independently at least three times and by at least two distinct molecular mechanisms during speciation of the family.
Keywords: Channichthyidae, oxygen transport, phylogenetics
Icefishes (family Channichthyidae) of the Southern Ocean surrounding Antarctica are unique among vertebrate animals; all 15 species lack hemoglobin (1, 2) and, despite their highly aerobic mode of metabolism, are believed also to lack the intracellular respiratory pigment, myoglobin (Mb) (3–7). Mb normally is present in high concentration in aerobic muscle tissues of vertebrate animals, where it functions both as an intracellular oxygen reservoir and to facilitate the transcellular diffusion of oxygen (8).
The Perciform suborder Notothenioidei (which includes the icefishes) arose and evolved in coastal Antarctic waters during the last 25 million years (9, 10). Antarctica became isolated at that time upon the opening of the Drake Passage and establishment of circumpolar currents that led to the rapid cooling of the Southern Ocean. At present, the ocean surrounding Antarctica is uniquely cold and thermally stable; water temperatures around the Antarctic Peninsula fluctuate only between +0.3 and −1.87°C annually (11, 12). Divergence of mitochondrial DNA sequences suggests that radiation of notothenioid families began 7 to 15 million years ago, but that speciation of channichthyid icefish began approximately 1 million years ago (13).
Icefishes exhibit several unique physiological features. Cardiovascular adaptations to compensate for lack of hemoglobin in the circulation include lower blood viscosity, increased heart size, greater cardiac output, and increased blood volume compared with their red-blooded notothenioid relatives (2, 14–16). Combined with the high aqueous solubility of oxygen at severely cold body temperature, these cardiovascular features are considered necessary to ensure that tissues obtain adequate amounts of oxygen carried in physical solution by the plasma. Although enhancing circulatory delivery of oxygen, these adaptations do not assist intracellular movement of oxygen within tissues. Heart ventricle (7, 14) and aerobic skeletal muscle (17) of icefishes contain among the highest mitochondrial densities (>40% of cell volume) of any vertebrate tissues. Consistent with this robust mitochondrial population, energy metabolism of the icefishes is highly and obligately aerobic (2), exacerbating the challenges to both circulatory and intracellular delivery of oxygen.
Pale coloration of icefish tissues has led most researchers to assume that Mb is not present in icefish oxidative muscles (3–7). However, we collected three channichthyid species off the Antarctic Peninsula, Chionodraco rastrospinosus, Pseudochaenichthys georgianus, and Chaenodraco wilsoni, that displayed distinctly rose-colored hearts. By contrast, hearts of two other icefish species common to Peninsular waters, Chaenocephalus aceratus and Champsocephalus gunnari, were pale yellow. Absorption spectrum of a clarified supernatant (40,000 × g) from P. georgianus ventricle exhibited maxima at 530 and 580 nm, characteristic of oxymyoglobin. Although not conclusive, these observations prompted us to ascertain whether Mb is expressed in aerobic muscle tissues of icefishes.
MATERIALS AND METHODS
Animals and Tissues.
Live specimens of icefishes C. aceratus, C. rastrospinosus, C. gunnari, P. georgianus, C. wilsoni, and red-blooded notothenioids Notothenia coriiceps and Gobionotothen gibberifrons were collected by Otter Trawl off the south shores of Low and Brabant Islands during the austral autumn in 1991, 1993, and 1995. Pagetopsis macropterus was captured off Brabant Island in March 1995. Cryodraco antarcticus tissues were the kind gift of A. L. DeVries (University of Illinois) (the specimen was captured from McMurdo Sound). Chionodraco hamatus tissues, from animals captured at Terra Nova Bay, were the kind gift of G. di Prisco and R. Acierno (Italian National Antarctic Program). Heart ventricle, auricle, pectoral adductor profundus, testes, and spleen tissues were dissected immediately upon sacrifice of the animals. Samples were frozen in liquid nitrogen, transported to the United States on dry ice, and stored at −70°C until isolation of protein and nucleic acids.
Protein Extraction and Immunoblot Analysis.
Soluble polypeptides were liberated from heart ventricles by homogenization in 20 mM Hepes (pH 7.8) at 4°C and centrifuged at 10,000 × g for 10 min. Protein in the supernatant was determined by BCA assay (Pierce). Polypeptides were denatured by boiling in the presence of 1% SDS/1 mM 2-mercaptoethanol, separated by electrophoresis through 17% Tricine–SDS/PAGE gels, and either electrophoretically transferred to polyvinylidene difluoride (PVDF) membranes (Micron Separations, Westboro, MA) or stained using Coomassie brilliant blue R-250. For slot blots, supernatants were diluted in 1× phosphate-buffered saline (PBS) (pH 7.4) and vacuum blotted onto PVDF membrane. Electroblot and slot-blot membranes were blocked by incubation overnight in 5% nonfat milk, washed in 1× PBS, and incubated with a 1:500 dilution of mouse anti-human Mb monoclonal antibody (Sigma). This antibody displayed strong cross-reactivity with Mb isolated from red-blooded notothenioid species known to express the pigment. Bound antibody was detected by a rabbit anti-mouse IgG secondary antibody (1:1000 dilution) conjugated to alkaline phosphatase (Bio-Rad), and visualized by subsequent incubation in stabilized Western blue substrate (Promega).
RNA Gel Blot Hybridization and PCR Amplification.
Total RNA was isolated from finely ground, frozen heart ventricles by acid guanidinium–thiocyanate–phenol chloroform extraction (18). RNA concentrations were determined in triplicate by spectrophotometric analyses. Equal amounts (5 μg) of total RNA were resolved by electrophoresis through 1% agarose/formaldehyde gels (19) and blotted to GeneScreen Plus nylon membranes (DuPont). Blots were probed with a 329-bp segment of N. coriiceps Mb cDNA corresponding to codons 5–114 that was 32P-labeled by the random primer method (Boehringer Mannheim). Blots received two successive 30-min washes in 0.1× SSC/0.1% SDS at 63°C. The N. coriiceps partial Mb cDNA was obtained by reverse transcriptase-PCR (20); first-strand synthesis was primed by oligo(dT)–NotI primer adapter (Pharmacia), and amplification employed degenerate oligonucleotides based on conserved segments of tuna and carp Mb polypeptide sequences (Swiss-Prot accession nos. PO2205 and PO2204, respectively). Sequencing confirmed that the amplified sequence encoded Mb (21). The distal two-thirds of the C. rastrospinosus Mb sequence was obtained by reverse transcriptase-PCR using the oligo(dT)–NotI primer adapter and nondegenerate primers based upon the N. coriiceps Mb sequence. The 5′ proximal segment was obtained by linker-mediated PCR (22) using the 5′ rapid amplification of cDNA ends system (CLONTECH) and a Mb-specific antisense oligonucleotide primer. Mb-specific oligonucleotides were used to prime dideoxy cycle-sequencing reactions (Perkin–Elmer), which were resolved with an Applied Biosystems 373A automated DNA sequencer.
Southern Blot Hybridization.
Genomic DNA was isolated from testes (or spleen of P. macropterus) by proteinase K digestion and phenol chloroform extraction as described by Sambrook et al. (19). Ten micrograms of each genomic DNA sample was digested with 2 units of HindIII overnight. Restriction fragments were resolved by electrophoresis through 0.7% agarose gels cast in 25 mM Tris-borate (pH 8.3) containing 1 mM EDTA, and transferred to Hybond-C nylon membranes (Amersham) by capillary blot. Blots were hybridized with the 32P-labeled, 329-bp N. coriiceps cDNA PCR product. Blots were washed twice in 0.2× SSPE at 62°C and exposed to Kodak X/AR5 film for 7–10 days with a tungsten intensifying screen.
RESULTS
Immunoblot analyses (Fig. 1) show that Mb is present in the heart ventricles of at least five icefish species: C. rastrospinosus, P. georgianus, C. wilsoni, C. antarcticus, and C. hamatus. Both polyclonal (not shown) and monoclonal antibodies (Fig. 1), which recognize Mb in a broad variety of vertebrates, identified a polypeptide of the expected size in these five icefish species. Teleost Mb is comprised of 146 amino acids (23), with an apparent molecular size of approximately 16 kDa. The more slowly migrating human Mb with a molecular size of 17.8 kDa contains 153 amino acids and is typical of mammalian Mbs (24, 25). Heart ventricles of the notothenioid species contained a 0.9-kb mRNA (Fig. 2), which hybridized under high stringency to a Mb cDNA probe isolated from the red-blooded Antarctic nototheniid N. coriiceps. Three icefish species, C. aceratus, C. gunnari, and P. macropterus, lacked the immunoreactive polypeptide (Fig. 1). C. aceratus and P. macropterus lacked Mb-hybridizing mRNA (Fig. 2). However, C. gunnari individuals reproducibly exhibited low steady-state quantities of the 0.9-kb Mb-hybridizing mRNA (Fig. 2) despite consistent absence of the Mb polypeptide in protein extracts from the same tissue samples. Neither immunoblot nor RNA gel blot hybridization detected Mb in the heart auricle or highly aerobic skeletal muscle (pectoral adductor profundus) of the icefishes or red-blooded Antarctic nototheniids tested (data not shown). This highly tissue-specific expression of Mb, confined exclusively to the heart ventricle, is unique among vertebrates and is a possible synaptomorphic character among notothenioid fishes.
Verification that the 0.9-kb mRNA encodes Mb was obtained by sequencing C. rastrospinosus cDNA products produced by reverse transcriptase-PCR and ligation-mediated PCR. The composite full-length sequence of C. rastrospinosus Mb cDNA (Fig. 3) is comprised of a short, 57 nucleotide 5′ untranslated leader, a 441 nucleotide coding sequence, a 78 nucleotide 3′ untranslated region proximal to the poly(A) tail, and includes a segment with 98% identity to the 329-bp N. coriiceps Mb PCR product. The predicted coding sequence contains 147 amino acids; this is consistent with the mature 146 residue Mb polypeptides of other teleosts from which the N-terminal methionine is removed. The sequence flanking the initiator methionine is consistent with the Kozak consensus sequence −3AynATGG+4 (26). When the complete (100%) deduced Mb polypeptide from C. rastrospinosus is compared with that of tuna (Thunnus albacares) Mb, 78% of the residues are identical and an additional 9% are similar. Key structural elements, such as his60 and his88, which mediate heme binding, are conserved. This high degree of homology leaves no doubt that some icefish species express Mb in cardiac ventricle.
Genomic DNA of all channichthyid species tested to date contain restriction fragments that hybridize to nototheniid Mb cDNA (Fig. 4). Red-blooded nototheniid species N. coriiceps and G. gibberifrons exhibited two or three Mb-specific restriction fragments indicating that Mb is a single-copy or low-copy gene. Both C. aceratus and P. macropterus contained Mb-specific fragments, although Mb mRNA was not detected in the heart ventricle of either species. Mb-hybridizing bands of similar size were apparent in genomic DNA of the Mb-expressing channichthyid species C. rastrospinosus and C. gunnari. These results indicate that loss of Mb expression in C. aceratus and P. macropterus did not result from large-scale deletion of the gene; deletion of the β-globin gene is the apparent mechanism by which hemoglobin expression was lost in C. aceratus (27).
DISCUSSION
Our data provide conclusive evidence that Mb is present and expressed in several species of Antarctic icefish that lack hemoglobin. The expression of Mb is extremely tissue-specific in both the icefishes and their red-blooded nototheniid relatives. Mb expression is confined to the heart ventricle and absent from the primary aerobic skeletal muscle used for labriform locomotion, the pectoral adductor profundus. Ability to express Mb in skeletal musculature was apparently lost early in the notothenioid lineage and prior to radiation of the channichthyid icefishes.
Ability to express Mb in the heart ventricle appears to have been lost by three independent mutational events during the evolution of icefishes. Comparison of Mb expression with a morphologically based phylogenetic tree for the channichthyids (9, 28) indicates that three icefish species that lack Mb expression in the heart ventricle, C. aceratus, C. gunnari, and P. macropterus, occupy distinct clades; two of these clades contain relatives that express Mb (Table 1). Our recent phylogenetic analysis of mitochondrial DNA control region sequences also unambiguously supports the occurence of three independent losses of Mb expression (I. Kornfield, Y. K. Tam, M.E.V., and B.D.S., unpublished data). Furthermore, the observation that C. gunnari contains Mb mRNA, whereas the other two species do not, suggests at least two distinct molecular mechanisms for the loss of Mb: failure to synthesize the Mb polypeptide in C. gunnari, and failure to transcribe the Mb gene or process Mb transcripts in C. aceratus and P. macropterus. All three species contain remnants of the Mb gene in the genome, in distinct contrast to the apparent mechanism for the loss of hemoglobin expression within the family (27). These results suggest that loss of Mb expression is both a recent and recurrent event in the evolutionary history of icefishes.
Table 1.
Genus | Total no. of species | Species examined | Polypeptide | RNA | |
---|---|---|---|---|---|
Champsocephalus | 2 | gunnari | − | + | |
Pseudochaenichthyes | 1 | georgianus | + | + | |
Neopagetopsis | 1 | ||||
Pagetopsis | 2 | macropterus | − | − | |
Dacodraco | 1 | ||||
Channichthys | 1 | ||||
Cryodraco | 1 | antarcticus | + | + | |
Chionobathyscus | 1 | ||||
Chaenocephalus | 1 | aceratus | − | − | |
Chionodraco | 3 | rastrospinosus | + | + | |
hamatus | + | + | |||
Chaenodraco | 1 | wilsoni | + | + |
All species contain discrete, single-copy genomic DNA fragments that hybridize to N. coriiceps Mb cDNA.
At present, it is unclear whether Mb plays a role in facilitated oxygen transport or storage at cold body temperatures. Maintenance of Mb expression in several icefish species could be either a vestigial characteristic or could indicate selective retention of the hemoprotein for aiding intracellular oxygen transport in the hearts of these species. Oxygen dissociation rates of mammalian and tuna Mbs are dramatically lengthened at cold temperature, suggesting that a similar respiratory pigment would not function at physiological temperatures of icefishes (29–31). However, sequence differences observed between the notothenioid and tuna Mbs may confer improved function at cold temperatures. If the latter possibility proves correct, it raises fascinating questions about how similar demands for storage and diffusion of oxygen are met in the heart ventricle of species without Mb and in the oxidative skeletal muscles of all Antarctic notothenioid species examined. No obvious differences have been observed among channichthyid species in either mitochondrial densities or in intracellular neutral lipid content, a factor that may play a role in aiding intracellular oxygen movement (16). Definitive resolution of the importance of Mb in these unusual animals may be aided by physiological studies of heart performance and determination of the kinetics of oxygen binding and dissociation for the hemoproteins from Antarctic icefishes.
Acknowledgments
We gratefully acknowledge the generous contribution of samples by Dr. A. L. DeVries (University of Illinois) and Drs. G. di Prisco and R. Acierno (Italian National Antarctic Program). This study would not have been possible without the excellent support provided by station personnel at the U.S. Antarctic Program’s Palmer Station, Antarctica, and the masters and crew of M/V Polar Duke. The work was supported by U.S. National Science Foundation (NSF) Grants OPP 92-20775 and 94-21657 to B.D.S. The University of Maine automated DNA sequencing facility was initiated by support through NSF EPSCoR Grant EHR91-08766 and Maine Science and Technology Foundation Grant 07G-GTSS930323.
ABBREVIATIONS
- Mb
myoglobin
- PVDF
polyvinylidene difluoride.
Footnotes
References
- 1.Ruud J T. Nature (London) 1954;173:848–850. doi: 10.1038/173848a0. [DOI] [PubMed] [Google Scholar]
- 2.Hemmingsen E A. In: Biology of Antarctic Fish. diPrisco G, Maresca B, Tota B, editors. New York: Springer; 1991. pp. 191–203. [Google Scholar]
- 3.Hamoir G. Comp Biochem Physiol B. 1988;90:557–559. [Google Scholar]
- 4.Feller G, Gerday C. Polar Biol. 1987;7:225–229. [Google Scholar]
- 5.Hamoir G, Gerardin-Otthiers N. Comp Biochem Physiol B. 1980;65:199–206. doi: 10.1016/0305-0491(79)90178-0. [DOI] [PubMed] [Google Scholar]
- 6.Walesby N J, Nicol C J M, Johnston I A. Br Antarct Surv Bull. 1982;51:201–214. [Google Scholar]
- 7.Johnston I A, Harrison P. Fish Physiol Biochem. 1987;3:1–6. doi: 10.1007/BF02183988. [DOI] [PubMed] [Google Scholar]
- 8.Wittenberg J B. Physiol Rev. 1970;50:559–636. doi: 10.1152/physrev.1970.50.4.559. [DOI] [PubMed] [Google Scholar]
- 9.Eastman J T. Antarctic Fish Biology: Evolution in a Unique Environment. San Diego: Academic; 1993. [Google Scholar]
- 10.Eastman J T. Am Zool. 1991;31:93–109. [Google Scholar]
- 11.DeWitt H H. Antarct Map Folio Ser. 1971;15:1–10. [Google Scholar]
- 12.Littlepage J L. In: Antarctic Research Series Vol. 5: Biology of the Antarctic Seas II. Llano G A, editor. New York: Am. Geogr. Soc.; 1965. pp. 1–10. [Google Scholar]
- 13.Bargelloni L, Ritchie P A, Patarnello T, Battaglia B, Lambert D M, Meyer A. Mol Biol Evol. 1994;11:854–863. doi: 10.1093/oxfordjournals.molbev.a040168. [DOI] [PubMed] [Google Scholar]
- 14.Johnston I A, Fitch N, Zummo G, Wood R E, Harrison P, Tota B. Comp Biochem Physiol A. 1983;76:475–480. [Google Scholar]
- 15.Fitch N A, Johnston I A, Wood R E. Respir Physiol. 1984;57:201–211. doi: 10.1016/0034-5687(84)90093-8. [DOI] [PubMed] [Google Scholar]
- 16.Hemmingsen E A, Douglas E L, Johansen K, Millard R W. Comp Biochem Physiol A. 1972;43:1045–1051. doi: 10.1016/0300-9629(72)90176-4. [DOI] [PubMed] [Google Scholar]
- 17.Londraville R L, Sidell B D. Respir Physiol. 1990;81:291–302. doi: 10.1016/0034-5687(90)90110-k. [DOI] [PubMed] [Google Scholar]
- 18.Chomczynski P, Sacchi N. Anal Biochem. 1987;162:156–159. doi: 10.1006/abio.1987.9999. [DOI] [PubMed] [Google Scholar]
- 19.Sambrook J, Fritsch E F, Maniatis T. Molecular Cloning: A Laboratory Manual. Plainview, NY: Cold Spring Harbor Lab.; 1989. [Google Scholar]
- 20.Saiki R K, Gelfand D H, Stoffel S, Scharf S, Higucci R, Horn R, Mullis K B, Elrich H A. Science. 1988;239:487–491. doi: 10.1126/science.2448875. [DOI] [PubMed] [Google Scholar]
- 21.Vayda M E, Small D J, Sidell B D. In: VI SCAR Antarctic Communities. Battaglia B, Valencia J, Walton D W H, editors. Cambridge, U.K.: Cambridge Univ. Press; 1997. in press. [Google Scholar]
- 22.Trout A B. In: PCR Technology: Current Innovations. Griffin H G, Griffin A M, editors. Boca Raton, FL: CRC; 1994. pp. 141–146. [Google Scholar]
- 23.Watts D A, Rice R H, Brown W D. J Biol Chem. 1980;255:10916–10924. [PubMed] [Google Scholar]
- 24.Akaboshi E. Gene. 1985;33:241–249. doi: 10.1016/0378-1119(85)90231-8. [DOI] [PubMed] [Google Scholar]
- 25.Blanchetot A, Price M, Jeffreys A J. Eur J Biochem. 1986;159:469–474. doi: 10.1111/j.1432-1033.1986.tb09909.x. [DOI] [PubMed] [Google Scholar]
- 26.Kozak M. Proc Natl Acad Sci USA. 1986;83:2850–2853. doi: 10.1073/pnas.83.9.2850. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cocca E, Ratnayake-Lecamwasam M, Parker S K, Camardella L, Ciaramella M, di Prisco G, Detrich H W. Proc Natl Acad Sci USA. 1995;92:1817–1821. doi: 10.1073/pnas.92.6.1817. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Iwami T. Mem Nat Inst Polar Res, Ser E. 1985;36:1–69. [Google Scholar]
- 29.Sato F, Shiro Y, Sakaguchi Y, Izukda T, Hayshi H. J Biol Chem. 1990;265:18823–18828. [PubMed] [Google Scholar]
- 30.Dowd M K, Murali R, Seagrave R C. Biophys J. 1991;60:160–171. doi: 10.1016/S0006-3495(91)82039-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Stevens E D, Carey F G. Am J Physiol. 1981;240:R151–R155. doi: 10.1152/ajpregu.1981.240.3.R151. [DOI] [PubMed] [Google Scholar]