H5N1 Genetic Instability in Karo Cluster in Sumatra Indonesia
Recombinomics Commentary
June 25, 2006
The recent workshop in Jakarta on human H5N1 in Indonesia provide additional detail on genetic changes associated with the Karo cluster. The cluster involved 8 members of the same family, the index case, her two sons, here three siblings, and two children of the siblings. Seven of the eight family members died, including the index case. A seven relatives of the index cases were H5N1 positive and the workshop presented sequence information on the six family members who died.
Five of the six developed symptoms are the time of death of the index case. The H5N1 from these five family members were remarkable similar. The H5N1 from the two son, IDN/534H/06 and IDN538H06 had one difference from the consensus sequence for all eight gene segments, NP G555A. the H5N1 from the daughter of the sister of the index case, IDN/535H/06 had one other change, M G953A. All other positions were identical in all five family members. A later isolate from the 10 year old nephew of the index case, IDN/546bH/06 had one additional change HA C29T. The father of the nephew, developed symptoms when his son died.
Analysis of the H5N1 from the nephew's father, IDB/560H/06 had the same change in HA, C29T. This sharing of the HA polymorphism suggested that the nephew infected his father, which was not unexpected because of the disease onset date for the father, as well as the close contact between father and son before the son died.
However, the H5N1 from the father had eight additional polymorphism in HA (C105A, T120C, C396A, G512A, A579G, G729A, C819A, C1276T). In addition, there were three polymorphisms in NA (T608C, T651C, T657C) as well as two polymorphisms in M (A651G and A918G). Thus, the other family member had zero or one difference with consensus H5N1 sequence of the cluster, while the H5N1 from the nephew's father had 14. The high number of changes suggest the involvement of a second virus resulting in a recombinant with the 14 differences.
This explosion in genetic changes over a short time period is cause for concern. The sharing of HA T29C between father and son suggests the isolate from the father was a recombinant, rather than an independent isolate. However, both alternatives require a second infection in a family member within a couple of weeks of the infection of the other family members, who were almost certainly infected by the index case.
The concentration of fatal H5N1 in this family is cause for concern. It was the largest and most deadly cluster in Indonesia to date and involved two distinct H5N1 sequences. Symptoms in the infected father were noted by a reporter, raising additional concern over the effectiveness of contact tracing.
Although there spread of these two viruses appears to be contained, the increased efficiency of H5N1 infections in Indonesia raises more red flags. The HA of human H5N1 isolates in Indonesia are represented in a phylogenetic tree. The Karo cluster (shadowed in pink) is distinct with the larger cluster of human cases, primarily in western Java (green isolates on bottom branch of the tree). The vast majority of these isolates have a novel cleavage site, RESRRRKKR, which is not present in the bird isolates from Indonesia. Moreover, the human and cat isolates are readily distinguished from the avian isolates raising serious question about the origin of the human bird flu infection.
Immunological cross-reactivity between the isolates in western Java and Sumatra are limited, suggesting two vaccines just for the human isolates from Indonesia may be required.
H5N1 in humans in Indonesia is clearly not under control. The sequences from the human isolates remain sequestered, and the number of animal isolates remains low. Moreover the bird isolates have not matched the human isolates in western Java.
An increased effort and additional transparency in Indonesia is long overdue.
Media Sources
Phylogenetic Trees