Emerging Novel GII.P16 Noroviruses Associated with Multiple Capsid Genotypes

Abstract

Noroviruses evolve by antigenic drift and recombination, which occurs most frequently at the junction between the non-structural and structural protein coding genomic regions. In 2015, a novel GII.P16-GII.4 Sydney recombinant strain emerged, replacing the predominance of GII.Pe-GII.4 Sydney among US outbreaks. Distinct from GII.P16 polymerases detected since 2010, this novel GII.P16 was subsequently detected among GII.1, GII.2, GII.3, GII.10 and GII.12 viruses, prompting an investigation on the unique characteristics of these viruses. Norovirus positive samples (n = 1807) were dual-typed, of which a subset (n = 124) was sequenced to yield near-complete genomes. CaliciNet and National Outbreak Reporting System (NORS) records were matched to link outbreak characteristics and case outcomes to molecular data and GenBank was mined for contextualization. Recombination with the novel GII.P16 polymerase extended GII.4 Sydney predominance and increased the number of GII.2 outbreaks in the US. Introduction of the novel GII.P16 noroviruses occurred without unique amino acid changes in VP1, more severe case outcomes, or differences in affected population. However, unique changes were found among NS1/2, NS4 and VP2 proteins, which have immune antagonistic functions, and the RdRp. Multiple polymerase-capsid combinations were detected among GII viruses including 11 involving GII.P16. Molecular surveillance of protein sequences from norovirus genomes can inform the functional importance of amino acid changes in emerging recombinant viruses and aid in vaccine and antiviral formulation.

Keywords: GII.4 Sydney; GII.P16; Norovirus; clinical outcomes; dual-typing; herd immunity; immune antagonism; molecular epidemiology; non-structural proteins; recombinants.

Figure 1

Outbreak distribution and emergence of a novel GII.P16 polymerase among (A) GII.4 Sydney (n = 2355), (B) GII.2 (n = 329) and (C) GII.1, GII.3, GII.10, GII.12, and GII.13* (n = 291). Noroviruses containing a novel GII.P16 polymerase corresponding to GII.4 Sydney, GII.2, GII.10, GII.1, GII.12, and GII.3 are shown in blue. Other polymerase types associated with GII.4 Sydney include GII.Pe (yellow), GII.P4 (grey), extant GII.P16 (orange), and unknown polymerase types (green). Other polymerase types associated with GII.2 include GII.P2 (yellow), GII.Pe (grey), extant GII.P16 (orange), and unknown polymerase types (green). GII.Pg-GII.1, GII.P12-GII.3, GII.Pc-GII.10, and GII.Pg-GII.12 viruses (yellow), GII.P21-GII.3, GII.Pg-GII.10, GII.Pq-GII.13 viruses (grey), GII.P16-GII.3 and GII.P16-GII.13 (extant) viruses (orange), and unknown polymerase types associated with GII.1, GII.3, and GII.13 capsids (green). The blue arrows indicate month and year of first report of novel GII.P16 polymerase associated with noted capsid. *GII.13 is only associated with extant GII.P16 and is included in the figure due to its association with GII.P16.

Figure 2

Time-scaled phylogenetic tree of complete GII.P16 polymerase nucleotide sequences constructed using the Bayesian MCMC method (BEAST v1.10.4) using the GTR + G + I substitution and uncorrelated relaxed clock model. Viruses in the Novel GII.P16 lineage were first detected in 2014 diverging from a common ancestor with Extant B GII.P16 viruses that were first reported in 2010. Viruses in the Extant A GII.P16 lineage were detected from 1975 to 2015. Figure includes all sequences from GenBank (Table S1) and CDC-generated (Table S2) included in this study.

Figure 3

Norovirus polymerase and capsid dual type combinations of (A) Genogroup I (n = 1343) and (B) Genogroup II (n = 8294). Polymerase-capsid combinations denoted with an asterisk (*) were found in the CDC collection only. Polymerase-capsid combinations denoted with a hash (#) were found in GenBank only. NA: Not Assigned. Total polymerase breakdown for each type: GI.P1 (101), GI.P2 (47), GI.P3 (333), GI.P4 (145), GI.P5 (143), GI.P6 (83), GI.P7 (192), GI.P8 (19), GI.P9 (21), GI.Pa (13), GI.Pb (57), GI.Pc (8), GI.Pd (164), GI.Pf (4), GI.Pg (5), GI.Ph (1), GI.Pi (2), GII.P1 (14), GII.P2 (340), GII.P3 (21), GII.P4 (1728), GII.P5 (7), GII.P6 (9), GII.P7 (851), GII.P8 (32), GII.P12 (349), GII.P13 (23), GII.P15 (22), GII.P16 (2755), GII.P17 (425), GII.P20 (3), GII.P21 (276), GII.P22 (86), GII.P23 (69), GII.P24 (20), GII.P26 (1), GII.Pa (15), GII.Pc (5), GII.Pe (1008), GII.Pf (2), GII.Pg (148), GII.Ph (3), GII.Pj (1), GII.Pk (1), GII.Pm (6), GII.Pn (2), GII.Pp (1), GII.Pq (3). GII.P4 includes Den Haag (49), New Orleans (217), and 1462 sequences not typed to the variant level. Total capsid breakdown for each type: GI.1 (101), GI.2 (47), GI.3 (520), GI.4 (121), GI.5 (175), GI.6 (140), GI.7 (197), GI.8 (18), GI.9 (21), GII.1 (124), GII.2 (1333), GII.3 (580), GII.4 (4247), GII.5 (86), GII.6 (579), GII.7 (183), GII.8 (32), GII.9 (5), GII.10 (22), GII.12 (143), GII.13 (174), GII.14 (88), GII.16 (11), GII.17 (519), GII.20 (8), GII.21 (20), GII.22 (5), GII.23 (69), GII.24 (20), GII.25 (3), GII.26 (1), GIX.1 (22). GII.4 includes Den Haag (43), New Orleans (52), Sydney (2211), and nine typed outside our variant level threshold. Polymerase-capsid combinations that are not shown include GI.P8-GII.4 (1) was only found in GenBank and GI.P untypeable-GI.3 (5) and GI.Pi-GI.untypeable (2) which were found in strains sequenced in this study only. Sequences were downloaded from CaliciNet on September 21, 2018 and from GenBank on October 12, 2018. Sequence length used to type polymerase and capsid was at least 509 nt for GI and 497 nt for GII. The number of sequences available for each combination were calculated into quartiles separately for GI and GII. Each quartile was assigned a color transitioning from lighter to darker shades of blue for GI and green for GII.

Figure 4

Informative and unique amino acid sequences detected among viruses with the novel GII.P16 polymerase. (A) Genomic organization of norovirus with informative sites (locations where >10% of sequences with a different polymerase type differed from those with a novel GII.P16 polymerase that emerged in 2015) are indicated with a vertical line (red, indicates changes at informative sites resulting in an amino acid that was unique to novel GII.P16 strains; green, indicates informative sites that are not unique to novel GII.P16 strains). (B–E) Sequence logos indicating the relative frequency of amino acid occurrence (bits) for informative sites unique to novel GII.P16 viruses within non-structural proteins when compared to Extant B and Extant A GII.P16 polymerases. Unique informative sites shown for (B) NS1/2, (C) NS4, (D) NS6, (E) NS7. (F) Sequence logos of informative sites unique to novel GII.P16-GII.4 Sydney viruses within VP2 when compared to GII.4 Sydney viruses with GII.Pe and GII.P4 New Orleans polymerases. Bottom numbers indicate amino acid position within each protein.

Figure 5

Informative amino acid sequences within the VP1 protein of GII.4 Sydney viruses with and without the novel GII.P16 polymerase. Sequence logos indicating the relative frequency of amino acid occurrence (bits) for locations where >10% of sequences with a different polymerase type differed from those with a novel GII.P16 polymerase for GII.4 Sydney strains with a novel GII.P16 (n = 91 sequences), GII.Pe (n = 172 sequences) or GII.P4 New Orleans (n = 47 sequences) polymerase. Bottom numbers indicate amino acid position within VP1 and colors indicate Shell (green), Protruding one (blue), Protruding 2 (peach) and C-terminal (grey) domains. Sites involved in neutralizing antibody recognition (epitopes A, B, C and D), the NERK motif, and histo-blood group antigen binding (#), as well as those under positive selection (+) are indicated.