The chimerization of the structural genes of DEN1 Puerto Rico/94 with DEN4 was performed in a manner similar to methods used to construct the rDEN2/4 and rDEN3/4 chimeric viruses (Figure 1) 1115. Construction of a stable p4-D1-CME cDNA clone required the addition of a linker sequence between the E and NS1 genes as had been observed for the construction of the rDEN3/4 chimeric cDNA clones 11. A total of four chimeric viruses were generated with either two (ME) or three (CME) DEN1 structural genes in the DEN4 background with or without the Δ30 mutation in the DEN4 3' UTR. These chimeric viruses are referred to as rDEN1/4(CME), rDEN1/4Δ30(CME), rDEN1/4(ME), and rDEN1/4Δ30(ME). Recombinant viruses were recovered in C6/36 cells and then were adapted to Vero cells by serial passage. Following adaptation, the viruses were biologically cloned by terminal dilution in Vero cells, and the complete nucleotide sequence of each chimeric virus was determined (Table 1). As expected, adventitious mutations were detected in the virus populations with each virus containing at least one amino acid change in NS4B, which is a previously identified locus for accumulation of mutations that enhance replication in Vero cells 17.

Chimeric viruses were assessed for replication in Vero cells, which are the intended cell substrate for vaccine manufacture (Figure 2). At a multiplicity of infection of 0.01, rDEN1/4(CME) and rDEN1/4(ME) each reached a virus titer of 7.3 log₁₀ PFU/ml, which approximates the yield of wild type DEN4 infection. Inclusion of the Δ30 mutation resulted in marginally reduced replication, but even rDEN1/4Δ30(CME), which had the lowest level of replication, reached a virus titer of 6.2 log₁₀ PFU/ml. Thus, after adaptation to growth in Vero cells, each chimeric virus replicates to a level that would allow for efficient manufacture of a vaccine virus.

The level of replication of rDEN1/4 chimeric viruses was first evaluated in SCID-HuH-7 mice since the level of replication in these mice is expected to correlate with the level of attenuation in rhesus monkeys 7. In this SCID mouse model, tumors which develop from implanted human hepatoma cells are infected with DEN viruses, and viremia is assessed at seven days post-infection. Wild type DEN4 has previously been shown to replicate to 6.0 log₁₀ PFU/ml in SCID-HuH-7 mice 18. Wild type DEN1 Puerto Rico/94 replicated to a mean peak virus titer of 6.6 log₁₀ PFU/ml (Table 2), which is similar to replication levels observed for the other wild type DEN viruses 1112. Replication of both CME chimeric viruses, rDEN1/4(CME) and rDEN1/4Δ30(CME), was significantly restricted when compared to the wild type DEN1 level of replication with a 40-fold and 159-fold reduction in replication, respectively. The addition of the Δ30 mutation did not appear to confer significant additional attenuation upon the rDEN1/4(CME) chimeric virus.

In contrast, the rDEN1/4(ME) virus reached a peak virus titer of 6.8 log₁₀ PFU/ml and therefore, was not attenuated for SCID-HuH-7 mice. rDEN1/4Δ30(ME) was significantly restricted in replication compared to wild type DEN1 with a 25-fold reduction in replication indicating that the Δ30 mutation was attenuating when added to the rDEN1/4(ME) virus. These results indicate that both the Δ30 mutation and the virus origin of the C gene play a role in the attenuation of rDEN1/4 viruses in SCID-HuH-7 mice.

Wild type DEN1 Puerto Rico/94 was found to replicate efficiently in rhesus monkeys comparable to previous studies with the DEN1 Western Pacific strain 10. All of the animals infected with DEN1 Puerto Rico/94 developed viremia with a mean duration of 2.8 days and mean peak virus titer of 2.0 log₁₀ PFU/ml (Table 3). The rDEN1/4(CME) virus appeared to be only marginally attenuated, if at all, when compared to DEN1 Puerto Rico/94 in terms of mean number of viremic days and mean peak virus titer. However, rDEN1/4Δ30(CME) was found to be attenuated when compared to wild type DEN1 virus and the parent rDEN1/4(CME) virus. Viremia was detected in only half of monkeys infected with rDEN1/4Δ30(CME), and both the mean duration (0.5 days) and mean peak virus titer (1.1 log₁₀ PFU/ml) were less than that of both wild type DEN1 and rDEN1/4(CME).

As had been observed in SCID-HuH-7 mice, infection of rhesus monkeys with rDEN1/4(ME) yielded no evidence of attenuation and the virus replicated to levels similar to that of wild type DEN1. In contrast, replication of rDEN1/4Δ30(ME) was found to be restricted when compared to either wild type DEN1 or rDEN1/4(ME) and was similar to that of rDEN1/4Δ30(CME). Fifty percent of monkeys infected with rDEN1/4Δ30(ME) developed detectable viremia. There was a four-fold reduction in the mean duration of viremia (0.7 days) and nearly a ten-fold reduction in mean peak virus titer (1.1 log₁₀ PFU/ml) compared to wild type DEN1.

The level of serum neutralizing antibody titer was determined in monkeys 28 days post-immunization (Table 3). Infection with wild type DEN1, rDEN1/4(CME), or rDEN1/4(ME) resulted in 100% seroconversion, defined as a four-fold or greater rise in serum neutralizing antibody levels after immunization. Neutralizing antibodies were not detected in any of the four monkeys immunized with rDEN1/4Δ30(CME) despite the presence of a low level of viremia in two monkeys, suggesting that this DEN1/4 chimeric virus may be over-attenuated. In contrast, four of six monkeys immunized with rDEN1/4Δ30(ME) seroconverted with a geometric mean serum neutralizing antibody titer of 1:23 which was approximately four-fold reduced from wild type DEN1 Puerto Rico/94 (1:103).

All immunized monkeys were challenged with wild type DEN1 virus on day 35 post-immunization and resulting viremia was quantified (Table 3). Both mock-immunized monkeys developed viremia after challenge with a mean peak virus titer of 2.1 log₁₀ PFU/ml. Prior immunization with rDEN1/4(CME) induced complete protection from detectable viremia after challenge. rDEN1/4Δ30(CME), which was poorly immunogenic, failed to confer protection in three of four challenged monkeys. This result confirms the over-attenuation of the rDEN1/4Δ30(CME) virus in rhesus monkeys. Both of the (ME) chimeric viruses, rDEN1/4(ME) or rDEN1/4Δ30(ME), conferred complete protection from viremia following wild type DEN1 challenge including those animals that did not develop detectable levels of serum antibody.

Based on the high level of attenuation of rDEN1/4Δ30(ME) virus in SCID-HuH-7 mice and in rhesus monkeys and on its protective efficacy in rhesus monkeys, it appeared that this chimeric virus might be suitable for further study in humans. Therefore, it was tested for infectivity and replication in Aedes aegypti mosquitoes to assess its potential for transmissibility. The infectivity and level of replication of rDEN1/4Δ30(ME) chimeric virus was compared to those of wild type parental DEN1 and DEN4 and of rDEN1/4(ME) following oral feeding of virus in a bloodmeal. Specifically, following ingestion of an infectious bloodmeal that contained 10^4.3 PFU/2 ul, virus replication in the midgut and dissemination to the salivary glands was determined by plaque titration of virus in the mosquito body and head, respectively. Wild type DEN1 and DEN4 replicated efficiently in the midgut of mosquitoes and dissemination to the salivary glands was robust with 88% and 71% of DEN1-infected or DEN4-infected mosquitoes experiencing disseminated infections, respectively (Table 4). 100% of mosquitoes with DEN1 midgut infections developed disseminated infections, while 81% of mosquitoes with DEN4 midgut infections had disseminated infections. DEN1 reached mean virus titers that were approximately 10-fold to 100-fold higher than DEN4 in both the body and head. rDEN1/4(ME) virus had slightly lower infectivity for the midgut, but dissemination to the head was comparable to that of its DEN4 wild type parent. The mean virus titer in the bodies (2.4 log₁₀ PFU/body) or heads (3.0 log₁₀ PFU/head) of rDEN1/4(ME)-infected mosquitoes approximated the values observed for wild type DEN4. However, presence of the Δ30 mutation in rDEN1/4Δ30(ME) conferred a profound reduction in mosquito infectivity. Virus infection was not detected in the bodies (or heads) of the 24 mosquitoes that were fed an infectious bloodmeal containing rDEN1/4Δ30(ME) suggesting that the rDEN1/4Δ30(ME) chimeric virus would be poorly transmissible by mosquitoes.

Vero cells (African green monkey kidney) were grown in OptiPro SFM (Invitrogen, Grand Island, NY) supplemented with 4 mM L-glutamine (Invitrogen). HuH-7 cells (human hepatoma) were maintained in D-MEM/F-12 (Invitrogen) supplemented with 10% fetal bovine serum (FBS), 1 mM L-glutamine and 0.05 mg/ml gentamicin (Invitrogen). C6/36 cells (Aedes albopictus mosquito) were maintained at 32°C in Minimal Essential Medium (MEM) containing Earle's salts and 25 mM HEPES buffer (Invitrogen) and were supplemented with 10% FBS, 2 mM L-glutamine, and 0.1 mM non-essential amino acids (Invitrogen). The DEN1 Puerto Rico/94 virus was provided by Dr. Duane Gubler (John A. Burns School of Medicine, University of Hawaii).

Chimeric viruses have been generated in which the CME and ME genes of the DEN4 p4 and p4Δ30 cDNA clones have been replaced with the corresponding genes derived from DEN1 Puerto Rico/94 (Figure 1). The genomic region within nucleotides 72 – 2353 was amplified by RT-PCR from the DEN1 Puerto Rico/94 genome using DNA primers to introduce a single nucleotide change at 2344 to create a translationally-silent XhoI restriction enzyme recognition site (Figure 1B). The resulting PCR fragment was cloned blunt-ended into vector pCR2.1 (Invitrogen) from which a 2255 bp fragment was obtained following partial digestion with BglII (DEN1 nucleotide 86) and XhoI (DEN1 nucleotide 2341). This 2255 bp fragment was cloned into a modified pUC118 vector containing the AscI – BglII fragment (SP6 promoter and DEN4 5'-UTR) from plasmid p4 (GenBank Accession number, AY648301). To generate a 5' untranslated region that would be identical to that of DEN4, site-directed mutagenesis was used to insert the nucleotides GAAAA immediately upstream of the AUG initiation codon for the polyprotein (Figure 1B, Junction 1), and the resulting plasmid was designated as pUCBXR-init.

To generate full-length CME chimeric cDNA plasmids, the AscI – XhoI region of p4 or p4Δ30 was replaced with the corresponding fragment derived from DEN1 subclone pUCBXR-init. Initially, ligation of the AscI – XhoI fragment of pUCBXR-init into p4 or p4Δ30 failed to yield genetically stable full-length cDNA plasmids. However, ligation in the presence of a synthetic DNA linker molecule containing termination codons in each of the forward and reverse coding frames and flanked by XhoI overhangs yielded stable plasmids p4-D1L-CME and p4Δ30-D1L-CME. The nucleotide and amino acid sequences of the resulting junctions are shown in Figure 1B.

To generate ME chimeric cDNA plasmids, site-directed mutagenesis of pUCBXR-init was used to introduce a unique PstI restriction enzyme recognition site at DEN1 nucleotide 400 by substituting nucleotides 399–401 (UGA → CAG) immediately downstream of coding sequence for the trypsin cleavage site separating the C-protein from its anchor region. Introduction of the PstI cleavage site is translationally silent for the DEN4 portion of the chimeric molecule; however, DEN1 C-protein amino acids 102 – 103 are changed from Val – Thr to Ala – Ala in the chimeric molecule (Figure 1B). A 1939 bp PstI – XhoI fragment derived by partial digestion of the resulting plasmid was cloned into the PstI – XhoI window of a pUC119 clone containing the AscI – XhoI region of DEN4 clone p4 to create plasmid pUC-D1(ME). To generate full-length ME chimeric cDNA plasmids, the AscI – XhoI region of p4 or p4Δ30 was replaced with the corresponding fragment derived from DEN1 subclone pUC-D1(ME). The resulting cDNA plasmids, p4-D1-ME and p4Δ30-D1-ME, were genetically stable in E. coli and did not require the XhoI synthetic linker fragment. The genomic region of each chimeric cDNA was sequenced as previously described 9 and GenBank accessions were assigned as follows (plasmid: accession numbers): p4-D1L-CME: EF456758, p4Δ30-D1L-CME: EF456759, p4-D1-ME: EF456756, p4Δ30-D1-ME: EF456757.

Plasmids were linearized with Acc65I (isoschizomer of KpnI which cleaves leaving only a single 3' nucleotide), and transcribed in vitro using SP6 polymerase. Prior to linearization, the XhoI linker was excised from p4-D1L-CME and p4Δ30-D1L-CME to restore the intact coding sequence. Briefly, the p4-D1L-CME and p4Δ30-D1L-CME plasmids were digested with XhoI and purified to remove the short linker. The plasmids were then circularized by DNA ligase and linearized by Acc65I.

Purified transcripts were transfected into Vero or C6/36 cells using DOTAP liposomes (Roche, Indianapolis, IN). After recovery, viruses were passaged in Vero cells until virus titers reached approximately 10⁶ PFU/ml. Viruses were subsequently biologically cloned by two or three terminal dilutions before experimental stocks were grown in Vero cells. The nucleotide sequence of each recovered virus was determined as previously described 9.

For analysis of replication in tissue culture, confluent 75 cm² flasks of Vero cells were infected with virus at a multiplicity of infection of 0.01. Aliquots of 0.5 ml were removed from flasks daily for seven days. After addition of SPG stabilizer (final concentration: 218 mM sucrose, 6 mM L-glutamic acid, 3.8 mM monobasic potassium phosphate, and 7.2 mM dibasic potassium phosphate, pH 7.2), samples were frozen on dry ice and stored at -80°C. Virus titer was determined by plaque assay on Vero cells. The limit of detection was 0.7 log₁₀ PFU/ml.

For analysis of virus replication in SCID-HuH-7 mice, four to six week-old SCID mice (Tac:Icr:Ha(ICR)-Prkdc^scid) (Taconic, Germantown, NY) were injected intraperitoneally with 0.1 mL phosphate-buffered saline containing 10⁷ HuH-7 cells 32. Five to six weeks after transplantation, tumor-bearing mice were infected by direct inoculation into the tumor with 10⁴ PFU of virus in 50 μl Opti-MEM I (Invitrogen). Serum was collected from infected mice on day 7 post-infection and frozen at -80°C. Virus titer was determined by plaque assay in Vero cells 9.

DEN1 viruses were evaluated for replication and immunogenicity in rhesus macaques using previously described methods 9. Dengue virus seronegative monkeys were inoculated subcutaneously with 10⁵ PFU of virus diluted in L-15 medium (Invitrogen) or with a mock inoculum of L-15 medium. Serum was collected on days 0–6, 8, 10 and 28 after inoculation and stored at -80°C. Virus titer in serum was determined for each day by plaque assay in Vero cells and serum neutralizing antibody titer was determined for days 0 and 28 by plaque reduction neutralization test in Vero cells 9. On day 35 post-infection, all monkeys were challenged by subcutaneous infection with 10⁵ PFU of DEN1 wild type virus. Serum was collected on days 0–8 and on day 10 for virus titration and frozen at -80°C.

Virus replication was evaluated in Aedes aegypti provided by Nancy McLean-Cooper (Walter Reed Army Institute for Research). Mosquitoes were fed an infectious bloodmeal and incubated for 21 days as previously described 33. Assuming a 2 ul bloodmeal, 4.3 log₁₀ PFU of virus was ingested by the mosquitoes. After freezing, heads and bodies were dissected and homogenized in 250 ul suspension buffer (Hank's balanced salt solution). Virus titer in homogenates was determined by plaque assay in Vero cells.