Virus Genes

, 37:31

Molecular characterization of the genes coding for glycoprotein and L protein of lymphocytic choriomeningitis virus strain MX


    • Institute of VirologySlovak Academy of Sciences
  • Martina Labudova
    • Institute of VirologySlovak Academy of Sciences
  • Juraj Kopacek
    • Institute of VirologySlovak Academy of Sciences
  • Silvia Pastorekova
    • Institute of VirologySlovak Academy of Sciences
  • Jaromir Pastorek
    • Institute of VirologySlovak Academy of Sciences

DOI: 10.1007/s11262-008-0240-2

Cite this article as:
Tomaskova, J., Labudova, M., Kopacek, J. et al. Virus Genes (2008) 37: 31. doi:10.1007/s11262-008-0240-2


Lymphocytic choriomeningitis virus (LCMV) is the prototype Arenavirus with ambisense coding strategy. We have previously described a new MX strain LCMV and determined the primary structure of the genes coding for the nucleoprotein and RING finger Z protein. In this report, we describe amplification and sequencing of the entire coding sequences of additional MX genes, the glycoprotein precursor (GPC) and L protein. The obtained MX GPC cDNA sequence was 1,615 nucleotides long and contained an ORF, which encodes the GPC precursor of 498 amino acids. MX L polymerase cDNA sequence was 6,668 nucleotides long and predicted ORF encodes the L polymerase of 2,209 amino acids. Nucleotide and deduced amino acid sequences were compared with the known GPC and L sequences and the comparison revealed that both genes shared the highest amino acid identity with Armstrong strain. Phylogenetic analysis confirmed that MX represents a separate LCMV strain. The GPC and L genes products contained several characteristic conserved regions. On the other hand, we have observed numerous differences in predicted protein sequences, which distinguish MX LCMV from other LCMV strains and might be of potential biological significance.


ArenavirusGlycoproteinLymphocytic choriomeningitis virusL polymeraseMX strainNegative strand RNA virus


Lymphocytic choriomeningitis virus (LCMV) represents the prototype member of the Arenaviridae family. Arenaviruses include clinically important human pathogens that cause severe, often lethal, hemorrhagic fever, such as Lassa fever virus and the South American hemorrhagic viruses [1, 2]. LCMV is an important model system to study acute and persistent viral infection, virus–host interactions and associated diseases [3, 4]. Moreover, increasing evidence indicates that LCMV might be a neglected human pathogen of clinical significance [58]. The genome of these enveloped negative-strand (NS) RNA viruses [1] consists of a small (S, 3.4 kb) and a large (L, 7.2 kb) segments. Each genomic segment uses an ambisense coding strategy to direct the synthesis of two polypeptides in opposite orientation, separated by an intergenic region with a predicted folding of a stable hairpin structure [9, 10]. The S RNA carries the open reading frames for the nucleoprotein (63 kDa) and viral glycoprotein precursor (GPC, 75 kDa), whereas the L RNA encodes the viral RNA-dependent RNA polymerase (RdRp, or L protein) (200 kDa), and a small RING finger protein Z (11 kDa) [10, 11]. This unusual ambisense arrangement may contribute to viral persistence, frequently established by arenaviruses, by allowing independent regulation of viral protein synthesis and virions production.

In the past, our group had contributed to the identification of a new MX strain of LCMV and determination of the primary structure of two genes NP and ZP. MX was initially considered an unusual transmissible agent of human carcinoma MaTu cells [12, 13]. However, we have shown that this agent is in fact persisting LCMV. The biological features of MX LCMV exhibited several similarities with other persistent LCMV. It is a non-cytolytic virus that does not form distinct virions and its transmission to uninfected cells is mediated by cell-to-cell contact or by cell-free cytoplasmic extract, but not by culture medium. Infected cells accumulate a high level of cytoplasmic NP and full-length as well as deleted viral RNAs. In contrast to other LCMV strains, LCMV MX has restricted host range. It can infect and replicate in HeLa cells, human embryo fibroblasts, human A549 cells, monkey VERO cells and rat PC12 cells, but not in AGS cells, canine MDCK cells, mouse NIH 3T3 fibroblasts, mouse embryo fibroblasts, and hamster CHO cells [14, 15].

In order to find out, whether these features of MX LCMV are determined at the genomic level, we decided to obtain and analyze the sequences of two additional LCMV genes. Using RT-PCR we obtained cDNAs encoding glycoprotein and L protein and assessed their primary structures. Comparative analysis with the known GP and L sequences from the other LCMV strains and arenaviruses provided additional evidence, which support the conclusions that MX represents a distinct strain of LCMV. Furthermore, our data add new information to the recently emerging concept of genetic diversity of LCMV isolates.

Materials and methods

Cell lines and virus

Human MaTu cells (as described in [12]) and uninfected and LCMV MX-infected cervical carcinoma HeLa cells were cultivated in Dulbecco’s Minimal Essential Medium (BioWhittaker, Verviers, Belgium) supplemented with 10% fetal calf serum and 40 μg/ml gentamicin (Lek, Slovenia) in a humidified atmosphere with 5% CO2 at 37°C.

LCMV MX strain was maintained as a persistent virus in MaTu cells. MaTu cell-free extract prepared by disruption of the cells in a hypotonic buffer followed by sonication, and three cycles of freezing thawing were used for infection of HeLa cells. Briefly, cells were plated at low density 16 h before use. At the time of infection, medium was removed from the cells, and 1 ml of cell-free extract was added. After 1 h at room temperature, the virus inoculum was removed and fresh medium was added. The cells were cultivated as described above and subsequently passaged. Spread of infection was followed by immunoflourescence detection of MX LCMV NP in recipient cells [15]. The RNA samples for sequencing were obtained from 37.–40. passage of LCMV MX infected HeLa cells (HeLa-MX).

RT-PCR and sequencing

Total RNA was isolated from cells by using InstaPure reagent (Eurogentec, Seraing, Belgium) according to the manufacturer’s instructions. Reverse transcription was performed with M-MuLV reverse transcriptase (Finnzymes; Espoo, Finland) using random hexamer primers (Invitrogen; Carlsbad, CA, USA). The mixture of 3 μg of total RNA from HeLa-MX cells and random primers (400 ng/μl) was heated for 10 min at 70°C, cooled quickly on ice and supplemented with dNTPs (each at 0.5 mM concentration, Finnzymes), reverse transcriptase buffer containing 6 mM MgCl2, 40 mM KCl, 1 mM DTT, 0.1 mg/ml BSA, and 50 mM Tris/HCl, pH 8.3. The mixture (final volume 24 μl) was further supplemented with 200 U of reverse transcriptase M-MuLV, incubated for 55 min at 42°C, heated for 15 min at 70°C and stored at −80°C until used.

Initial PCR was performed with Dynazyme EXT polymerase (Finnzymes) in an automatic DNA thermal cycler (Eppendorf AG) with primers designed on the basis of known sequences of LCMV strains WE and Armstrong. Obtained PCR fragments were gel-purified and sequenced, strain-specific primers were designed and used for additional RT-PCR. The 25 μl reaction premix contained 1 μl of cDNA, 1× buffer with 1.5 mM MgCl2, 200 μM dNTP, 0.2 μM of each plus and minus strand primer, and 0.5 U of Dynazyme EXT polymerase. The protocol of PCR consisted of initial denaturation at 94°C for 3 min followed by 35 cycles with denaturation at 94°C for 30 s, annealing for 40 s (the annealing temperature depended on sets of primers) and extension at 72°C (time of extension depended on size of PCR product), followed by a final extension at 72°C for 5 min.

5′ end of GPC gene was reverse transcribed and amplified using SMART RACE cDNA Amplification Kit (Clontech, Palo Alto, CA, USA) and GPC-specific primers (GPRACE 5′-GTGGGAATTATTGGCTGAACAT-3′, GPRACEN 5′-CCAGAACTCCCCATGCTGATG-3′) according to the protocol of the manufacturer.

PCR products were gel-purified using the Wizard® SV Gel and PCR Clean-Up System (Promega, Madison, WI, USA) to eliminate the remnant primers and dNTP. The purified PCR products were directly sequenced on both strands using an ABI 3100 automated sequencer (Applied Biosystems). Sequence coverage was two to sixfold per nucleotide position. A total of six GPC-specific and 44 L-specific primers were used for PCR and sequencing (see Fig. 1 for sequencing strategy and supplementary data for sequences of selected primers). Sequences of GPC and L polymerase have been sent to GenBank and assigned the accession numbers EU195888 and EU195889.
Fig. 1

Scheme of the sequencing strategy of MX GPC and MX L protein genes. Horizontal lines represent obtained PCR products. Numbers indicate the corresponding primers. The primer sequences are listed according to these numbers in the supplementary data for this article

Northern blot hybridization

Twenty micrograms of total RNA from MX-infected and uninfected HeLa cells was separated on 1.2% agarose-formaldehyde gel and transferred to HybondTM-X plus membrane (Amersham Biosciences; Little Chalfont, Buckinghamshire, UK) by capillary blotting. The membrane was probed with appropriated PCR fragments of LCMV strain MX glycoprotein and L protein, respectively. Probes were labeled with [α-32P] dCTP using the Megaprime DNA Labeling System (Amersham GE Healthcare; Little Chalfont, Buckinghamshire, UK). Prehybridization, hybridization, and washing under stringent condition were performed according to Northern Analysis Protocol from Promega. Northern blot was analyzed by autoradiography. RNA sizes were estimated using a 0.25–10.0 kb RNA ladder (Invitrogen).

Sequence analysis

Nucleotide sequences were compiled using the CAP program ([16], Isotopically averaged molecular weight and pI of GPC and L protein gene, respectively, were estimated from deduced amino acid sequences, using ProteinCalculator v3.3 program (available at

Nucleotide and amino acid sequences of the GPC and L genes were aligned by the program CLUSTAL W [17] ( and manually adjusted. In addition to the genes sequenced in this work, the following sequences were included (accession number): LCMV Armstrong (M20869, J04331), LCMV CH-5692 (AF325214), LCMV CH-5871 (AF325215), LCMV CHV1 (U10157), LCMV CHV2 (U10158), LCMV CHV3 (U10159), LCMV LE (EF264923), LCMV M1 (AB261991), LCMV Marseille #12 (DQ286931, DQ286932), LCMV WE (M22138, AF004519), LCMV Y (DQ118959), Pichinde (K02734, NC_006439), Junin (D10072, AY216507), Lassa (J04324, NC_004297), Tacaribe (M20304, J04340), Mopeia (AY772170, AY772169), Guanarito (AY497548, AY216504), Mobala (AY342390, DQ328876), and Ippy (DQ328877, DQ328878). Multiple alignments of amino acid sequences generated by Clustal W (default setting) were used for phylogenetic analysis. The phylogenetic trees were constructed by the neighbor-joining method using the program Clustal W (default setting) [17]. Analysis of the potential post-translation modification sites were done on the complete sequence of the GPC using Prosite database (


Sequencing of GPC and L protein

Two-step strategy was employed to sequence GP and L protein of LCMV MX. First, primers were designed based on known sequences of LCMV strains Armstrong and WE. Numerous short regions were transcribed and amplified using these primers. Subsequently, the obtained PCR fragments were sequenced to design strain-specific primers for additional RT-PCR. The fragments of the predicted lengths were gel-purified and sequenced from both sites using the PCR primers. Remaining gaps were closed after designing new sets of strain-specific primers (Fig. 1).

The obtained MX GPC cDNA sequence was 1,615 nucleotide (nt) long and contained an ORF starting at nt 78 and terminating at nt 1,574. The ORF represents the GPC precursor gene, encoding 498 amino acids (aa) long protein with predicted molecular weight of 56.2 kDa and estimated pI 8.67. MX L polymerase cDNA sequence was 6,668 nts in length. Predicted coding region (nt 30–6,659) encodes the L polymerase of 2,209 aa with predicted molecular weight of 254.1 kDa and isoelectric point of 7.39.

Northern blot analysis

Northern blot analysis confirmed the presence of LCMV RNA in MX-infected HeLa cells. GP and L probe, respectively, derived from LCMV strain MX hybridized to RNA isolated from MX-infected HeLa cells, but not from control, uninfected HeLa cells (Fig. 2). The hybridization with GP probe yielded two major bands of 1.6 and 2.5 kb, respectively, and one larger faint band of about 3.5 kb. The faint band was at the position of a full-length S RNA (Fig. 2a). Out of two major bands, the smaller one corresponded to GP mRNA and the larger one corresponded to an RNA molecule that was smaller than the genomic S RNA, but larger than GP mRNA-sized and most probably represents subgenomic S RNA. This latter assumption is supported by our observation that primers designed for amplification of the full-length GPC from HeLa MX cDNA yielded not only an expected GPC amplicon, but also a smaller PCR product with the size of about 700 bp. Sequence analysis revealed that this PCR product represented a deleted form of GPC that was lacking the 882 nt fragment with the break points at nt 469 and 1351. Genomic S RNA with such a deletion roughly corresponded to the major subgenomic S RNA of 2.5 kb detected in the Northern blot.
Fig. 2

Northern blot analysis of MX LCMV RNA with MX GPC and MX L protein probes. Twenty micrograms of total RNA from MX-infected [1] and uninfected [2] HeLa cells was separated on 1.2% agarose-formaldehyde gel and transferred to membrane by capillary blotting. The membrane was hybridized with 32P-labeled PCR fragments of LCMV strain MX glycoprotein (a) and L protein (b), respectively, and analyzed by autoradiography. RNA sizes were estimated using a 0.25–10.0 kb RNA ladder. The identified S RNA, GP mRNA, and L RNA are indicated

S RNA-related molecules of similar sizes were previously detected in MX LCMV-infected cells with NP probe [15] and they were also found by Bruns et al. [18] in mouse L cells persistently infected with LCMV strain Armstrong and by van der Zeijst et al. [19] in persistently infected BHK cells.

A single band of 7.5 kb was detected by hybridization with L probe. This position corresponded to the full-length L RNA suggesting the absence of major deletions in this part of the viral genome (Fig. 2b).

Phylogenetic analysis

In order to determine the overall degree of sequence identity among MX and other LCMV strains, nucleotide and amino acid sequences of GPC and L polymerase gene, respectively, were aligned and compared (Table 1). The identity values showed the same trend as those observed before for MX NP and MX ZP genes in comparison to other LCMV strains, and emphasized the status of MX as an individual LCMV strain. Both GPC and L genes shared the highest amino acid identity with the Armstrong strain. In agreement with observation previously published by Vieth et al. [20] and Emonet et al. [21], the polymerases exhibited a higher genetic divergence than the GPC proteins (Table 1). Similarly, when MX GPC and MX L genes were compared with other Old World and New World arenaviruses, L genes were generally less conserved than GPC genes.
Table 1

Identity at the nucleotide and amino acid levels among LCMV MX and other LCMV strains

% Identity

LCMV strain


GPC gene

L gene















Marseille #12













































nt, nucleotide level; aa, amino acid level; ND, not defined

Phylogenetic relationships between LCMV MX strain and the other arenaviruses species were analyzed using entire amino acid sequences of GPC and L genes. The neighbor-joining method was applied to construct unrooted trees (Fig. 3). The phylogenetic tree derived from GPC gene and L gene, respectively, revealed that the strain MX formed a group with other LCMV strains (Fig. 3a, b). These results also supported the conclusion that MX represents a distinct LCMV strain.
Fig. 3

Phylogenetic relationship among arenaviruses based on the analysis of complete amino acid sequences of the GPC (a) and L protein (b) genes generated by neighbor-joining method

Sequence comparison

Since MX strain represents persistent LCMV with restricted host range, comprehensive analysis has been done with the goal of finding those GPC and L polymerase gene regions that most likely account for the differences responsible for these biological features. It has been shown that cleavage of GPC to GP1 and GP2 by SKI-1/S1P is not required for cell surface expression of LCMV GPs on infected cells but is essential for their incorporation into virions and for the production of infectious virus particles. Furthermore, proteolytic processing of LCMV GPC depends on the presence of a cluster of basic amino acids at the C-terminus of the cytoplasmic domain of GP2, a structural motif that is conserved in Old World arenaviruses [22].

Sequence comparison revealed that MX GP, like the GPs of other LCMV strains, contains an RRLA recognition site for cleavage as well as a conserved sequence WKRR at the C-terminus of GP2. Post-translation modifications obviously played the major role in function of the processed proteins. Therefore, we analyzed the potential N-glycosylation and N-myristoylation sites within the complete sequence of the GPC. For LCMV it was shown that both cleavage and cell surface expression of LCMV GPC are impaired in parallel either when no N-glycosylation occurs or a proline residue is present at position 110 [23, 24]. In our study, glycosylation sites were predicted and showed to be identical with ARM and WE strain, respectively, and a leucine residue was present at position 110. In addition, the predicted N-myristoylation sites were identical. A recent study revealed that stable signal peptide (SSP) of LCMV surface glycoprotein is involved not only in efficient glycoprotein expression, processing and cell surface localization, but also in particle formation and GP-mediated cell fusion [25]. We observed that MX SSP sequence is largely conserved, it contains a conserved N-terminal myristoyl moiety acceptance site (G2), an N-terminal extension, two hydrophobic domains (H1 10–32, and H2 34–53) separated by a single positive charge (K33), and a C-terminal SP cleavage site (54–58). We found out that the valine residue at position 5 is substituted by a similar amino acid methionine.

However, the abundant subgenomic S mRNA with internal deletion in GPC gene between nts 469 and 1351 could be translated to a defective 16 kDa GPC protein composed of 146 N-terminal amino acids but missing the rest of the molecule due to a frameshift-generated stop codon. Because such truncated GPC is lacking the recognition site for cleavage, important C-terminal sequences, and a hydrophobic transmembrane domain, it is well possible that it can be neither expressed at the cell surface nor contribute to assembly of virions.

The arenavirus L protein contains the characteristic sequence motifs conserved among the RdRp L proteins of the NS RNA viruses. The proposed polymerase active site of L is located within its domain III, which contains highly conserved A–D motifs. Several regions along the whole protein exhibit a considerable degree of conservation as found in the polymerase domain, suggesting functional or structural relevance of these regions [26, 27].

In this study, the alignment analysis revealed a high degree of homology between all available LCMV L proteins with 1,765 identical amino acids, while MX L protein had 126 amino acid substitutions, which differed from the other strains. The most variable regions (399–484 and 873–946 residues) were located within the N-terminal portion of the protein. These regions contained condensed clusters of conserved and non-conserved substitutions. Within the polymerase region, MX L protein exhibited all characteristic highly conserved motifs. It was shown that the presence of SDD sequence, a characteristic of C motif of segmented NS RNA viruses, as well as the presence of the conserved aspartate (D) residue within A motif of L proteins, is strictly required for the polymerase activity of the LCMV L protein [28]. We found that SDD sequence within C motif either D residue within A motif was preserved in MX L protein. In contrast, the carboxy-terminus of the MX L protein contained a cluster of five amino acid residues, which were different from all other strains.


LCMV can easily induce persistent infection in a broad range of cells derived from various species. Since the virus is usually maintained without any apparent signs of cytopathic effect, the infection might remain unrecognized. This was the case for the human carcinoma cell line MaTu that was found to contain a transmissible agent (MX), whose nature was elucidated as a novel strain of LCMV [1215]. Biological features of LCMV MX are very similar to LCMV detected by other authors in a number of persistently infected cell lines but differ by the restricted host range [18, 19, 29]. Previous studies have shown that the genes coding for NP and ZP virus proteins are intact, but differ from the homologous proteins in other viral strains to such extent as the other LCMV strains differ from each other.

In this study, we have determined the primary structure of two additional LCMV MX genes GPC and L protein. Comparative analysis with the known GP and L sequences from the other LCMV strains and arenaviruses confirmed the status of MX as an individual LCMV strain.

It is essential to understand the principles by which persistence is initiated and maintained, as well as the pathologic consequences of virus replication in a host in terms of causing disease. Small differences in either viral or host genes seem to influence the course of infection and the resulting disease state [30].

The GP of LCMV serves as virus attachment protein to its receptor on host cells and is the key determinant for cell tropism, pathogenesis, and epidemiology of the virus. In persistent infection, surface expression of GP is significantly down-regulated relative to that in acutely infected cells [18]. Correct proteolytic processing of GPC is essential for its incorporation into virions and for the production of infectious virus particles [22].

Comprehensive sequence analysis revealed that MX GP, like the GPs of other LCMV strains, contains all conserved gene regions responsible for efficient glycoprotein expression, processing, and cell surface localization, with the exception of the change of phenylalanine to leucine at amino acid 260. This change is identical as in LCMV ARM clone13, an isolate that suppresses the CTL response in adult immunocompetent mice and thereby causes a persistent infection [31].

It is known that defective interfering (DI) RNAs have an important role in down-regulating viral protein expression and establishing persistent infection, probably by competing during replication with the standard genome for virus-encoded proteins. Common types of DI RNAs of RNA viruses contain sequence rearrangements or large deletions in protein-coding regions. This results in RNA templates that support replication, but do not support transcription of mRNAs that can direct the synthesis of functional protein [32].

Northern blot analysis of HeLa MX cells has detected heterogeneous population of viral RNA species. Our data are consistent with numerous observation of persistent LCMV infection from other laboratories [18, 19, 29]. Moreover, we detected a GPC-carrying RNA subpopulation with a large deletion in the central part of GPC gene that corresponds to the major subgenomic S RNA of 2.5 kb.

Accumulation of such defective viral genomes in cells is likely to reduce transcription of the full-length genomic RNAs by a competitive occupation of the transcriptional machinery, ultimately leading to down-regulation of the viral protein synthesis and production of infectious progeny virions. This would result in infections with the same characteristic as those previously reported for LCMV persistence and attributed to LCMV DI activity [29, 32]. Because the proportion of the truncated to full-length GPC mRNAs is relatively equal, the truncated RNAs might be critical in maintaining persistence by keeping viral glycoprotein expression low. It is also conceivable that the truncated GPC presumably produced from the aberrant subgenomic S RNA hinders formation of new virions. However, generation of the subgenomic viral RNAs may depend on both the particular cell line and physiological status of infected cells. We are currently examining this hypothesis in order to determine, whether this major subgenomic S RNA species contributes to MX persistence.

Recent studies showed that L and Z genes exhibited a higher variability than structural genes NP and GPC. This phenomenon was observed when comparing the Old World arenaviruses not only from different species, but also between different representatives of the same species [20, 21]. Noteworthy, similar analyses in other RNA viruses demonstrated that distances between structural genes are always higher than between genes implicated in viral replication [21]. It was shown that the heterogeneity is not spread homogenously along the L protein. The most conserved regions were previously defined as functional domains. These regions contain sequences, which are conserved among all arenaviruses [20, 21]. Within domain III are found the conserved A–D motifs which are thought to form the module containing the active site in RNA synthesis [27]. Recent studies confirmed the key role of the highly conserved amino acid residues within motifs A and C on L polymerase activity and provided evidence that oligomerization of L is required for its function [28]. Conserved regions are interspersed by stretches of high sequence variability [20]. Our findings supported previous assumption. MX L protein exhibited all characteristic highly conserved motifs as well as the conserved aspartate (D) residue within motif A of L proteins that is strictly required for the polymerase activity of the LCMV L protein [28]. The most variable regions (399–484 and 873–946 residues) were located within the N-terminal portion of the protein. The carboxy-terminus of the MX L protein contained cluster of five amino acid residues, which were different from all other strains. It needs to be determined whether these substitutions could have effects on polymerase activity.

In summary, we have completed the genetic characterization of the MX agent by providing the sequence of the GPC and L genes. The results of this work confirmed that MX represents a distinct strain of LCMV. They also fit with a current concept that the genetic diversity of LCMV is higher as believed before and support accumulating data that each strain or isolate differs from the other ones. The future investigations will show whether there is a distinct relationship between the genomic characteristics and the pathogenic properties of LCMV strains/isolates.


This work was supported by the grant 2/5081/26 from the Scientific Grant Agency of the Ministry of Education of Slovak Republic and Slovak Academy of Sciences.

Supplementary material

11262_2008_240_MOESM1_ESM.doc (71 kb)
MOESM1 (DOC 71 kb)

Copyright information

© Springer Science+Business Media, LLC 2008