H5N1 Evolution in Indonesian Alternative Reservoirs
Recombinomics Commentary
June 7, 2007
In Jakarta, Bayu Krisnamurthi, the head of Indonesia’s avian flu control commission, told reporters that it appears recent human cases have become infected from less intensive exposure to the virus than previously had been the case.
That raises the suspicion, he suggested, that the virus has adapted to more easily infect humans.
It was not clear which cases Krisnamurthi was referring to or how exposure was measured.
It’s not even clear how exposure could be measured in many of Indonesia’s cases. A substantial portion of the 99 human infections there have occurred in people for whom no link to infected poultry was ever discovered.
In a presentation last fall, officials of the Ministry of Health told the WHO that in at least one-third of their cases, they could find no firm proof of how the people had come in contact with the virus.
"Even the updates I’ve seen from the Ministry of Health, still there’s a large percentage of the cases they can’t find any connection" to infected birds, Perdue said.
In a related issue a microbiologist from Krisnamurthi’s commission told Reuters news agency that molecular study of the viruses has revealed changes.
"Virus samples from poultry cases have increasingly shown a similarity in their amino acid structure to virus samples extracted from humans," Wayan Teguh Wibawan said.
Perdue said it would be important to know if the Indonesian scientists are drawing the conclusions by studying the entire genetic blueprints of viruses isolated from human cases, or if they are looking only at a portion of the hemagglutinin gene.
The above comments raise additional questions about the evolution of H5N1 in Indonesia. There are many unresolved issues, which have been exacerbated by the withholding of samples and sequences from recent human and bird cases. There are also issues linked to the isolation of the H5N1 from human samples, as well as a lack of information of H5N1 in wild birds or mammals in Indonesia.
The problem on match failures between poultry and patient isolates has created concerns that were discussed in the summer of 2006 in Jakarta. The poultry sequences did not match the human sequences. These differences began to diminish when more recent poultry samples were sequenced, but did not really address the issue of an alternative reservoir of H5N1 in Indonesia. Last summer, the only match of the human sequences was with H5N1 from a cat infected in early 2006 in Indramayu.
Media reports have suggested that H5N1 infections of cats may be widespread, which could lead to independent evolution of H5N1 that is more compatible with mammalian infections in general. However, no additional feline sequences have been made public. Similarly, there have also been media reports of H5N1 infections in swine in Indonesia, but those sequences have also not been made public.
The one cat sequence did not have changes in the receptor binding domain and neither have human H5N1 sequences. However, human sequences have been generated by WHO affiliated reference labs in Hong Kong and Atlanta which grew H5N1 in chicken eggs instead of mammalian cells such as MDCK. These growth conditions could select against changes in the receptor binding domain.
Changes have been noted in isolates from Vietnam as well as the Middle East. However, in both locations, there was strong evidence for mixtures in the hosts that had H5N1 with receptor binding domain changes, or positions that interacted with the receptor binding domain. In Vietnam, some plaque purified clones had changes, while others, from the same patient, did not. Similarly, the receptor binding domain change S227N was present in Turkey in the first reported human case infected with the Qinghai (Clade 2.2) strain of H5N1. However, the change was not in the public sequence from the sister of the index case. Similarly, in Egypt S227N was also detected last year in one patient, but sequences from patients infected at about the same time in the same area were not released. Similarly, S227N was detected in Egypt this year, but the position encoding S227N was mixed, indicating the patient was infected with H5N1 with and without S227N.
In addition, many of the polymorphisms in Indonesia are found in H5N1 isolates from China and Qinghai isolates in countries in Europe, the Middle East, and Africa, suggesting transport by wild birds infected with Qinghai H5N1. Indeed, one Indonesian isolate did have the Qinghai H5N1 cleavage site, raising additional questions about H5N1 in various Indonesian reservoirs.
Thus, the above comments on H5N1 evolution in Indonesia highlight the need for more information of H5N1 sequences in patients, poultry, and alternative reservoirs such as cats, dogs, pigs, and wild birds.
Media sources
Recombinomics Presentations