Amino acid residues that are important for Hyal2 function as a receptor for jaagsiekte sheep retrovirus
© Duh et al; licensee BioMed Central Ltd. 2005
Received: 01 September 2005
Accepted: 28 September 2005
Published: 28 September 2005
Infection by jaagsiekte sheep retrovirus (JSRV) and by enzootic nasal tumor virus (ENTV) depends on cell-surface expression of the virus entry receptor, hyaluronidase 2 (Hyal2). Human Hyal2 binds the envelope (Env) proteins of these viruses and is functional as a receptor, but Hyal2 from mice does not bind Env nor does it mediate entry of either virus. Here we have explored the amino acid determinants that account for the difference in receptor function.
Analysis of human-mouse Hyal2 chimeric proteins showed that amino acid differences responsible for the difference in Hyal2 receptor activity were localized to the central third of Hyal2. Human Hyal2 mutants containing single or double amino acid replacements with the respective mouse amino acids were generated across this region and were assayed for activity. None of the single or double mutation reduced the receptor activity of human Hyal2 by more than 10-fold, whereas mouse Hyal2 activity is reduced 1,000-fold from that of human Hyal2. While the 3-dimensional structures of mammalian Hyal2 proteins are unknown, bee venom hyaluronidase shows significant amino acid similarity to human and mouse Hyal2 and its structure has been determined. Many mutations having the largest negative effects on human Hyal2 function mapped to a small region of the bee venom hyaluronidase close to but not overlapping the active site of the enzyme, suggesting that this site represents the binding site for Env. Analysis of synonymous and non-synonymous nucleotide substitutions in the coding sequences of multiple mammalian Hyal2 proteins shows that the proteins are undergoing strong selection for amino acid conservation. We found no evidence for positive selection of amino acid changes that might reflect evolution of mammalian hosts to resist JSRV or ENTV infection.
These results show that the greatly reduced receptor activity of mouse Hyal2 in comparison to that of human Hyal2 is determined by multiple amino acid changes acting in concert. In particular, no one amino acid change blocks infection. However, the most important amino acids map to a small patch on a predicted 3-dimensional Hyal2 structure, which may represent the binding site for Env.
JSRV and ENTV are closely-related retroviruses that induce tumors in the lower airways and nasal epithelium, respectively, of sheep and goats . Both viruses utilize the glycosylphosphatidylinositol-anchored cell-surface protein Hyal2 as a receptor for cell entry . Hyal2 is a member of a family of proteins, some of which exhibit high hyaluronidase activity and are capable of rapid degradation of hyaluronan, a component of the extracellular matrix. However, Hyal2 exhibits only weak hyaluronidase activity  and its primary biological role in mammals is not known. Evidence for Hyal2 function as the virus receptor is provided by experiments showing that expression of sheep or human Hyal2 in cells that are not normally susceptible to infection renders the cells fully infectable by retroviral vectors bearing the JSRV or ENTV Env proteins [4–7]. For example, the titers of JSRV and ENTV vectors on mouse cells expressing human Hyal2 are >1,000-fold higher than those on cells expressing mouse Hyal2. In addition, the receptor-binding surface (SU) domains of the JSRV and ENTV envelope proteins bind with high affinity to cells expressing human Hyal2 but not to cells expressing nonfunctional receptors such as mouse Hyal2 [6, 8, 9], indicating that Hyal2 is the primary receptor that binds virus prior to entry.
Here we have used a transient transfection assay to analyze the amino acid differences between human and mouse Hyal2 that account for the large difference in receptor activity of the two proteins. Non-susceptible NIH 3T3 mouse cells were transfected with the Hyal2 expression constructs and receptor function was quantitated by measuring transduction of the cells with a JSRV vector. We could localize most of the difference in receptor function to the central 38% of the Hyal2 proteins. Single and double amino acid changes made throughout this region in human Hyal2 to convert the residues to those in the mouse sequence were made and tested. None of the changes reduced human Hyal2 activity more than 10-fold, but a combination of three of the most important mutations reduced the activity by 17-fold. We conclude that the difference in receptor activity between human and mouse Hyal2 is explained by the effects of multiple amino acid changes acting together. However, the most important amino acids mapped to a small surface region of the known bee venom hyaluronidase suggesting that this is the binding site for Env.
Deletions of the amino or carboxy termini of human Hyal2 abrogate receptor activity
The carboxy end of Hyal2 is predicted to contain a GPI addition signal, and experimental data confirms the presence of the GPI linkage of Hyal2 to the cell surface . Complete removal of the GPI addition signal (amino acids 440–473) or only the GPI addition site (amino acids 440–453) abrogated the receptor activity of human Hyal2 . We have not explored whether internal deletions upstream of the GPI signal are compatible with receptor function or whether other membrane linkages might yield functional receptors, but it is clear that the carboxy end of Hyal2 is important for normal receptor function.
Analysis of chimeric mouse/human Hyal2 proteins localizes amino acids responsible for the difference in receptor activity to the central region of the proteins
Analysis of human Hyal2 mutants containing single and double mouse amino acid substitutions shows cooperative effects of the changes
Interestingly, the mutation R301G increased the receptor activity of human Hyal2 by 65%. A glycine is present at this position in both the mouse and the sheep receptor. The sheep receptor functions better as a JSRV receptor than does the human receptor in previous assays , suggesting that this glycine is important to achieve the highest receptor activity.
We also made several changes to mouse Hyal2 to see if its activity could be increased by replacement of mouse amino acids with the corresponding amino acids from human Hyal2. Based on the results above, we made the mutations F327L, N189E, and both mutations together in mouse Hyal2. All of these mutants had receptor activities (0.2% of human Hyal2, means of 2 experiments for each construct) similar to mouse Hyal2 (0.1%). The result for the N189E mutation is particularly interesting since replacement of the second domain of mouse Hyal2 with that of human Hyal2 restores nearly full receptor activity (construct mhmm in Fig. 2), and none of the human Hyal2 mutants containing single and double replacements with the corresponding mouse amino acids significantly reduced human Hyal2 activity besides the E189N replacement (Fig. 3). Again, these results indicate that alteration of receptor activity is due to multiple amino acid changes acting in concert.
Mouse Hyal2 accumulates at the cell surface at levels at least as high as those of human Hyal2
Mapping of human Hyal2 amino acids important for receptor function to the bee venom hyaluronidase crystal structure indicates a potential site for Hyal2/Env interaction
Of the seven human Hyal2 amino acids that when mutated to the mouse sequence showed the largest negative effects (colored red in Fig. 6), 4 mapped to one surface patch on Hyal2 (E307, M308, D345, and R349) (Fig. 6, left structure), two mapped under this patch (A322 and L327) (Fig. 6, right structure), and one mapped to a distant surface site (E189) (Fig. 6, right structure). Two of the residues in the surface patch are on the same loop of a helix (E307 and M308) while the other two are on adjacent loops of another helix (D345 and R349). The L327F mutation in human Hyal2 that results in a 5-fold reduction in Hyal2 receptor activity lies just underneath this surface patch with the side chain of this amino acid pointing toward the patch, and the increased size of phenylalanine compared to leucine might cause movement of the 2 helices that compose the surface patch resulting in loss of receptor activity. It is more difficult to explain the 9-fold reduction of Hyal2 receptor activity caused by the A322S mutation, but this amino acid also maps close to the surface patch and may cause distortion of the patch. This residue is unlikely to be involved directly in Env binding, even though it has a surface localization, because additional amino acids at the N- and C- termini of Hyal2 that cannot be mapped to the Hya structural model would likely interfere with Env binding to this amino acid residue (see Fig. 6, right structure).
The human Hyal2 amino acid that when mutated to the mouse sequence showed the largest positive effect (R301, colored green in Fig. 6, middle structure) maps to the surface near the active site of Hya and could interact with Env, while E189 is distant from the proposed binding patch (Fig. 6, right structure). Thus a plausible binding site for Env that involves many of the important amino acid residues in human Hyal2 is predicted by this analysis.
Comparison of mammalian Hyal2 sequences show strong pressure to conserve the protein sequence but no evidence for selective pressure to avoid JSRV infection
Viruses bearing the JSRV Env protein can infect cells from multiple mammalian species, and we considered the possibility that there might be selection during the evolution of mammals for Hyal2 variants that cannot act as receptors for these viruses. Evidence for such selection has been found for other proteins involved in virus defense, including APOBEC3G and TRIM5α [11, 12]. The location of positively selected amino acid variations might provide additional data regarding the Env binding site on Hyal2. In particular, we previously found that bovine Hyal2 acts as a weak receptor for JSRV and ENTV compared to sheep or human Hyal2, but bovine Hyal2 is much more closely related to sheep Hyal2 than to human Hyal2, suggesting that there might be specific selection in cattle for resistance to transmission of JSRV or ENTV from sheep and goats .
Positive selection for alteration in protein function can be detected by measuring the rate of non-synonymous (Ka) versus synonymous (Ks) nucleotide changes in DNA during species evolution. Ka/Ks = 1 indicates a lack of selection for or against the protein sequence, Ka/Ks < 1 indicates negative selection, or selection for amino acid conservation, and Ka/Ks > 1 indicates positive selection for protein alteration. We examined the Ka/Ks ratios for pairwise combinations of Hyal2 proteins from sheep, cattle, pig, dog, human, mouse, and rat (see Additional files 1, 2, 3 for DNA sequences used and amino acid alignment) by using the K-estimator 6.1 program created by JM Comeron . The overall Ka/Ks values ranged from 0.11 to 0.23, indicating strong selection for amino acid conservation in Hyal2 proteins from these species. We also analyzed Ka/Ks ratios using a sliding window approach to detect particular regions of Hyal2 that might be under positive (diversifying) selection, but again found no evidence for such selection (data not shown).
Strong selection for amino acid conservation can mask small numbers of amino acids undergoing positive selection for altered protein function, so we looked more closely for amino acid selection at the single amino acid level. Analysis for selection at individual amino acid sites using the Single Likelihood Ancestor Counting (SLAC) program [14–16] identified 48 negatively-selected (conserved) sites and no positively-selected sites at the 0.1 level of significance. Further analysis using the Random Effects (REL) program [14–16], that provides a less conservative analysis for selection at individual amino acid sites, found 35 negatively-selected (conserved) sites and no positively-selected sites at the default program significance setting of 50. Together these results provide strong evidence for conservation of Hyal2 protein sequence but no evidence for positive selection to resist virus infection.
We have found that the difference in activity of human and mouse Hyal2 as entry receptors for viruses bearing the JSRV Env are determined by multiple amino acid differences between the two proteins. This is in contrast to results of a similar analysis of the activity of human and mouse CAT1 as entry receptors for ecotropic murine leukemia viruses. In the latter case, mouse CAT1 is active while human CAT1 is inactive as a receptor. Substitution of only one amino acid in the mouse receptor with the corresponding amino acid from the human receptor was enough to abrogate receptor activity, and substitution of 2 amino acid residues in human CAT1 with those of mouse CAT1 could convert human CAT1 into a functional receptor . In other examples, amino acid changes resulting in receptor glycosylation were found to be responsible for the lack of receptor activity in receptor orthologs from cells that were resistant to virus infection [18, 19]. In the case of mouse and human Hyal2 there are no predicted N- or O-linked glycosylation sites in the middle region of these proteins that we found was responsible for the difference in receptor activity, indicating that the difference in receptor activity is not due to a difference in glycosylation.
We used amino-terminal Flag tags to show that the mouse and human Hyal2 proteins were both processed to the cell surface. Levels of mouse Hyal2 were somewhat higher than those of human Hyal2, indicating that the lack of mouse Hyal2 receptor activity was not due to poor processing of the protein to the cell surface. We did not perform this analysis for the mouse/human Hyal2 chimeras or for the mutant Hyal2 proteins, and some proportion of reduced receptor activity we detected in some constructs could be due to improper folding and or processing of the proteins to the cell surface. We attempted to measure cell-surface processing of mouse and human Hyal2 at 2 days after Hyal2 plasmid transfection, at the same time we assay for JSRV vector transduction, but the cell-surface Flag signal was too low to allow reliable quantitation. Selection for cells expressing the transfected plasmid was required for Flag detection, and even then the signal was heterogeneous (Fig. 4). In retrospect, a better way to perform these studies might have been to Flag-tag all of the Hyal2 constructs and insert these cDNAs into a retroviral vector, to generate virus using packaging cell lines, to transduce the target cells at a multiplicity of infection such that each cell received one vector copy, and to select the cells for the presence of the vector. These selected populations would then stably express the Hyal2 constructs and could then have been assayed for susceptibility to JSRV vector transduction and for cell-surface Flag expression to provide more accurate data on cell-surface protein processing and receptor activity. Regardless, results presented here allow us to draw some strong conclusions about the determinants of Hyal2 receptor function. For example, if any of the human Hyal2 mutants containing the corresponding mouse Hyal2 residues had no activity, it would have been important to show that this result was not due simply to a processing defect. However, all mutants had activity well above the mouse Hyal2 level, and were thus functional to some extent, allowing us to conclude that multiple amino acid differences contribute to the low activity of mouse Hyal2 in comparison to that of human Hyal2.
We have mapped four of the amino acid residues that are important for human Hyal2 function as a virus receptor to a surface patch on the bee venom Hya crystal structure, and two others map to positions underneath this patch and could influence the structure of the patch. We hypothesize that this patch represents the binding site for Env, but it is also possible that this site is required for Hyal2 binding to another protein involved in virus entry. Furthermore, this structural extrapolation to the bee venom hyaluronidase may not be entirely valid and determination of a crystal structure for mouse or human Hyal2 is important to confirm these predictions. A soluble form of human Hyal2 has been made that can bind tightly to JSRV Env, as measured by surface plasmon resonance analysis, and that can block JSRV vector transduction by binding to JSRV vector virions . We are currently working to crystallize this protein to allow more definitive analysis of results presented here.
We show that the greatly reduced receptor activity of mouse Hyal2 in comparison to that of human Hyal2 is determined by multiple amino acid changes acting in concert, and that no one amino acid change blocks infection. However, the most important amino acids map to a small patch on a predicted 3-dimensional Hyal2 structure which we hypothesize is the binding site for Env.
Cell culture and virus production
NIH 3T3 (TK-) Swiss mouse embryo fibroblasts , 208F Fischer rat embryo fibroblasts , and PJ4/LAPSN vector-producing cells  were grown in Dulbecco's modified Eagle medium (DMEM) with high glucose (4.5 g/L) and 7% fetal bovine serum at 37°C in a 10% CO2-air atmosphere at 100% relative humidity. PJ4/LAPSN cells produce virus containing the LAPSN retroviral vector  that encodes human placental alkaline phosphatase (AP) and neomycin phosphotransferase (Neo). Virus produced by PJ4/LAPSN cells contain Moloney murine leukemia virus Gag-Pol proteins and have a JSRV Env protein coat. LAPSN(PJ4) vector was produced by feeding confluent PJ4/LAPSN cells and harvesting the medium 12 to 24 h later. Virus-containing medium was centrifuged at 3,000 × g for 5 min to remove cells and debris and was frozen at -70°C until use.
All Hyal2 plasmids tested were constructed in either the pFLAG-CMV-1 mammalian expression vector (Sigma-Aldrich, Saint Louis, MO) or the pCR3.1 TOPO eukaryotic expression vector (Invitrogen, Carlsbad, CA). Each of the human Hyal2 amino or carboxy terminus deleted plasmids was cloned into the EcoRI site of the pFLAG-CMV-1 vector after PCR amplification of the desired coding regions. The human and mouse Hyal2 mutants containing single, double or triple amino acid substitutions were created by using the QuikChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA) following recommended protocols. Human and mouse Hyal2 cDNAs share two common unique restriction sites, PflMI (located at codon 158), and BsaAI (located at codon 354). The mouse Hyal2 cDNA contains 3 BlpI restriction sites (located near codons 305, 321, and 413), which are completely absent in human Hyal2. By performing site-directed mutagenesis without altering the encoded amino acids, a BlpI site near codon 305 was created in human Hyal2 and the BlpI site near codon 321 was eliminated in mouse Hyal2. Both the human and mouse Hyal2 can therefore be divided into 4 domains by the 3 restriction sites: PflMI (codon 158), BlpI (codon 305), and BsaAI (codon 354). The chimeric mouse/human Hyal2 plasmids were obtained by exchanging restriction fragments between human and mouse Hyal2. The Hyal2 coding regions of all plasmids were sequenced after construction to confirm the expected sequences.
Receptor activity assay
For assay, NIH 3T3 mouse cells or 208F rat cells were seeded at 2.5 × 105 per well (d = 3.5 cm) of 6-well plates. The next day the cells were transfected with 4.5 μg of the test Hyal2 plasmid plus 0.5 μg of a plasmid encoding β-galactosidase (β-gal) per well by calcium phosphate coprecipitation as previously described . One day after transfection the cells were trypsinized and divided 1:6 into 6-well plates. Two days after transfection the cells were exposed to serial dilutions of LAPSN(PJ4) vector in the presence of 4 μg/ml Polybrene. Four days after transfection the cells exposed to the LAPSN(PJ4) vector were stained for AP to determine the vector titer and cells not exposed to the vector were stained for β-gal expression to evaluate transfection efficiency for each plasmid. Experiments for which β-gal staining was low or variable were not included in the results. In each experiment the LAPSN(PJ4) vector titer was measured on cells transfected with an unmodified human Hyal2 expression plasmid, and results for the test plasmids are expressed as a percentage of that for the human Hyal2 plasmid (typically 104 to 3 × 104 AP+ focus-forming units (FFU) per ml following transfection of NIH 3T3 mouse cells and 103 to 2 × 103 FFU/ml following transfection of 208F rat cells with the human Hyal2 plasmid). Several Hyal2 constructs were tested in both mouse and rat cells with similar results, but most of the receptor activity assays shown were performed only in the mouse cells. Results are means of at least two independent experiments.
Cell-surface expression of amino-terminal Flag-tagged Hyal2 proteins was evaluated by FACS analysis after incubation of cells with anti-Flag mouse monoclonal primary antibody followed by incubation in goat anti-mouse IgG secondary antibody. Propidium iodide was added to the cells before analysis and only cells that excluded the dye (live cells) were included in the analysis. 30,000 live cells were analyzed per sample.
We thank Neal Van Hoeven for assistance with the FACS analysis, Roland Strong, Vladimir Vigdorovich, and Shervin Bahrami for help with the Hyal2 structural analysis, and Harmit Malik for help with the protein evolution analysis. CD and ADM were funded by NIH grants DK47754, HL36444 and HL66947. FMD and MIL were funded by the National Cancer Institute, NIH under contracts No. NO1-CO-56000 and NO1-CO-12400, and by the Intramural Research Program of the NIH, NCI. The content of the publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
- Fan H: Jaagsiekte sheep retrovirus and lung cancer. Curr Top Microbiol Immunol. 2003, Berlin , Springer, 275: 1-248.Google Scholar
- Miller AD: Identification of Hyal2 as the cell-surface receptor for jaagsiekte sheep retrovirus and ovine nasal adenocarcinoma virus. Curr Top Microbiol Immunol. 2003, 275: 179-199.PubMedGoogle Scholar
- Vigdorovich V, Strong RK, Miller AD: Expression and characterization of a soluble, active form of the jaagsiekte sheep retrovirus receptor, Hyal2. J Virol. 2005, 79 (1): 79-86. 10.1128/JVI.79.1.79-86.2005.PubMed CentralView ArticlePubMedGoogle Scholar
- Rai SK, DeMartini JC, Miller AD: Retrovirus vectors bearing jaagsiekte sheep retrovirus Env transduce human cells by using a new receptor localized to chromosome 3p21.3. J Virol. 2000, 74 (10): 4698-4704. 10.1128/JVI.74.10.4698-4704.2000.PubMed CentralView ArticlePubMedGoogle Scholar
- Rai SK, Duh FM, Vigdorovich V, Danilkovitch-Miagkova A, Lerman MI, Miller AD: Candidate tumor suppressor HYAL2 is a glycosylphosphatidylinositol (GPI)-anchored cell-surface receptor for jaagsiekte sheep retrovirus, the envelope protein of which mediates oncogenic transformation. Proc Natl Acad Sci USA. 2001, 98 (8): 4443-4448. 10.1073/pnas.071572898.PubMed CentralView ArticlePubMedGoogle Scholar
- Dirks C, Duh FM, Rai SK, Lerman MI, Miller AD: Mechanism of cell entry and transformation by enzootic nasal tumor virus. J Virol. 2002, 76 (5): 2141-2149. 10.1128/jvi.76.5.2141-2149.2002.PubMed CentralView ArticlePubMedGoogle Scholar
- Alberti A, Murgia C, Liu SL, Mura M, Cousens C, Sharp M, Miller AD, Palmarini M: Envelope-induced cell transformation by ovine betaretroviruses. J Virol. 2002, 76 (11): 5387-5394. 10.1128/JVI.76.11.5387-5394.2002.PubMed CentralView ArticlePubMedGoogle Scholar
- Liu SL, Duh FM, Lerman MI, Miller AD: Role of virus receptor Hyal2 in oncogenic transformation of rodent fibroblasts by sheep betaretrovirus env proteins. J Virol. 2003, 77 (5): 2850-2858. 10.1128/JVI.77.5.2850-2858.2003.PubMed CentralView ArticlePubMedGoogle Scholar
- Van Hoeven NS, Miller AD: Improved enzootic nasal tumor virus pseudotype packaging cell lines reveal virus entry requirements in addition to the primary receptor Hyal2. J Virol. 2005, 79 (1): 87-94. 10.1128/JVI.79.1.87-94.2005.PubMed CentralView ArticlePubMedGoogle Scholar
- Markovic-Housley Z, Miglierini G, Soldatova L, Rizkallah PJ, Muller U, Schirmer T: Crystal structure of hyaluronidase, a major allergen of bee venom. Structure Fold Des. 2000, 8 (10): 1025-1035. 10.1016/S0969-2126(00)00511-6.View ArticlePubMedGoogle Scholar
- Sawyer SL, Emerman M, Malik HS: Ancient adaptive evolution of the primate antiviral DNA-editing enzyme APOBEC3G. PLoS Biol. 2004, 2 (9): E275-10.1371/journal.pbio.0020275.PubMed CentralView ArticlePubMedGoogle Scholar
- Sawyer SL, Wu LI, Emerman M, Malik HS: Positive selection of primate TRIM5alpha identifies a critical species-specific retroviral restriction domain. Proc Natl Acad Sci U S A. 2005, 102 (8): 2832-2837. 10.1073/pnas.0409853102.PubMed CentralView ArticlePubMedGoogle Scholar
- Comeron JM: K-Estimator: calculation of the number of nucleotide substitutions per site and the confidence intervals. Bioinformatics. 1999, 15 (9): 763-764. 10.1093/bioinformatics/15.9.763.View ArticlePubMedGoogle Scholar
- Pond SL, Frost SD: Not so different after all: a comparison of methods for detecting amino acid sites under selection. Mol Biol Evol. 2005, 22 (5): 1208-1222. 10.1093/molbev/msi105.View ArticleGoogle Scholar
- Pond SL, Frost SD: Datamonkey: rapid detection of selective pressure on individual sites of codon alignments. Bioinformatics. 2005, 21 (10): 2531-2533. 10.1093/bioinformatics/bti320.View ArticlePubMedGoogle Scholar
- Pond SL, Frost SD: Fast Positive and Negative Selection Detection. [http://www.datamonkey.org/]
- Yoshimoto T, Yoshimoto E, Meruelo D: Identification of amino acid residues critical for infection with ecotropic murine leukemia retrovirus. J Virol. 1993, 67 (3): 1310-1314.PubMed CentralPubMedGoogle Scholar
- Eiden MV, Farrell K, Wilson CA: Glycosylation-dependent inactivation of the ecotropic murine leukemia virus receptor. J Virol. 1994, 68 (2): 626-631.PubMed CentralPubMedGoogle Scholar
- Marin M, Lavillette D, Kelly SM, Kabat D: N-linked glycosylation and sequence changes in a critical negative control region of the ASCT1 and ASCT2 neutral amino acid transporters determine their retroviral receptor functions. J Virol. 2003, 77 (5): 2936-2945. 10.1128/JVI.77.5.2936-2945.2003.PubMed CentralView ArticlePubMedGoogle Scholar
- Wei CM, Gibson M, Spear PG, Scolnick EM: Construction and isolation of a transmissible retrovirus containing the src gene of Harvey murine sarcoma virus and the thymidine kinase gene of herpes simplex virus type 1. J Virol. 1981, 39 (3): 935-944.PubMed CentralPubMedGoogle Scholar
- Quade K: Transformation of mammalian cells by avian myelocytomatosis virus and avian erythroblastosis virus. Virology. 1979, 98 (2): 461-465. 10.1016/0042-6822(79)90569-5.View ArticlePubMedGoogle Scholar
- Miller DG, Edwards RH, Miller AD: Cloning of the cellular receptor for amphotropic murine retroviruses reveals homology to that for gibbon ape leukemia virus. Proc Natl Acad Sci USA. 1994, 91 (1): 78-82.PubMed CentralView ArticlePubMedGoogle Scholar
- Thompson JD, Higgins DG, Gibson TJ: CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 1994, 22 (22): 4673-4680.PubMed CentralView ArticlePubMedGoogle Scholar
- Delano WL: The PyMOL Molecular Graphics System. [http://www.pymol.org]
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.