Polyphyletic origin of MERS coronaviruses and isolation of a novel clade A strain from dromedary camels in the United Arab Emirates

Figures & data

Figure 1 Geographical distribution of dromedaries in the United Arab Emirates infected with MERS-CoV strains that were isolated in the present study. The number of dromedaries corresponds to the number of MERS-CoV isolates from the sampling location.

Table 1 Clinical and epidemiological data of the 10 dromedaries with MERS-CoV strains isolated and sequenced in this study

Strains	Clade/lineage	Sex/age	Date of sample collection	Clinical samples	Live/dead dromedary (causes of death)	Origin of dromedary
D1164.1/14	B1	F/5 months	6 Jun, 2014	Nasal swab	Live	Breeding herd returning from Saudi Arabia
D2597.2/14	B1	NA/6 months	13 Dec, 2014	Nasal swab	Live	Dairy farm in Dubai
D252/15	B1	F/1 month	30 Jan, 2015	Nasal swab	Dead (white muscle disease, isosporosis abdominal edema disease)	Dairy farm in Dubai
D374/15	B1	M/1 month	12 Feb, 2015	Nasal swab	Dead (colisepticemia)	Dairy farm in Dubai
D383/15	B1	F/1 month	14 Feb, 2015	Nasal swab	Dead (colisepticemia)	Dairy farm in Dubai
D389/15	B1	M/15 days	15 Feb, 2015	Nasal swab	Dead (septicemia)	Dairy farm in Dubai
D998/15	A2	F/1 month	23 April, 2015	Nasal swab	Dead (white muscle disease and clostridiosis)	Dairy farm in Dubai^a
D1157/15	B5	F/1 month	12 May, 2015	Nasal swab	Dead (septicemia)	Dairy farm in Dubai
D1271/15	B3	M/3 month	29 May, 2015	Nasal swab	Dead (high fever, purulent nasal discharge, pneumonia)	Dairy farm in Dubai
D1189.1/15	B3	F/6-9 months	18 May, 2015	Nasal swab	Dead (nasal discharge)	Umm Al Quwain

Abbreviation: Not available, NA.

^a A special dairy farm with camel mothers from Pakistan.

Figure 2 Maximum-likelihood phylogeny based on the complete genome sequences of 182 MERS-CoV strains. A general time-reversible model of nucleotide substitution with estimated base frequencies, the proportion of invariant sites, and the gamma distribution of rates across sites were used in the maximum-likelihood analysis. Bootstrap values are shown next to the branches. The scale bar indicates the number of nucleotide substitutions per site. MERS-CoVs from dromedaries are indicated in black circles. The ten MERS-CoV strains sequenced in the present study are colored: blue, Dubai; brown, Umm Al Quwain; green, Saudi Arabia. MERS-CoVs isolated from the United Arab Emirates are indicated by gray boxes.

Figure 3 Maximum-likelihood phylogeny based on ORF1a, ORF1b and S gene sequences of 182 MERS-CoV strains. TIM1+I+G, GTR+I+G, and TK 2+I substitution models were selected for the ORF1a, ORF1b and S gene trees, respectively. Bootstrap values are shown next to the branches. The scale bar indicates the number of nucleotide substitutions per site. MERS-CoVs from dromedaries are indicated in black circles. The 10 MERS-CoV strains sequenced in the present study are colored: blue, Dubai; brown, Umm Al Quwain; green, Saudi Arabia. MERS-CoVs isolated from the United Arab Emirates are indicated by gray boxes.

Table 2 Comparison of amino acid substitutions between different clades and lineages

Clade/lineage	Species	Strain	nsp2		nsp3									nsp4	S1		S2
			588	726	1000	1024	1055	1066	1370	1574	1717	2114	2481	2780	26	194	1020	1158
										CD	CD				NT	NT	HR1
A/1	Human	EMC/2012	A	K	T	S	P	K	M	G	L	A	F	A	A	Y	Q	S
A/1	Human	Jordan-N3 /2012	A	K	T	S	P	K	M	G	L	A	F	A	A	Y	Q	S
A/2	Camel	D998/15	T	K	T	F	P	E	M	K	L	V	C	A	V	Y	Q	S
A/2	Camel	NRCE- HKU205	A	K	T	F	P	E	M	R	L	V	C	A	V	Y	Q	S
B/5	Human and Camel	All	T	N	I	S	S	K	I	K/E	I	V	F	V	V	H	H/R	A
B/6	Human and Camel	All	T	N	V	S	S	K	I	K	I	V	F	V	V	H	R	A
B/4	Human and Camel	All	T	N	V	S	S	K	I	K/G/E	I	V	F	V	V	H	R	A
B/3	Human and Camel	All	T	N	V	S	S	K	I	K/G	I	V	F	V	V	H	R	A
B/2	Human and Camel	All	T	N	V	S	S	K	I	K	I	V	F	V	V	H	R	A
B/1	Human and Camel	All	T	N	V	S	S	K	I	K/E	I	V	F	V	V	H	R	A

Abbreviations: Catalytic domain, CD; N-terminal, NT; heptad region 1, HR1.

Figure 4 Detection of potential recombination events by bootscan analysis and multiple alignments. Bootscanning was conducted with Simplot version 3.5.1 (F84 model; window size, 1,500 bp; step, 200 bp) on a gapless nucleotide alignment. (A) D1271/15 and D1189.1/15 were used as the query sequences and compared with the genome sequences of a lineage 4 MERS-CoV strain Hafr-Al-Batin 1 (red, KF600628), a lineage 1 MERS-CoV strain UAE/Abu Dhabi_UAE_9 (green, KP209312) and a lineage 5 MERS-CoV strain D1157/15 (blue, KX108944). (B) D2731.3/14 was used as the query sequence and compared with the genome sequences of a lineage 1 MERS-CoV strain, FRA/UAE (green, KF745068), a lineage 4 MERS-CoV strain, KFMC-4 (red, KT121575) and a lineage 5 MERS-CoV strain, England-Qatar/2012 (blue, KC667074).

Figure 5 Maximum-likelihood trees constructed using nucleotide sequences of non-recombination regions (A), and recombination regions (B) detected using RDP4. The non-recombination region is approximately positions 1 to 14957 and 28836 to 30087, and the recombination region is approximately positions 14958 to 28835, with position numbering based on the D1189.1/15 genome sequence. Lineages are indicated by colored boxes: lineage 1, green; lineage 3, red; and lineage 5, blue.

Figure 6 Estimation of time to the most recent common ancestor for MERS-CoV. The time-scaled phylogeny was summarized from all Markov chain Monte Carlo phylogenies of the concatenated coding regions (ORF1ab, S, E, M and N) of 90 phylogenetically distinct MERS-CoV genomes, which were analyzed under the relaxed-clock model with an uncorrelated lognormal distribution in BEAST version 1.8.0. The 10 MERS-CoV genomes sequenced in this study are in boldface.