Bovine Rotavirus Nonstructural Protein 4 Produced by Lactococcus lactis Is Antigenic and Immunogenic

Abstract

Rotavirus nonstructural protein 4 (NSP4) can induce diarrhea in mice. To get insight into the biological effects of NSP4, production of large quantities of this protein is necessary. We first tried to produce the protein in Escherichia coli, but the nsp4 gene proved to be unstable. The capacity of the generally regarded as safe organism Lactococcus lactis to produce NSP4 either intra- or extracellularly was then investigated by using the nisin-controlled expression system. Production of recombinant NSP4 (rNSP4) was observed in L. lactis for both locations. In spite of a very low secretion efficiency, the highest level of production was obtained with the fusion between a lactococcal signal peptide and rNSP4. Cultures of the rNSP4-secreting strain were injected into rabbits, and a specific immune response was elicited. The anti-rNSP4 antibodies produced in these rabbits recognized NSP4 in MA104 cells infected by rotavirus. We showed that L. lactis is able to produce antigenic and immunogenic rNSP4 and thus is a good organism for producing viral antigens.

Rotavirus is the major etiologic agent of severe diarrhea in infants and young children around the world (6). Rotavirus infects mature villus enterocytes in the small intestine. Nonstructural protein 4 (NSP4), encoded by gene 10, has been shown to be an intracellular receptor for double-layered particles. Purified NSP4 and a 22-amino-acid peptide (amino acids 114 to 135) were both capable of inducing dose-related diarrhea after intraperitoneal or intraduodenal administration to 6- to 10-day-old mice (1). NSP4 is thought to act as an enterotoxin which triggers chloride secretion by a calcium-dependent signal transduction pathway. To study the biological properties of NSP4, it is necessary to produce large quantities of NSP4 protein. Previously, Estes et al. produced and purified full-length NSP4 from Spodoptera frugiperda 9 cells infected with a recombinant baculovirus expressing rotavirus gene 10 of strain SA11 (22). Newton et al. (12) produced part of NSP4 (amino acids 86 to 175) as a fusion with a 36-kDa domain of glutathione S-transferase.

Developing efficient gene expression and protein secretion systems in nonpathogenic gram-positive lactic acid bacteria is an original approach for producing proteins of therapeutic interest (15, 25) and a new strategy for rotavirus vaccination. These lactic acid bacteria possess many properties which make them good candidates for oral vaccination purposes; e.g., they have generally regarded as safe status or adjuvant properties (15). They have already been used to produce several bacterial antigens and interleukins (19). Some viral antigens or parts of viral antigens have been produced in lactic acid bacteria; antigen M6-gp41E has been produced in Lactobacillus plantarum (5), a fragment of the human immunodeficiency virus type 1 envelope protein has been produced in Streptococcus gordonii (16), 250 amino acids of rotavirus protein VP7 have been produced in L. plantarum, an epitope of foot-and-mouth disease virus protein VP1 (amino acids 137 to 162) has been produced in Lactobacillus casei (15), and a short epitope of bovine coronavirus has been produced in Lactoccocus lactis (9).

In this study, we constructed strains of L. lactis that produce recombinant NSP4. Both intra- and extracellular locations of recombinant NSP4 (rNSP4) were examined by using the nisin-inducible expression system (3). In spite of a very low efficiency of secretion of rNSP4, the highest level of production of rNSP4 was observed when the nsp4 gene was fused to a lactococcal signal sequence. The recombinant viral protein showed antigenic and immunogenic properties (i.e., it was recognized by specific antibodies and was able to induce an immune response). In addition, the recombinant L. lactis strains should allow study of the biological properties of NSP4 without interference from lipopolysaccharides or inflammatory reactions like those observed with Escherichia coli.

MATERIALS AND METHODS

Bacterial strains, media, and reagents.

The bacterial strains and plasmids used in this work are listed in Table 1. L. lactis and E. coli were grown in M17 medium (21) and in Luria-Bertani (17) medium, respectively. When required, antibiotics were added at the following concentrations: 50 μg of ampicillin (Roche) per ml and 5 μg of chloramphenicol (Sigma) per ml for L. lactis and 10 μg of chloramphenicol per ml for E. coli.

TABLE 1.

L. lactis strains and plasmids

Strain or plasmid	Relevant propertiesa	Reference
Strains
MG1363	Plasmid-free strain
NZ9000	Derivative of MG1363 carrying regulatory genes nisR and nisK	7
Plasmids
pRF10	Apr; pBS+ carrying gene 10 of rotavirus RF encoding NSP4	Unpublished data
pNSP4	Apr; pBS+ carrying the EcoRI-NSP4-XbaI fragment	This study
pMAL:NSP4	Apr; pMAL-c2 carrying the EcoRI-NSP4-XbaI fragment	This study
pET:NSP4	Apr; pET23b+ carrying the EcoRI-NSP4-XbaI fragment	This study
pSEC	Cmr; Usp45 signal leader fused to the nuc gene mature moiety under control of the nisin-inducible promoter PnisA	Langella et al.b
pCYT	Cmr; nuc gene mature moiety under control of the nisin-inducible promoter PnisA	Langella et al.b
pSEC:NSP4	Cmr; nuc gene replaced by NSP4 gene in pSEC	This study
pCYT:NSP4	Cmr; nuc gene replaced by NSP4 gene in pCYT	This study

^aCm^r, resistance to chloramphenicol; Ap^r, resistance to ampicillin. 

^bLangella et al., unpublished data. 

Cloning procedures and PCR.

Strains of E. coli and L. lactis were transformed by electroporation by using standard procedures (4, 26). Plasmid DNA was isolated from L. lactis as described previously (24). PCR amplifications with Taq polymerase (Promega) were performed by using 25 cycles, with each cycle consisting of a denaturation step at 94°C for 30 s, a primer annealing step at 55°C for 30 s, and a primer extension step at 72°C for 1 min, with a DNA thermocycler (Perkin-Elmer GeneAmp PCR system 2400). All other DNA manipulations were performed by established procedures (17).

Construction of plasmids for rNSP4 production in E. coli.

Two systems were used to produce rNSP4 in E. coli. An EcoRI-XbaI DNA fragment encoding NSP4 was isolated from pNSP4 (as described previously) and inserted into pMAL-c2 (New England Biolabs, Hitchin, United Kingdom), an E. coli bacterial expression vector used to express a maltose binding protein-NSP4 fusion protein.

In the second approach, we used the pET system (Novagen, Genetics Institute, Cambridge, United Kingdom). EcoRI and NotI restriction sites were introduced into the nsp4 gene to allow cloning in pET23b+. Nucleotides were substituted by performing PCR with pRF10 and the following primers flanking the NSP4 coding sequence: forward primer 5′-TACCGGAATTCCGGAATGGAAAAGCTTACCGAC-3′ and reverse primer 5′-ATAGTTTAGCGGCCGCCATCGCTGCAGTCACTTCTTTTGG-3′. The amplified fragment was cloned into pET23b+. Two recombinant plasmids, pMAL:NSP4 and pET:NSP4, were obtained in E. coli TG1. The nsp4 gene cloned with the pET system was expressed in E. coli B121.

Construction of plasmids for expression of rNSP4 in L. lactis.

Unless otherwise indicated, plasmid constructions were first established in E. coli and then transferred to L. lactis. cDNA of the bovine rotavirus (RF strain) gene 10 encoding NSP4 was obtained by reverse transcription-PCR, cut with EcoRI, and cloned in EcoRI-cut pBluescript SK+ (pBS; Stratagene), resulting in pRF10. An NsiI restriction site was introduced into the nsp4 coding sequence to allow in-frame cloning with the ATG start codon of the nuc gene (10) contained on the pCYT and pSEC plasmids (P. Langella, Y. Le Loir, S. Gilbert, J. Commissaire, R. L'Haridon, and G. Corthier, unpublished data). pSEC and pCYT are derived from pNZ8010 (3), which contains the gus gene under transcriptional control of the promoter of nisA (P_nisA). The gus gene was deleted by XhoI digestion and replaced by a BamHI-XhoI-cut DNA fragment encoding the ribosome-binding site (RBS_Usp45) and signal sequence (SP_Usp45) of the usp45 gene (23) and the mature part of the staphylococcal nuclease protein (Nuc) (18) to obtain plasmid pSEC. The DNA fragment encoding SP_Usp45 was then deleted by reverse PCR to obtain pCYT. An NsiI site was introduced at the 3′ end of RBS_Usp45 to allow replacement of the nuc coding sequence by the DNA fragment encoding the rNSP4 protein. Nucleotide substitutions were made by performing PCR with pRF10 as the template DNA. This procedure required a mutagenic primer, 5′-GGCGAATTCGATGCATCCGAAAAGCTTACCGAC-3′ containing EcoRI and NsiI sites (underlined), and a reverse primer, 5′-GTGACTGCAGCGATGTAATGAGATATCTAGAGCC-3′ containing EcoRV and XbaI sites (underlined and boldface). The amplified fragment was digested with EcoRI and XbaI and cloned into pBS+ (Stratagene) which had been digested with EcoRI and XbaI, generating pNSP4. Digestion of pNSP4 with NsiI and EcoRV allowed purification of a fragment containing the nsp4 gene, which was cloned in two L. lactis vectors, pSEC and pCYT (Fig. 1). Thus, the nuc gene was replaced by the modified cDNA fragment encoding NSP4. The sequence of each fusion was checked. The two resulting plasmids, pSEC:NSP4 and pCYT:NSP4, were introduced into L. lactis NZ9000, which is a derivative of L. lactis MG1363 carrying the two regulatory genes of P_nisA, nisR and nisK (3).

FIG. 1.

Construction of pCYT:NSP4 and pSEC:NSP4. , nuc gene; , NSP4 gene; , chloramphenicol resistance (Cm^r); , ampicillin resistance (Ap), ▧, SP_Usp45; ——, ribosome-binding site (RBS); ▩, P_nisA. Restriction enzyme sites are also indicated; only relevant restriction sites are shown. The direction of transcription is indicated.

Nisin induction and analysis of rNSP4 production in L. lactis NZ9000(pSEC:NSP4) and NZ9000(pCYT:NSP4).

For induction and production of rNSP4, 50-ml cultures of L. lactis NZ9000(pSEC:NSP4) and NZ9000(pCYT:NSP4) were grown until the optical density at 600 nm (OD₆₀₀) was 0.5 and were induced with 1 ng of nisin (Sigma France) per ml. Noninduced cultures were used as controls. After induction, the cells were grown for 3 h and were harvested and resuspended in 5 ml of water. Sterile glass beads were added, and the cells were disrupted with a Mini-Beadbeater-8 (Biospec Products). Cellular debris and glass beads were removed by centrifugation, and the protein-containing supernatant was then concentrated fivefold with a Speed-Vac. Samples to be compared were prepared in parallel and loaded on the same gel. Ten microliters of each sample was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (15% polyacrylamide) (8), followed by Western blotting. rNSP4 was detected by using a rabbit anti-NSP4 antiserum directed against the C-terminal part of NSP4 (dilution, 1:1,000; kindly supplied by L. Svensson, Stockholm, Sweden) and alkaline phosphatase-labelled anti-rabbit immunoglobulin G (heavy plus light chains; BioSys, Compiègne, France). Staining was performed with a standard alkaline phosphatase substrate, 5-bromo-4-chloro-3-indolylphosphate (BCIP)-nitroblue tetrazolium (Life Technologies, Gaithersburg, Md.). After staining, different exposures of nonsaturated film were scanned and compared by using ImageQuant programs to obtain average values.

To determine the cell distribution of rNSP4, cell fractionation was performed as described previously (11). Briefly, cell and supernatant fractions were separated and concentrated 10-fold by using trichloroacetic acid. The equivalent of 1 ml of culture at an OD₆₀₀ of 1 was concentrated in 100 μl (final volume), and 10 μl of each sample was loaded on an SDS-PAGE gel. To determine the location of the precursor preUspNSP4, the NZ9000(pSEC:NSP4) cell fraction was divided into cytoplasmic and cell wall fractions as previously described (14).

Cell culture and viral infections.

Fetal rhesus monkey kidney cell line MA 104 was grown in confluent monolayers in Eagle's minimal essential medium (Life Technologies, Paisley, Scotland) supplemented with 10% fetal calf serum and antibiotics. MA104 monolayers were washed and inoculated with bovine rotavirus (RF strain) at a multiplicity of infection of about 5 PFU per cell. Cells were processed for Western blotting or immunoprecipitation at 8 h postinfection. Spodoptera frugiperda 9 cells were infected with recombinant baculovirus expressing rotavirus gene 10 (BNSP4-SA11; kindly supplied by M. Estes) to produce NSP4. Cells were collected 3 to 4 days postinfection.

Immunization of rabbits with a culture of L. lactis NZ9000(pSEC:NSP4).

Before two 6-month-old rabbits were injected, blood samples were taken to prepare a preimmune serum and to check (by enzyme-linked immunosorbent assay) that the rabbits were not infected with rotavirus. Two hundred milliliters of a nisin-induced culture of L. lactis NZ9000(pSEC:NSP4) containing 10⁸ bacteria/ml was concentrated 200-fold and lysed. To immunize the rabbits, we used a 1:1 mixture composed of 500 μl of the concentrated culture lysate (containing approximately 10¹⁰ bacteria/ml) and 500 μl of Inject Alum's adjuvant (Pierce, Rockford, Ill.). The rabbits were immunized four times intramuscularly at 3-week intervals. One week after the last injection, the rabbits were bled, and sera (dilution, 1:50) were used in Western blot experiments.

RESULTS

Production of rNSP4 in E. coli and L. lactis.

The DNA fragment encoding rNSP4 (nsp4) was cloned and expressed in E. coli by using two different expression vectors, pET and pMAL. The insert in both plasmids was sequenced, and the expected sequence was obtained only with the pET:NSP4 plasmid. Several mutations were observed in the pMAL:NSP4 plasmid. Expression of rNSP4 was induced in E. coli containing pET:NSP4. We observed that growth was significantly reduced after this induction, suggesting that rNSP4 is toxic for E. coli. Total protein extracts of this strain were subjected to SDS-PAGE followed by Western blotting. No rNSP4 was detected on the Western blot with anti-NSP4 antibodies (data not shown).

To produce rNSP4 in L. lactis, the nsp4 gene was cloned in L. lactis on pCYT and pSEC, expression plasmids that allow intra- and extracellular location of the protein of interest, respectively. In both plasmids, expression of nsp4 was under transcriptional control of the nisin-inducible promoter P_nisA (3). The two resulting plasmids, pCYT:NSP4 and pSEC:NSP4, were introduced into L. lactis NZ9000, a derivative of L. lactis MG1363 containing the two regulatory genes of P_nisA, nisR and nisK. We first checked that nisin addition did not inhibit cell growth of L. lactis NZ9000, the control strain. Strains NZ9000(pCYT:NSP4) and NZ9000(pSEC:NSP4) were induced by nisin when sufficient biomass was present (see Materials and Methods) (Fig. 2a). Compared to the control strain, growth of the two L. lactis strains producing rNSP4 was different; the growth of L. lactis NZ9000(pCYT:NSP4) was reduced, and growth of L. lactis NZ9000(pSEC:NSP4) even stopped.

FIG. 2.

Production of rNSP4 in L. lactis. (a) Kinetics of nisin induction: bacterial growth rates in extracts of L. lactis NZ9000(pCYT:NSP4) and NZ9000(pSEC:NSP4) before and after induction with 1 ng of nisin per ml. Times t0, t1, and t2 were the times when samples used to measure OD₆₀₀ and samples for Western blotting experiments were obtained. Symbols: ▴, noninduced NZ9000(pCYT:NSP4) and NZ9000(pSEC:NSP4); ■, induced NZ9000(pCYT:NSP4); ●, induced NZ9000(pSEC:NSP4). (b) Immunoblotting of lysates from L. lactis NZ9000(pCYT:NSP4) and NZ9000(pSEC:NSP4) induced by nisin with rabbit antiserum to NSP4 C-terminal peptide.

Analysis of rNSP4 produced in L. lactis.

Next, production of rNSP4 in the two rNSP4-producing L. lactis strains was analyzed. Whole-cell protein contents were analyzed at the time of induction with nisin and 2 and 4 h after induction (Fig. 2a). The protein extracts were analyzed by SDS-PAGE, followed by Western blot assays in which the blots were developed with anti-NSP4 antibodies (Fig. 2b). In the absence of nisin, no signal was detected, indicating that rNSP4 was not produced. After 2 and 4 h of induction with nisin, the results showed that rNSP4 was present only in cell fractions of L. lactis NZ9000(pCYT:NSP4). In the case of L. lactis NZ9000(pSEC:NSP4), the low proportion of mature NSP4 detected in the total fraction and the absence of a band in the supernatant fraction (data not shown) suggested that the secretion efficiency of mature NSP4 was very poor. Two other major differences between secreted production and cytoplasmic production were observed. (i) One difference was the number of bands detected. In the cellular fraction of L. lactis NZ9000(pSEC:NSP4) four protein forms were distinguished, including a 22-kDa band corresponding to the precursor preUspNSP4 form, a 20-kDa band corresponding to the mature form of rNSP4 (175 amino acids), and 18- and 16.5-kDa bands probably corresponding to degradation products of rNSP4. In the cellular fraction of L. lactis NZ9000(pCYT:NSP4), only a faint 20-kDa band corresponding to the expected size of mature rNSP4 and an 18-kDa major band were observed, suggesting that there was intracellular degradation of mature rNSP4. (ii) The other difference was the intensity of the bands detected. The NSP4 content of NZ9000(pSEC:NSP4) was significantly greater (around 10-fold greater) than that of NZ9000(pCYT:NSP4). The antibodies used for these Western blot experiments recognized the C-terminal region of NSP4, suggesting that degradation of rNSP4 produced in L. lactis occurred at the N-terminal end.

To localize the precursor preUspNSP4 more precisely, the cellular fraction of L. lactis NZ9000(pSEC:NSP4) was divided into protoplast and cell wall fractions by using the protocol of Piard et al. (14). The occurrence of rNSP4 in the different fractions was estimated by SDS-PAGE followed by Western blotting. We observed only four rNSP4 bands in the lane corresponding to the cytoplasmic fraction, suggesting that translocation of preUspNSP4 could be blocked in the bacterial membrane (data not shown).

Immunogenicity of rNSP4 produced in L. lactis.

To test whether the rNSP4 produced in L. lactis can induce a humoral immune response with production of antibodies in rabbits, a culture of L. lactis NZ9000(pSEC:NSP4) was lysed and used to immunize rabbits. The immune serum obtained, LacNS4, was compared with antibodies directed against the C-terminal part of NSP4. The two sera were tested with a Western blot of lysates of S. frugiperda 9 cells infected with baculovirus producing rNSP4 (Fig. 3). As expected, no signal was observed with preimmune serum from rabbits injected with the lactococcal lysate (Fig. 3, lane 1). The NSP4 protein was detected with both anti-NSP4 sera (Fig. 3, lanes 2 and 3). The four bands were observed in both cases, indicating that immunization of rabbits with lysates of rNSP4-producing L. lactis strains leads to production of antibodies that recognize the same NSP4 forms as the antibodies raised against NSP4 produced in baculovirus.

FIG. 3.

Immunogenicity of rNSP4: immunoblotting of lysates of S. frugiperda 9 cells exposed to different sera, including preimmune rabbit serum (lane 1), LacNS4 (lane 2), and rabbit anti-NSP4 C-terminal region serum (lane 3).

Characterization of rNSP4 protein produced in L. lactis.

Four different types of NSP4 were tested in Western blot experiments with the rabbit anti-NSP4 C terminus: NSP4 produced in MA104 cells infected with rotavirus, NSP4 produced in S. frugiperda 9 cells infected with baculovirus, and NSP4 produced in induced cultures of L. lactis NZ9000(pSEC:NSP4) and NZ9000(pCYT:NSP4) (Fig. 4a). No signal was detected in the noninfected MA104 cell control (Fig. 4a and b, lane 1). The anti-NSP4 C-terminal serum allowed us to identify bands specific for NSP4 in the two other samples (Fig. 4a, lanes 2 and 3). Identical results were obtained with the LacNS4 serum (Fig. 4b, lanes 2 and 3). The apparent molecular weights of rNSP4 produced in L. lactis were different than the molecular weights of NSP4 produced in MA104 cells infected with rotavirus or in recombinant baculovirus-infected S. frugiperda 9 cells (Fig. 4a, lanes 2 to 5). The differences were probably due to the absence of glycosylation in L. lactis. For MA104 cells and S. frugiperda 9 cells, we detected a major band around 25 kDa corresponding to an NSP4 form with two glycosylations and a band around 24 kDa corresponding to an NSP4 form with only one glycosylation (Fig. 4a, lanes 2 and 3). A common 20-kDa band corresponding to native NSP4 was detected with the four sources of NSP4. The 22-kDa band corresponding to the precursor preUspNSP4 was observed only with the rNSP4 form fused with SP_Usp45 (Fig. 4a, lane 4). We concluded that LacNS4, the anti-NSP4 antibodies produced in rabbits immunized with lactococci, recognize NSP4 produced in MA104 and S. frugiperda 9 cells.

FIG. 4.

Characterization of rNSP4 protein. (a) Antigenic characterization: immunoblotting of MA104 lysates (lane 1, uninfected; lane 2, infected), S. frugiperda 9 infected lysate (lane 3), and induced L. lactis lysates [lane 4, NZ9000(pSEC:NSP4); lane 5, NZ9000(pCYT:NSP4)]. NSP4 was detected by using a rabbit anti-NSP4 C-terminal region serum. (b) Immunogenic characterization: Western blotting of MA104 lysates (lane 1, uninfected; lane 2, infected) and S. frugiperda 9 infected lysate (lane 3). NSP4 was detected by using LacNS4.

DISCUSSION

In this paper, we describe for the first time production of an entire viral antigen in the food-grade bacterium L. lactis. Our results indicate that L. lactis is able to produce recombinant bovine rotavirus NSP4 that possesses antigenic and immunogenic properties. Previously, Newton et al. (12) produced a fragment of NSP4 in E. coli with the GST fusion protein system. In our hands, it was impossible to express the entire rNSP4 with pMAL and pET systems in E. coli. This failure could have been due to toxicity of NSP4 for E. coli, as previously described with a slightly leaky inducible promoter (20). Two L. lactis strains were constructed to produce rNSP4 in two locations under control of an inducible lactococcal promoter. To produce intra- and extracellular rNSP4, we used the ribosome binding site of L. lactis secreted protein Usp45, and the nsp4 gene was placed under transcriptional control of the nisin-inducible promoter P_nisA. In strain NZ9000(pSEC:NSP4), the nsp4 gene was fused to the signal sequence of Usp45 (SP_Usp45). After induction with nisin, the rates of growth were very different for the two strains and were correlated with rNSP4 production. In L. lactis NZ9000(pSEC:NSP4), the presence of SP_Usp45 allowed 10-fold enhancement of production of rNSP4 compared to intracellular production. Our hypothesis to explain the difference is that the precursor preUsp:rNSP4 could be translated more efficiently than intracellular forms of rNSP4 produced in NZ9000(pCYT:NSP4). Furthermore, recognition of preUsp:rNSP4 by the secretion machinery of L. lactis could allow it to escape intracellular proteolysis. The same positive effect of SP_Usp45 was recently observed in L. lactis for production of bovine β-lactoglobulin, the major cow's milk allergen (Langella, unpublished data), and for production of the Brucella abortus ribosomal L7/L12 protein (L. Ribeiro, personal communication).

It has been found that the signal peptide is necessary, but not sufficient, for mature protein export. However, all proteins bearing this sequence are not secreted (2, 27). Hydrophobic domains of NSP4 could result in localization of rNSP4 in the bacterial membrane and prevent secretion. In Western blot experiments with NZ9000(pSEC:NSP4), we showed that the unprocessed intracellular precursor preUsp:rNSP4 and mature forms of rNSP4 were present in the cellular fraction. The low level of maturation of preUsp:rNSP4 could be due to suboptimal recognition of this hybrid precursor by the secretion machinery of L. lactis, leading to aggregate formation. At this point, we note that (i) only Usp45, a protein with an unknown function, seems to be secreted efficiently by L. lactis (23) and (ii) only bacterial heterologous proteins have been efficiently secreted by L. lactis (19). Using the staphylococcal nuclease (Nuc) as the secreted model protein, Le Loir et al. (11) showed that a positive net global charge in the first 10 amino acid residues of the Nuc N terminus resulted in a dramatic decrease in Nuc secretion efficiency. Analysis of the peptide sequence of the N terminus of rNSP4 revealed the presence of an Arg residue at position +3, resulting in a net global charge of +1 in the first five amino acid residues of the N terminus of rNSP4. Experiments are currently in progess to investigate how insertion of a synthetic propeptide (19) and fusion with Nuc can affect rNSP4 secretion efficiency.

NSP4 produced in recombinant baculovirus-infected S. frugiperda 9 cells and infected MA104 cells was characterized by three major bands. These bands may represent full-length NSP4 with two, one, and no sites glycosylated (13, 22). Unlike the NSP4 produced in a eucaryotic system, the NSP4 protein produced in L. lactis was not glycosylated. This posttranslational function is probably responsible for the difference in the molecular weights of bands.

The rNSP4 protein produced in L. lactis has the same immunogenic properties as the viral protein. We may use these properties to develop a rapid diagnostic method for biomedical analysis, like an enzyme-linked immunosorbent assay. Recovery of rNSP4 protein with antigenic and immunogenic properties offers possibilities for producing other proteins which are produced in E. coli with difficulty. The L. lactis strains provide a good way to produce NSP4 without interference from lipopolysaccharide, as in E. coli. This production could be performed either in vitro or in the digestive tract of mice in order to induce mucosal immunity against this enterotoxin.