Monkeypox virus genomic accordion strategies

Abstract

The 2023 monkeypox (mpox) epidemic was caused by a subclade IIb descendant of a monkeypox virus (MPXV) lineage traced back to Nigeria in 1971. Person-to-person transmission appears higher than for clade I or subclade IIa MPXV, possibly caused by genomic changes in subclade IIb MPXV. Key genomic changes could occur in the genome's low-complexity regions (LCRs), which are challenging to sequence and are often dismissed as uninformative. Here, using a combination of highly sensitive techniques, we determine a high-quality MPXV genome sequence of a representative of the current epidemic with LCRs resolved at unprecedented accuracy. This reveals significant variation in short tandem repeats within LCRs. We demonstrate that LCR entropy in the MPXV genome is significantly higher than that of single-nucleotide polymorphisms (SNPs) and that LCRs are not randomly distributed. In silico analyses indicate that expression, translation, stability, or function of MPXV orthologous poxvirus genes (OPGs), including OPG153, OPG204, and OPG208, could be affected in a manner consistent with the established "genomic accordion" evolutionary strategies of orthopoxviruses. We posit that genomic studies focusing on phenotypic MPXV differences should consider LCR variability.

Fig. 1. De novo assembly of subclade IIb lineage B.1 monkeypox virus (MPXV) genome sequence 353R.

A visual representation of the fully annotated MPXV isolate 353R genome (based on the subclade IIb lineage A reference isolate MPXV-M5312_HM12_Rivers genome sequence annotation). Shown are (from the outside to the inside): high-quality genome (HQG) hybrid assembly (wide outer light blue ring); sequencing coverage distribution graph (thin ragged line [green: ≥10,000x, 99.42%; black: 1000x–10,000, 0.28%; orange: <1000x–10, 0.1%; red: <10–0, 0.2%]); orthologous poxvirus gene (OPG) annotations according to the standardized nomenclature (lettering and shaded boxes [orange: ANK/PRANC (N-terminal ankyrin protein with PRANC domain) inverted terminal repetition [ITR] regions; gold: Bcl-2 domain; blue: BTB/Kelch domains; green: housekeeping; purple: other; pink: TNFR and/or PIE domains); contigs from NovaSeq, MiSeq, and nanopore sequencing (wide inner gray rings). Additionally, radial lines and shading that originate in the center and reach outward on the white background indicate low-complexity regions (LCRs; royal blue) and areas with more change (light blue).

Fig. 2. Characterization and validation of non-randomly distributed low-complexity regions (LCRs) in monkeypox virus (MPXV) genome sequence 353R.

a LCR3 sequence validation using MPXV 353R nanopore sequencing data and 15 additional raw data sequencing reads downloaded from the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA). Box plots representing median (middle line), 25th and 75th percentile (box), and 5th and 95th percentile (whiskers) outliers are plotted as single dots. Blue dashed lines represent trends of LCR3 repeats in MPXV lineages (n = 16 sequences). (Wilcoxon rank-sum test p-values are indicated in the graph.) b LCR pair 1/4 sequence validation using MPXV 353R nanopore sequencing data and 20 additional raw data sequencing reads downloaded from NCBI SRA. Box plots representing median (middle line), 25th and 75th percentile (box), and 5th and 95th percentile (whiskers) outliers are plotted as single dots. Blue dashed lines represent trends of LCR1/4 repeats in MPXV lineages (n = 21 sequences). (Wilcoxon rank-sum test p-values are indicated in the graph.) Detailed information on the represented materials, along with originator and epidemiological data, is provided in Supplementary Data 6.

Fig. 3. Low-complexity regions (LCRs) are non-randomly distributed in the monkeypox virus (MPXV) genome.

a Frequencies (mean +/- standard error) at which LCRs occur in orthologous poxvirus genes (OPGs) of different functional groups are represented in different colors (n = 236 sequences). Shown are functional classes in which pairwise comparisons had a significantly different frequency than other groups. (Multiple pairwise Wilcoxon test false discovery rate [FDR] corrected p-value < 0.05 and specific p-values are indicated in the graph.) b Entropy value distribution for short tandem repeats (STRs) in LCRs (left) and single-nucleotide polymorphisms (SNPs; right). Box plots representing median (middle line), 25th and 75th percentile (box), and 5th and 95th percentile (whiskers) outliers are plotted as single dots; blue = LCR, yellow=SNPs. (Student’s t-test-specific p-values are indicated in the graph.) c Distributions of the pairwise inter-sample Euclidean distances for each STR in LCRs (Supplementary Data 3). SNPs in the box plot represent the distribution of average Euclidean distances of each variable position along the MPXV genome. Box plots representing median (middle line), 25th and 75th percentile (box), and 5th and 95th percentile (whiskers) outliers are plotted as single dots; blue = LCR, yellow = SNPs. (Multiple pairwise Wilcoxon test false discovery rate [FDR] corrected p-value < 0.05 and specific p-values are indicated in the graph.).

Fig. 4. Low-complexity regions (LCRs) might be more phylogenetically informative than single-nucleotide polymorphisms (SNPs) for inter-host sequence analysis.

Monkeypox virus (MPXV) population genomics within the biological specimen (intra-host) and across different specimens (inter-host). a Panel shading indicates the number of reads supporting each LCR for each sample. Only paired reads that include a perfect match to both flanking regions were counted; the gradient shows the maximum value in black and the minimum value (n = 1) in the lightest blue. Samples without coverage are indicated in gray. b Comparison of LCR allele frequency for samples 353R and 349R. Only LCRs with at least 10 supporting paired reads including both flanking regions were counted; only alleles with a frequency of 0.03 or higher were considered. The gradient shows the maximum value in black and the minimum value (n = 0.03) in the lightest blue. c Comparison of LCR allele frequency in all samples for LCRs with good coverage (7, pair 10/11, 12, 13, 14, 19, 20, and 21). Only LCRs with at least 10 supporting paired reads including both flanking regions were counted; only alleles with a frequency of 0.03 or higher were considered. The gradient shows the maximum value in black and the minimum value (n = 0.03) in the lightest blue.

Fig. 5. Conservation and variation in proteins encoded by orthologous poxvirus gene (OPG) and codon usage analysis in OPG low-complexity regions (LCRs).

MetaLogo visualization of conserved and varying amino-acid residues in OPG-encoded proteins among monkeypox virus (MPXV) clade I, subclade IIa, and subclade IIb, with homologous and nonhomologous sites highlighted. Genome sequences used for each alignment can be found in Supplementary Data 5. a OPG153 LCR7-derived variability. CMLV, camelpox virus; VACV, vaccinia virus; VARV, variola virus; CPXV, cowpox virus; MPXV, monkeypox virus. b OPG208 LCR3-derived variability; c Frequencies (mean +/− standard error) of TAC codon usage in the OPG208 ORF using an alternative start codon (ATG) in MPXV clade IIb (n = 7 sequences, accession versions: OX044336.2 [generated in this study], ON563414.3, ON568298, ON736420, ON674051, ON649879, and ON676707.1) to average TAC usage in MPXV and humans (one-tailed Wilcoxon test p-value = 0.0156) (*=p-value < 0.05); and d OPG204 LCR21-derived variability. Visualizations created at Biorender.com and Geneious version 2022.2 created by Biomatters.

Fig. 6. Read coverage analyses with RNA-seq data suggest that low-complexity regions (LCRs) 3 and 21 are transcribed during monkeypox virus (MPXV) infection.

a–g The positional depth of translation starts, panning from the 40 nucleotides upstream to the 10 nucleotides downstream of the first alternative start codon (ATG), of top early genes. The vertical blue line represents the first ATG of the gene, and the arrow shows the spans of the first 10 (canonical) coding nucleotides. g, h If present, red lines represent ATGs, evidenced from transcriptomic data demonstrating that both OPG204 and OPG208 ATGs, located upstream of the canonical ones, are transcribed during MPXV subclade IIb infection. ORF, open reading frame.