Paradigm Shift Intervention Monitoring
twitter
Commentary
South African
Bat Beta2c Distinct From Human MERS-CoV
Recombinomics
Commentary 22:15
July 25, 2013
The above comments are from an EID paper describing a beta2c isolate, PML/2011/Neozul/RSA/2012, which is the bat sequence that is most closely related to human MERS-CoV sequences. Although the actual sequence at the above link has not been released, it is likely that it is marginally more related to the human sequences than the three bat sequences from Europe (UKR-G17/Pipnat/UKR/2011, 8-691/Pipnat/ROU/2009, 8-724/Pippyg/ROU/2009) based on the phylogenetic tree of a 138 nt segment from the spike gene.
There currently are nine full sequences from human MERS-CoV sequences which are closely related to each other, but are decades or centuries away from the bat sequences, including the isolate from South Africa. The human sequences form two major sub-clades composed of the 2 earliest sequences (Jordan-N3 and EMC) in one subclade and the 7 more recent sequences in a second clade, which can be further divided into two sub-grioups involving England1 and the UAE sequence on one arm, with England2 and the four sequence from Al-Hasa (Al-Hasa_1, Al-Hasa_2, Al-Hasa_3, Al-Hasa_4) on the other arm. The differences between the two subclades are clustered in the polymerase gene, signaling evolution by recombination, but all of the human sequences have an identity of at least 99.6% with the human consensus sequence, based on the nucleotide sequence.
The above comment on the South African bat sequence, like the earlier paper on the European bat sequence, largely ignores the significant nucleotide difference between the bat and human sequences and instead uses the predicted protein sequence of a highly conserved region of a highly conserved gene to highlight similarities between the bat and human fragments.
The human sequences however, demonstrate the superior value of the data provided by the nucleotide sequences, which is greater than 30,000 positions for each of the human isolates. The four Al Hasa sequences differ from each other by 3 nucleotide or less, yielding an identity greater than 99.99%. Similarly, the identities between th e 7 more recent sequences is approximately 99.9%, once again showing the strong nucleotide conservation between sequences from cases in Saudi Arabia and Qatar, while the 99.6% extends the conservation to 2012 sequences from Jordan and Saudi Arabia.
The differences between the bat and human sequences are easily seen using the short segments described in the recent papers on the bat sequences. The 816 nt segment of the polymerase gene (RdRp) described above is identical in all 7 of the most recent human sequences and has 1 difference with Jordan-N3 and 2 differences with EMC. In contrast the most closely related bat sequence, UKR-G17/Pipnat/UKR/2011, has 101 differences (87.6% identity).
Similarly the spike sequence of 138 nucleotides is identical in all 9 human sequences, but there are 30 differences (78.3% identity) in the most closely bat sequence from Europe.
Although the South African sequences have not been released yet, the RdRp identity is likely to be near 90% while the spike fragment is likely to be near 80%, indicating the South African bat sequence, like the European bat sequences, is decades or centuries away from the human sequences, signaling the absence of any recent jumps to humans to produce the public MERS-CoV sequences.
Recombinomics
Presentations
Recombinomics
Publications
Recombinomics
Paper
at Nature Precedings