Full genome sequence analysis of a 1-7-4-like PRRSV strain in Fujian Province, China

Clinical case

In September 2018, severe reproductive and respiratory disease was observed in piglets in a farm. The affected pigs had respiratory distress, and high abortion rate (25%) and mortality (40%).

Strain identification and nucleotide sequencing

PRRSV infection was confirmed by Real-time RT-PCR testing of the serum of affected pigs according to the manufacturer’s instructions. The ORF5 sequence RFLP pattern was inferred according to Wesley et al. The complete genomic sequences of FJ0908 were amplified as described previously (Zhang et al., 2018). The PCR products were purified and cloned into pGEM-T Easy according to the manufacturer’s instructions (Promega, Madison, WI, USA) and three recombinant clones of every fragment were sequenced by Ruibo Life Technologies Corporation (Beijing, China).

Complete genomic sequence analysis

Fifty-four representative type 2PRRSV sequences in GenBank were utilized in phylogenetic analyses (Table 1). Multiplex sequence alignments were performed using CLUSTAL X (version 1.83) and the phylogenetic relationships were assessed by MEGA 6.0 as described (Liu et al., 2017b). The ORF5 sequences were classified according to the global PRRSV classification systems (Shi et al., 2010).

Recombination events were detected using Simplot v 3.5.1 and the boot scanning analysis was performed with a 200-bp window, sliding along the genome alignments with a step size of 20 bp.

Complete genomic sequence analysis

The genomes of FJ0908 (GenBank accession no. MK202794) was 15,112 nt in length, excluding the poly (A) tail at the 3′ end, 300 nt shorter than the genome of the prototypic VR2332. Genome alignments revealed that FJ0908 shared 83.6% identity with JXA1, 84.7% with VR2332, 82.2% with QYYZ, 86.3% with NADC30, 97.3–97.6% with 1-7-4 PRRSV family (IA/2014/NADC34, IA/2015/NADC35) and 98.7% with LNWK130 (Table 2).

The results also showed that 5′-UTR, ORF1a, ORF1b, ORFs 2-7 and the 3′-UTR of FJ0908 shared 97.2–98.2% nucleotide homology with 1-7-4 PRRSV family (IA/2014/NADC34, IA/2015/NADC35), which was higher than the homology shared with other representative strains, indicating that FJ0908 strain belonged to 1-7-4 PRRSV. ORF1a and ORF1b encode 16 nsps of PRRSV, Nsp1β and Nsp2 are the most variable protein products among these nsps (Table 2). ORFs 2 to 7 encode the PRRSV structural proteins, among these structural proteins, GP2, GP3, GP4, GP5a and GP5 exhibited the most variance (Table 2).

Amino acid analysis of Nsp2

Nsp2 contains different deletions and insertions, as compared to VR2332 and is the most variable protein in PRRSV genome Liu et al., 2017b; Li et al., 2011). Strikingly, the nsp2 gene of the FJ0908 was 2,640 nt in length and encoded 880 aa, with a 100 aa deletion (aa329–428) within nsp2, as compared to VR2332, and the deletion pattern was consistent with most of 1-7-4 PRRSVs.

Antigenic analysis of GP2-GP5

The antigenic regions (ARs) and glycosylation sites within the GP2, GP3, GP4, and GP5 proteins of FJ0908 were predicted and compared to OH/2014/ISU-7 IA/2014/ISU-8, IA/2014/NADC34, IA/2015/NADC35, IA/2015/NADC36, LNWK96, LNWK130, NV-21, NV-25, VR2332, CH-1a, JXA1 and NADC30.

In GP2, two antigenic regions (AR 41–55 and AR123–135) were confirmed in type 2 PRRSV (De Lima et al., 2006). The predicted AR at aa 41–55 were highly conserved and no aa substitution was detected in AR123–135. In GP3, four predicted antigenic regions (AR32–46, AR 51–105, AR 111–125, and AR 137–159) were proven (De Lima et al., 2006; Zhou et al., 2006; Wang et al., 2014). The AR comprising aa32–46, aa51–105, aa111–125, and aa137–159 of FJ0908 was most similar to 1-7-4 representative PRRSVs including LNWK130, but differed from VR2332, CH-1a, JXA1 and NADC30. GP4 has one predicted AR at aa51–65 (De Lima et al., 2006). FJ0908 had 1-2 aa substitutions as compared to the 1-7-4 representative PRRSVs, but had 2-6 aa substitutions as compared to VR2332, CH-1a, JXA1 and NADC30 and LNWK96. Additionally, no aa substitution was detected in AR51–65 between FJ0908 and LNWK130. The putative glycosylation sites in GP2-4 were completely conserved among the investigated strains, except for GP2 of isolate IA/2015/NADC36 that had one substitution N¹⁸⁴D.

GP5 is the major envelope protein encoded by ORF5 and is the most variable PRRSV protein. Sequences alignments of GP5 revealed that FJ0908 shares 87.0%, 86.0–86.5%, 85.5%, 88.0–90.5% and 97.5% amino acid identity with VR-2332-like (VR2332 and BJ-4), JXA1-like (JXA1 and HuN4), QYYZ-like (FJFS and QYYZ), NADC30-like strains (NADC30 and ISU30), and 1-7-4-like (IA/2014/NADC34, IA/2015/NADC35), respectively (Table 2). Furthermore, the restriction sites analysis showed that the ORF5 RFLP of FJ0908 has the same 1-7-4 pattern [1 (MluI = 0 sites), 7 (HindII = nt 88, 219, 360), 4 (SacII = nt 24, 555)] (Fig. S1).

GP5 has six ARs (AR1–15, AR27–35, AR37–51, AR149–156, AR166–181, and AR192–200) (De Lima et al., 2006; Zhou et al., 2009). In GP5, the N-terminus ARs (AR1-15 and AR27-35) were very variable among all strains and only three aa substitutions (K⁴N, Q¹³R and L¹⁵P) were found in the two antigenic regions between FJ0908 and LNWK130, whereas the other four ARs (AR 37–51, AR 149–156 , AR 166–181, and AR 192–200) were conserved. GP5 had differential predicted N-glycosylation among PRRSV strains. FJ0908 possessed five predicted sites (N32, N44, N51, N57 and N59). Although N-glycosylation at N57 was not a novel finding, N57 was detected in most of the 1-7-4 sequences. The results also revealed that FJ0908 had the same N-glycosylation pattern as LNWK96 and LNWK130.

Phylogenetic analysis

PRRSV ORF5 is the most variable and has been used as a marker of PRRSV genetic variability. Based on global PRRSV classification systems, type 2 PRRSV was divided into nine monophyletic lineages (1-9) and lineage1 was further classified into nine sublineages (1.1–1.9) (Shi et al., 2010; Guo et al., 2018). The ORF5-based phylogenetic tree showed that FJ0908, as well as 1-7-4 isolates including LNWK130, were clustered in sublineage 1.5 (Fig. 1A). Whole genome phylogenetic analysis also indicated that FJ0908 was most closely related to a genetic cluster in 1-7-4-like lineage 1 (Fig. 1B).

Recombination analysis

To test for possible recombinant events within FJ0908 strain, we performed recombinant detection using SimPlot v3.5.1 software. From the similarity plot, two recombination breakpoints within the FJ0908 genome were identified, which were located in Nsp1 (nt 760 and nt 1,300) (Fig. 2A). To further confirm the putative recombination events, phylogenetic trees for each of the sequence regions identified during the analysis were generated, we identified two recombination breakpoints located in nsp1 (nt 760 and nt 1,300) (Figs. 2B and 2C). The two breakpoints separated the genome of FJ0908 into 3 regions. For FJ0908, the region between the breakpoints (nt 760–1,300) is closely related to ISU30 strain, the two regions between the breakpoints (nt 1–759 and nt 1,301–15,534) are closely related to IA/2014/NADC34. Collectively, the above results suggested that FJ0908 derived from recombination between IA/2014/NADC34 and ISU30 (Fig. 2). Moreover, LNWK130 strain (1-7-4-like PRRSV) firstly identified in Liaoning Province, China was also deriving from the recombination of 1-7-4 isolates and ISU30 (Zhang et al., 2018). However, the recombination pattern was different between LNWK130 and FJ0908. For LNWK130, the two breakpoints (nt 760–1,300 and nt 1–759) are closely related to ISU30 strain, one breakpoints (nt 1301–15534) is closely related to IA/2014/NADC34. Additionally, sequences alignments of the recombination region (nt 1-759) revealed that FJ0908 shares 98.8%, 94.8% and 92.6% nucleotide identity with IA/2014/NADC34, ISU30 and NADC30, respectively, in contrast, LNWK130 shares 95.8%, 95.9% and 93.1% nucleotide identity with IA/2014/NADC34, ISU30, and NADC30, respectively. Sequences alignments of the recombination region (nt 760-1,300) revealed that FJ0908 shares 92.8%, 94.3% and 82.0% nucleotide identity with IA/2014/NADC34, ISU30 and NADC30, respectively, similarly, LNWK130 shares 89.4%, 93.0% and 80.2% nucleotide identity with IA/2014/NADC34, ISU30, and NADC30, respectively. For all of the nsp sequences and structural proteins, the most variable regions were found in nsp1β, nsp3, nsp5, nsp6, nsp8, GP3 and GP5 between FJ0908 and LNWK130.