Amplicon and Metagenomic Analysis of Middle East Respiratory Syndrome (MERS) Coronavirus and the Microbiome in Patients with Severe MERS

Abstract

Middle East respiratory syndrome coronavirus (MERS-CoV) is a zoonotic infection that emerged in the Middle East in 2012. Symptoms range from mild to severe and include both respiratory and gastrointestinal illnesses. The virus is mainly present in camel populations with occasional zoonotic spill over into humans. The severity of infection in humans is influenced by numerous factors, and similar to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), underlying health complications can play a major role. Currently, MERS-CoV and SARS-CoV-2 are coincident in the Middle East and thus a rapid way of sequencing MERS-CoV to derive genotype information for molecular epidemiology is needed. Additionally, complicating factors in MERS-CoV infections are coinfections that require clinical management. The ability to rapidly characterize these infections would be advantageous. To rapidly sequence MERS-CoV, an amplicon-based approach was developed and coupled to Oxford Nanopore long read length sequencing. This and a metagenomic approach were evaluated with clinical samples from patients with MERS. The data illustrated that whole-genome or near-whole-genome information on MERS-CoV could be rapidly obtained. This approach provided data on both consensus genomes and the presence of minor variants, including deletion mutants. The metagenomic analysis provided information of the background microbiome. The advantage of this approach is that insertions and deletions can be identified, which are the major drivers of genotype change in coronaviruses. IMPORTANCE Middle East respiratory syndrome coronavirus (MERS-CoV) emerged in late 2012 in Saudi Arabia. The virus is a serious threat to people not only in the Middle East but also in the world and has been detected in over 27 countries. MERS-CoV is spreading in the Middle East and neighboring countries, and approximately 35% of reported patients with this virus have died. This is the most severe coronavirus infection so far described. Saudi Arabia is a destination for many millions of people in the world who visit for religious purposes (Umrah and Hajj), and so it is a very vulnerable area, which imposes unique challenges for effective control of this epidemic. The significance of our study is that clinical samples from patients with MERS were used for rapid in-depth sequencing and metagenomic analysis using long read length sequencing.

Keywords: MERS-CoV; MinION; metagenomics; sequencing.

FIG 1

Location of conserved primer pairs (Table 4) on the MERS-CoV genome and position compared with MERS-CoV genes. Primer pairs can be used to generate amplicons of various lengths, including 30, 15, and 8 amplicons as indicated. Appropriate genes are indicated on the MERS-CoV genome.

FIG 2

Agarose gel electrophoresis of amplicons generated using 30 (A), 15 (B), and 8 (C) combinations of primers pairs. These primer pairs were used to generate amplicons in combination with reverse transcription of RNA extracted from MERS-CoV-infected cells and are indicated above the appropriate lane on the gel image. Molecular size markers are indicated to the left.

FIG 3

Agarose gel electrophoresis of amplicons generated using 30 (A) and 15 (B) combinations of primers pairs. These primer pairs were used to generate amplicons in combination with reverse transcription of RNA extracted from nasal aspirates taken from patients with MERS. Primer pairs are indicated above the appropriate lane on the gel image. Molecular size markers are indicated to the left.

FIG 4

Read depth analysis of 30 (A) and 15 (B) amplicons sequenced on a single flow cell using an Oxford Nanopore MinION device. Coverage of each position on the MERS-CoV genome is indicated on the y axis. The nucleotide length of the MERS-CoV genome is indicated by the genome position on the x axis. The dashed line represents 20× coverage, indicating that above this line each nucleotide was sequenced at least 20 times.

FIG 5

A flow diagram of the bioinformatic pipeline that was used to derive viral genome coverage and minor variation information of viral genomes within patients.

FIG 6

The sequencing reads were mapped to the patient consensus viral genome sequence. The custom script counted the number of each base at each genome position with a quality score of >10. Positions with a depth of <20 were removed from the analysis. This figure shows the proportion of base changes observed compared with the patient’s dominant consensus reference genome. Overall, transitions were observed more frequently than transversions, where C > U is the most observed base change. Patient 10, dark gray; patient 115, light gray; outliers not visualized.

FIG 7

Genome-wide representation of standardized indel polymorphisms for patient (PX) 110 and PX 115. Indels are designated by bars and anchored at the x axis, corresponding to each gene. The standardized indel proportion is ranked in each plot. Where the difference in scale hinders visualization, radial subplots are used to provide a view of genes with the lowest proportion of indels.

FIG 8

Genome-wide representation of indel polymorphisms, demonstrated by smoothened densities (adjust, 0.05; alpha, 0.7) for PX 110 and PX 115.

FIG 9

The top 20 species categorized into genus from 15 patients with severe MERS infection. Human reads were removed from sequencing libraries, and viral and bacterial transcripts were identified using Kraken2. Kraken2 outputs were converted into biom format before importing into R with Phyloseq. The relative abundance of each species is plotted for each patient. +, MERS-CoV reads detected; −, MERS-CoV reads not detected.

FIG 10

Increased and decreased abundance of bacteria in fatal cases of MERS-CoV infections compared with nonfatal cases derived from transcript abundance data. A Phyloseq object was converted into a DESeq2 object, a contrast between fatal and nonfatal outcome was used to calculate log₂ fold change, and values with a false discovery rate (FDR) of <0.01 were plotted. The x axis represents the genus of identified transcripts, unassigned refers to transcripts that could not be assigned at the genus level, and color illustrates the phyla.