Processing of N-linked oligosaccharides from precursor- to mature-form herpes simplex virus type 1 glycoprotein gC.

Immature and mature forms of glycoprotein gC were purified by immunoadsorbent from herpes simplex virus type 1-infected BHK cells labeled with [3H]mannose for a 20-min pulse or for 11 h followed by a 3-h chase. The nature of N-asparagine-linked oligosaccharides carried by the immature form, pgC (molecular weight = 92,000), and the mature gC (molecular weight = 120,000) has been investigated. All pronase-digested glycopeptides of pgC were susceptible to endo-beta-N-acetylglucosaminidase H treatment; thus they have a high-mannose structure. Using thin-layer chromatography to separate endo-beta-N-acetylglucosaminidase H-cleaved oligosaccharides, polymannosyl chains of different sizes, ranging from Man9GlcNAc to Man5GlcNAc, were separated. The major components were Man8GlcNAc and Man7GlcNAc, suggesting that pgC labeled in a 20-min pulse represents the form of glycoprotein already routed to the Golgi apparatus. Analysis of glycopeptides of mature gC showed that the majority (95%) of N-linked glycans were converted to complex-type glycans. Ion-exchange chromatography and affinity chromatography on concanavalin A-Sepharose and leucoagglutinin-agarose revealed that diantennary and triantennary glycans predominated, whereas tetrantennary chains were not present. Parts of the di- and triantennary chains were not fully sialylated. The high heterogeneity of complex-type chains found in mature gC may be related to the high number of N-glycosylation sites of the glycoprotein as predicted by DNA sequencing studies (Frink et al., J. Virol. 45:634-647, 1983).     

Characterization of a herpes simplex virus type 2 75,000-molecular-weight glycoprotein antigenically related to herpes simplex virus type 1 glycoprotein C.

Evidence is presented that the herpes simplex virus type 2 glycoprotein previously designated gF is antigenically related to herpes simplex virus type 1 gC (gC-1). An antiserum prepared against type 1 virion envelope proteins immunoprecipitated gF of type 2 (gF-2), and competition experiments revealed that the anti-gC-1 component of the antiserum was responsible for the anti-gF-2 cross-reactivity. An antiserum prepared against fully denatured purified gF-2, however, and three anti-gF-2 monoclonal antibodies failed to precipitate any type 1 antigen, indicating that the extent of cross-reactivity between gC-1 and gF-2 may be limited. Several aspects of gF-2 synthesis and processing were investigated. Use of the enzymes endo-beta-N-acetylglucosaminidase H and alpha-D-N-acetylgalactosaminyl oligosaccharidase revealed that the fully processed form of gF-2 (about 75,000 [75K] apparent molecular weight) had both complex-type N-linked and O-linked oligosaccharides, whereas newly synthesized forms (67K and 69K) had only high-mannose N-linked oligosaccharides. These last two forms were both reduced in size to 54K by treatment with endo-beta-N-acetylglucosaminidase H and therefore appear to differ only in the number of N-linked chains. Neutralization tests and radioiodination experiments revealed that gF-2 is exposed on the surfaces of virions and that the 75K form of gF-2 is exposed on cell surfaces. The similarities and differences of gF-2 and gC-1 are discussed in light of recent mapping results which suggest collinearity of their respective genes.     

Extensive homology between the herpes simplex virus type 2 glycoprotein F gene and the herpes simplex virus type 1 glycoprotein C gene.

The region of the herpes simplex virus type 2 (HSV-2) genome which maps colinearly with the HSV-1 glycoprotein C (gC) gene has been cloned, and the DNA sequence of a 2.29-kilobase region has been determined. Contained within this sequence is a major open reading frame of 479 amino acids. The carboxyterminal three-fourths of the derived HSV-2 protein sequence showed a high degree of sequence homology to the HSV-1 gC amino acid sequence reported by Frink et al. (J. Virol. 45:634-647, 1983). The amino-terminal region of the HSV-2 sequence, however, showed very little sequence homology to HSV-1 gC. In addition, the HSV-1 gC sequence contained 27 amino acids in the amino-terminal region which were missing from the HSV-2 protein. Computer-assisted analysis of the hydrophilic and hydrophobic properties of the derived HSV-2 sequence demonstrated that the protein contained structures characteristic of membrane-bound glycoproteins, including an amino-terminal signal sequence and carboxy-terminal hydrophobic transmembrane domain and charged cytoplasmic anchor. The HSV-2 protein sequence also contained seven putative N-linked glycosylation sites. These data, in conjunction with mapping studies of Para et al. (J. Virol. 45:1223-1227, 1983) and Zezulak and Spear (J. Virol. 49:741-747, 1984), suggest that the protein sequence derived from the HSV-2 genome corresponds to gF, the HSV-2 homolog of HSV-1 gC.     

Characterization of baculovirus-expressed herpes simplex virus type 1 glycoprotein K.

The DNA region encoding the complete herpes simplex virus type 1 (HSV-1) glycoprotein K (gK) was inserted into a baculovirus transfer vector, and recombinant viruses expressing gK were isolated. Four gK-related recombinant baculovirus-expressed peptides of 29, 35, 38, and 40 kDa were detected with polyclonal antibody to gK. The 35-, 38-, and 40-kDa species were susceptible to tunicamycin treatment, suggesting that they were glycosylated. The 38- and 40-kDa species corresponded to partially glycosylated precursor gK (pgK) and mature gK, respectively. The 29-kDa peptide probably represented a cleaved, unglycosylated peptide. The 35-kDa peptide probably represented a cleaved, glycosylated peptide that may be a precursor to pgK. Indirect immunofluorescence with polyclonal antibody to gK peptides indicated that the recombinant baculovirus-expressed gK was abundant on the surface of the insect cells in which it was expressed. Mice vaccinated with the baculovirus-expressed gK produced very low levels (< 1:10) of HSV-1 neutralizing antibody. Nonetheless, these mice were partially protected from lethal challenge with HSV-1 (75% survival). This protection was significant (P = 0.02). Despite some protection against death, gK-vaccinated mice showed no protection against the establishment of latency. Surprisingly, gK-vaccinated mice that were challenged ocularly with a stromal disease-producing strain of HSV-1 had significantly higher levels of ocular disease (herpes stromal keratitis) than did mock-vaccinated mice. In summary, this is the first report to show that vaccination with HSV-1 gK can provide protection against lethal HSV-1 challenge and that vaccination with an HSV-1 glycoprotein can significantly increase the severity of HSV-1-induced ocular disease.     

Fusogenic domains in herpes simplex virus type 1 glycoprotein H

Infection of eukaryotic cells by enveloped viruses requires fusion between the viral envelope and the cellular plasma or endosomal membrane. The actual merging of the two membranes is mediated by viral envelope glycoproteins, which generally contain a highly hydrophobic region termed the fusion peptide. The entry of herpesviruses is mediated by three conserved proteins: glycoproteins B, H (gH), and L. However, how fusion is executed remains unknown. Herpes simplex virus type 1 gH exhibits features typical of viral fusion glycoproteins, and its ectodomain seems to contain a putative internal fusion peptide. Here, we have identified additional internal segments able to interact with membranes and to induce membrane fusion of large unilamellar vesicles. We have applied the hydrophobicity-at-interface scale proposed by Wimley and White (Wimley, W. C., and White, S. H. (1996) Nat. Struct. Biol. 3, 842-848) to identify six hydrophobic stretches within gH with a tendency to partition into the membrane interface, and four of them were able to induce membrane fusion. Experiments in which equimolar mixtures of gH peptides were used indicated that different fusogenic regions may act in a synergistic way. The functional and structural characterization of these segments suggests that herpes simplex virus type 1 gH possesses several fusogenic internal peptides that could participate in the actual fusion event.     

Localization of epitopes of herpes simplex virus type 1 glycoprotein D.

We previously defined eight groups of monoclonal antibodies which react with distinct epitopes of herpes simplex virus glycoprotein D (gD). One of these, group VII antibody, was shown to react with a type-common continuous epitope within residues 11 to 19 of the mature glycoprotein (residues 36 to 44 of the predicted sequence of gD). In the current investigation, we have localized the sites of binding of two additional antibody groups which recognize continuous epitopes of gD. The use of truncated forms of gD as well as computer predictions of secondary structure and hydrophilicity were instrumental in locating these epitopes and choosing synthetic peptides to mimic their reactivity. Group II antibodies, which are type common, react with an epitope within residues 268 to 287 of the mature glycoprotein (residues 293 to 312 of the predicted sequence). Group V antibodies, which are gD-1 specific, react with an epitope within residues 340 to 356 of the mature protein (residues 365 to 381 of the predicted sequence). Four additional groups of monoclonal antibodies appear to react with discontinuous epitopes of gD-1, since the reactivity of these antibodies was lost when the glycoprotein was denatured by reduction and alkylation. Truncated forms of gD were used to localize these four epitopes to the first 260 amino acids of the mature protein. Competition experiments were used to assess the relative positions of binding of various pairs of monoclonal antibodies. In several cases, when one antibody was bound, there was no interference with the binding of an antibody from another group, indicating that the epitopes were distinct. However, in other cases, there was competition, indicating that these epitopes might share some common amino acids.     

Characterization of oligosaccharides of highly purified glycoprotein gC of herpes simplex virus type 1 (HSV-1)

The envelope membrane glycoprotein gC of HSV-1 was purified from Triton X-100 extracts of virus-infected BHK-21 or HEp-2 cells by a single step immuno-affinity column using monoclonal anti-gC antibody. The analysis of the purified [³H]G1cN labeled glycoprotein gC (by gel filtration on Bio-Gel P4) before and after digestion with endo-β-N-acetylglucosaminidase (endo D) indicated that gC contains Asn-linked “complex type” oligosaccharides. No “high mannose” type oligosaccharides were detected. Fractionation of radio-labeled glycopeptides of gC on a column of concanavalin A-sepharose suggested that glycopeptides have “diantennary” and “triantennary” and/or “tetra antennary” structures. Tunicamycin inhibited the incorporation of [¹⁴C]GalN or [³H]GlcN into gC in HSV-1 infected BHK-21 or HEp-2 cells. Gel filtration analysis of [³H]GlcN labeled gC following β-elimination reaction failed to indicate O-glycosidically linked oligosaccharides.     

Cross-linking of glycoprotein oligomers during herpes simplex virus type 1 entry.

Herpes simplex virus (HSV) has 10 glycoproteins in its envelope. Glycoprotein B (gB), gC, gD, gH, and gL have been implicated in virus entry. We previously used chemical cross-linking to show that these five glycoproteins were close enough to each other to be cross-linked into homodimeric and hetero-oligomeric forms; hetero-oligomers of gB-gC, gC-gD, gD-gB, gH-gL, gC-gL and gD-gL were found in purified virions. To better understand the roles of these glycoproteins in viral entry, we have modified a standard HSV penetration assay to include cross-linkers. This allowed us to examine changes in associations of viral glycoproteins during the entry process. HSV-1(KOS) was adsorbed at 4 degrees C to human neuroblastoma cells (SY5Y). The temperature was raised to 37 degrees C and cells were treated with cross-linker at various times after the temperature shift. Cytoplasmic extracts were examined by Western blotting (immunoblotting) for viral glycoproteins. We found that (i) as in virus alone, the length and concentration of the cross-linking agent affected the number of specific complexes isolated; (ii) the same glycoprotein patterns found in purified virions were also present after attachment of virions to cells; and (iii) the ability to cross-link HSV glycoproteins changed as virus penetration proceeded, e.g., gB and gD complexes which were present during attachment disappeared with increasing time, and their disappearance paralleled the kinetics of penetration. However, this phenomenon appeared to be selective since it was not observed with gC oligomers. In addition, we examined the cross-linking patterns of gB and gD in null viruses K082 and KOSgD beta. Neither of these mutants, which attach but cannot penetrate, showed changes in glycoprotein cross-linking over time. We speculate that these changes are due to conformational changes which preclude cross-linking or spatial alterations which dissociate the glycoprotein interactions during the penetration events.     

Temporal aspects of O-glycosylation of glycoprotein C from herpes simplex virus type-1.

Herpes simplex virus type-1 glycoprotein C (gC1) contains several O-linked oligosaccharides clustered near N-linked chains, and Pronase digestion produces glycopeptides carrying both oligosaccharide types. We have taken advantage of this fact to investigate the temporal relationship between the initiation of O-linked chains and the processing of N-linked oligosaccharides. gC1 was isolated from herpes-simplex-virus-infected BHK (baby-hamster kidney) cells after short labelling periods with [3H]glucosamine, and the labelled Pronase-cleaved glycopeptides fractionated on concanavalin A-Sepharose. N-[3H]Acetylgalactosamine, mostly convertible into free N-[3H]acetylgalactosaminitol on mild alkaline-borohydride treatment, was found in glycopeptides with an affinity to concanavalin A-Sepharose corresponding to that of glycopeptides carrying Man8GlcNAc2 or larger N-linked chains. Since there is evidence that the processing of N-linked chains up to Man8GlcNAc2 involves enzymes located in the rough endoplasmic reticulum, current results strongly suggest that gC1 acquires O-linked N-acetylgalactosamine before the glycoprotein routing to the Golgi apparatus. The addition of the second sugar to the nascent O-linked chain appeared to occur after a relatively long lag time.     

Release of a virus coded glycoprotein from herpes simplex virus type 1 infected cells

HEp-2 cells, which were infected with HSV-1, excrete besides other proteins a soluble glycoprotein (Mr 125000–130000) related to the virus protein gC. The excretion of the glycoprotein and the production of extracellular virus particles is reduced to a similar extent when the cells were treated with monensin. Possible consequences of the excretion of soluble viral proteins to a modulation of the immune response are discussed.Abbreviations HSV-1 Herpes simplex virus type 1 - PAGE Polyacrylamide gel electrophoresis - SDS Sodium dodecylsulfate     

Identification of C3b-binding regions on herpes simplex virus type 2 glycoprotein C.

Glycoprotein C from herpes simplex viruses types 1 and 2 (gC-1 and gC-2) acts as a receptor for the C3b fragment of the third component of complement. Our goal is to identify domains on gC involved in C3b receptor activity. Here, we used in-frame linker-insertion mutagenesis of the cloned gene for gC-2 to identify regions of the protein involved in C3b binding. We constructed 41 mutants of gC-2, each having a single, double, or triple insertion of four amino acids at sites spread across the protein. A transient transfection assay was used to characterize the expressed mutant proteins. All of the proteins were expressed on the transfected cell surface, exhibited processing of N-linked oligosaccharides, and bound one or more monoclonal antibodies recognizing distinct antigenic sites on native gC-2. This suggested that each of the mutant proteins was folded into a native structure and that a loss of C3b binding by any of the mutants could be attributed to the disruption of a specific functional domain. When the panel of insertion mutants was assayed for C3b receptor activity, we identified three distinct regions that are important for C3b binding, since an insertion within those regions abolished C3b receptor activity. Region I was located between amino acids 102 and 107, region II was located between residues 222 and 279, and region III was located between residues 307 and 379. In addition, region III has some structural features similar to a conserved motif found in complement receptor 1, the human C3b receptor. Finally, blocking experiments indicated that gC-1 and gC-2 bind to similar locations on the C3b molecule.     

Oligomer formation of the gB glycoprotein of herpes simplex virus type 1.

Oligomer formation of the gB glycoprotein of herpes simplex virus type 1 was studied by sedimentation analysis of radioactively labeled infected cell and virion lysates. Fractions from sucrose gradients were precipitated with a pool of gB-specific monoclonal antibodies and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Pulse-labeled gB from infected cell was synthesized as monomers and converted to oligomers posttranslationally. The oligomers from infected cells and from virions sedimented as dimers, and there was no evidence of higher-molecular-weight forms. To identify amino acid sequences of gB that contribute to oligomer formation, pairs of mutant plasmids were transfected into Vero cells and superinfected with a gB-null mutant virus to stimulate plasmid-specified gene expression. Radioactively labeled lysates were precipitated with antibodies and examined by SDS-PAGE. Polypeptides from cotransfections were precipitated with an antibody that recognized amino acid sequences present in only one of the two polypeptides. A coprecipitated polypeptide lacking the antibody target epitope was presumed to contain the sequences necessary for oligomer formation. Using this technique, two noncontiguous sites for oligomer formation were detected. An upstream site was localized between residues 93 and 282, and a downstream site was localized between residues 596 and 711. Oligomer formation resulted from molecular interactions between two upstream sites, between two downstream sites, and between an upstream and a downstream site. A schematic diagram of a gB oligomer is presented that is consistent with these data.     

Detailed analysis of the portion of the herpes simplex virus type 1 genome encoding glycoprotein C.

We previously showed that the right third of HindIII fragment L (0.59 to 0.65) of herpes simplex virus type 1 (HSV-1) encodes a family of mRNAs some members of which appear to be related by splicing. In the experiments described in this communication, we determined the nucleotide sequence of the DNA encoding this mRNA family and precisely located the mRNAs associated with this DNA sequence. The major mRNA species is unspliced and encoded by a 2.520-nucleotide region. Just upstream of the 5' end are TATA and CAT box sequences characteristic of HSV-1 promoters. The 3' end maps near a region containing a nominal polyadenylation signal. Three minor species (2,400, 2,200, and 1,900 bases, respectively) appear to share a very short leader sequence with the 5' end of the major mRNA and are then encoded by uninterrupted DNA sequences beginning about 100, 400, and 625 bases downstream of the 5' end of the major unspliced mRNA. These positions map at or very near positions which agree reasonably well with consensus splice acceptor sequences. The fourth mRNA is encoded by a contiguous 730-nucleotide sequence at the 3' end of the major unspliced mRNA and has its 5' end just downstream of recognizable TATA and CAT box sequences. We suggest that this mRNA is controlled by its own promoter. The nucleotide sequence data, in combination with the mRNA localization, demonstrate four potential polypeptides encoded by the region. The largest is 1,569 bases long and defines a 523-amino acid protein with sequence features characteristic of a glycoprotein. This was confirmed to be HSV-1 glycoprotein C by immune precipitation of the in vitro translation product of the major unspliced mRNA, performed with a polyspecific antibody to HSV-1 envelope glycoproteins (anti-env-1 serum), and by comparison of tryptic peptides of this translation product with those of authentic HSV-1 glycoprotein C. Polypeptides encoded by some of the minor species also were tentatively identified.     

Antibodies to a synthetic oligopeptide that react with herpes simplex virus type 1 and 2 glycoprotein C

Nucleotide sequence and mRNA localization studies have allowed the prediction of the amino acid sequence of herpes simplex virus type 1 (HSV-1) glycoprotein C (gC). We immunized a rabbit with a conjugate of bovine serum albumin and a synthetic peptide having the same sequence as that deduced for amino acids 128 through 139 of HSV-1 gC. A very similar amino acid sequence has been predicted to exist in the related product, herpes simplex virus type 2 (HSV-2) gC, which was formerly designated gF. Preparations of crude antiserum and immunoaffinity-purified antibodies were obtained and shown to react in enzyme-linked immunosorbent assays with purified HSV-1 gC and HSV-2 gC. Although these antibodies did not detectably immunoprecipitate proteins from radiolabeled infected cell extracts, they reacted with HSV-1 gC and HSV-2 gC that were electrophoretically transferred to nitrocellulose membranes from polyacrylamide gels. These results confirm that HSV-1 gC and HSV-2 gC are immunologically related and also define a specific portion of HSV-1 gC that is conserved.     

Pathogenicity in mice of herpes simplex virus type 2 mutants unable to express glycoprotein C.

Herpes simplex virus type 2 (HSV-2) mutants that were unable to express glycoprotein C (gC-2) were isolated. Deletions were made in a cloned copy of the gC-2 gene, and recombinant viruses containing these deletions were screened by using an immunoreactive plaque selection protocol. The viruses did not display a syncytial phenotype. Intravaginal inoculation of BALB/cJ mice with one of the HSV-2 gC-2- viruses produced local inflammation followed by a lethal spread of the viral infection into the nervous system in a manner identical to that produced by parental HSV-2 strain 333. Similarly, intracerebral inoculation of DBA-2 mice with the gC-2- virus produced a lethal neurological disease paralleling that caused by HSV-2 strain 333. These results indicate that gC-2 is not required for the spread of HSV-2 infections in mice.     

The cytoplasmic domain of herpes simplex virus type 1 glycoprotein C is required for membrane anchoring.

The herpes simplex virus type 1 (HSV-1) glycoprotein C (gC) gene was altered so that it encoded a truncated glycoprotein lacking a cytoplasmic domain but retaining 20 of 23 amino acids of the transmembrane domain. No additional amino acid residues were introduced into the glycoprotein encoded by the altered gene. The gene was recombined into the HSV-1 genome by marker transfer. Two recombinant viruses, dl1 and dl2, that expressed the mutant gene were isolated. Characterization of these viruses showed that a substantial fraction of the mutant glycoprotein was secreted from infected cells. Pulse-chase experiments showed that the kinetics of posttranslational modification of the mutant glycoprotein were similar to those of the wild type. However, comparison of the kinetics of secretion of gC by dl2 and gC-3, a gC mutant lacking both the transmembrane and cytoplasmic domains, showed that dl2 gC was secreted much more slowly than gC-3 gC. Iodination of plasma membrane glycoproteins showed that dl2 gC was initially expressed on the cell surface as a membrane protein and subsequently was slowly released from the membrane into the medium. These data indicate that a major function of the cytoplasmic domain of gC is to ensure the stable anchoring of the glycoprotein in plasma membranes. In contrast to these major changes in the membrane-anchoring properties of gC, characterization of the virions produced by dl1 and dl2 showed that they contain significant amounts of gC. Thus the cytoplasmic domain does not appear to be essential for incorporation of this glycoprotein into virions.     

Absence of asparagine-linked oligosaccharides from glycoprotein D of herpes simplex virus type 1 results in a structurally altered but biologically active protein.

Glycoprotein D (gD) of herpes simplex virus contains three utilized sites (Asn-X-Ser/Thr) for addition of asparagine-linked carbohydrates (N-CHO). Previously, we used oligonucleotide-directed mutagenesis to alter serine or threonine residues to alanine at each N-CHO addition site. Studies with monoclonal antibodies showed that a mutant protein lacking all three sites (now designated AAA) was structurally altered because of the amino acid change at residue 96 as well as the absence of the N-CHO. In this study, we constructed additional single mutations at site 1 (residues 94 and 96) and found that in most cases, the amino acid change itself adversely affected the conformation of gD. However, changing asparagine 94 to glutamine (Q) at site 1 had the least effect on gD. We constructed a second triple mutant, QAA, which lacked all three N-CHO signals. The antigenic conformation of QAA was similar to that of gD produced in the presence of tunicamycin (TM-gD). However, binding of MAbs to the AAA protein or to single mutants altered at site 1 was reduced compared with TM-gD. Wild-type gD and QAA proteins were equally susceptible to digestion by trypsin or Staphylococcus aureus V8 protease. In contrast, the AAA protein was more sensitive to trypsin but less sensitive to V8, again suggesting conformational alterations of the AAA protein. Despite what appeared to be large changes in structure, each mutant complemented the infectivity of a virus lacking gD (F-gD beta). We conclude that the N-CHO and amino acids at N-CHO site 1 play an important role in forming and/or maintaining gD structure, but none of the N-CHO are required for gD to function in the complementation assay.     

Disulfide bonds of herpes simplex virus type 2 glycoprotein gB.

Glycoprotein B (gB) is the most highly conserved envelope glycoprotein of herpesviruses. The gB protein is required for virus infectivity and cell penetration. Recombinant forms of gB being used for the development of subunit vaccines are able to induce virus-neutralizing antibodies and protective efficacy in animal models. To gain structural information about the protein, we have determined the location of the disulfide bonds of a 696-amino-acid residue truncated, recombinant form of herpes simplex virus type 2 glycoprotein gB (HSV gB2t) produced by expression in Chinese hamster ovary cells. The purified protein, which contains virtually the entire extracellular domain of herpes simplex virus type 2 gB, was digested with trypsin under nonreducing conditions, and peptides were isolated by reversed-phase high-performance liquid chromatography (HPLC). The peptides were characterized by using mass spectrometry and amino acid sequence analysis. The conditions of cleavage (4 M urea, pH 7) induced partial carbamylation of the N termini of the peptides, and each disulfide peptide was found with two or three different HPLC retention times (peptides with and without carbamylation of either one or both N termini). The 10 cysteines of the molecule were found to be involved in disulfide bridges. These bonds were located between Cys-89 (C1) and Cys-548 (C8), Cys-106 (C2) and Cys-504 (C7), Cys-180 (C3) and Cys-244 (C4), Cys-337 (C5) and Cys-385 (C6), and Cys-571 (C9) and Cys-608 (C10). These disulfide bonds are anticipated to be similar in the corresponding gBs from other herpesviruses because the 10 cysteines listed above are always conserved in the corresponding protein sequences.