- Split View
-
Views
-
Cite
Cite
Adrian Creanga, Nguyen Le Khanh Hang, Vuong Duc Cuong, Ha T Nguyen, Hoang Vu Mai Phuong, Le Thi Thanh, Nguyen Co Thach, Pham Thi Hien, Nguyen Tung, Yunho Jang, Amanda Balish, Nguyen Hoang Dang, Mai Thuy Duong, Ngo Thu Huong, Do Ngoc Hoa, Nguyen Dang Tho, Alexander Klimov, Bryan K Kapella, Larisa Gubareva, James C Kile, Nguyen Tran Hien, Le Quynh Mai, C Todd Davis, Highly Pathogenic Avian Influenza A(H5N1) Viruses at the Animal–Human Interface in Vietnam, 2003–2010, The Journal of Infectious Diseases, Volume 216, Issue suppl_4, 15 September 2017, Pages S529–S538, https://doi.org/10.1093/infdis/jix003
- Share Icon Share
Abstract
Mutation and reassortment of highly pathogenic avian influenza A(H5N1) viruses at the animal–human interface remain a major concern for emergence of viruses with pandemic potential. To understand the relationship of H5N1 viruses circulating in poultry and those isolated from humans, comprehensive phylogenetic and molecular analyses of viruses collected from both hosts in Vietnam between 2003 and 2010 were performed. We examined the temporal and spatial distribution of human cases relative to H5N1 poultry outbreaks and characterized the genetic lineages and amino acid substitutions in each gene segment identified in humans relative to closely related viruses from avian hosts. Six hemagglutinin clades and 8 genotypes were identified in humans, all of which were initially identified in poultry. Several amino acid mutations throughout the genomes of viruses isolated from humans were identified, indicating the potential for poultry viruses infecting humans to rapidly acquire molecular markers associated with mammalian adaptation and antiviral resistance.
The emergence of highly pathogenic avian influenza (HPAI) A(H5N1) virus during 2003 in several southeastern Asian countries, including China, Indonesia, Thailand, and Vietnam, was marked by the expansion of several different hemagglutinin (HA) gene clades. Vietnam has been one of the more severely affected countries. A total of 119 confirmed human H5N1 virus infections were recorded in Vietnam between 2003 and 2010, 59 (49.6%) of which were fatal [1]. Despite incomplete epidemiological data for all cases, the large majority occurred after reported exposure to sick or dead poultry. Currently, there is little evidence that any particular HPAI H5N1 virus is more virulent in or transmissible among or between birds and humans. And, while certain protein motifs and amino acid residues have been implicated in enhanced virulence and transmissibility, overall, these traits appear to be multifactorial [2]. In Vietnam, most poultry infections were caused by viruses that belonged to clade 1 (genotype Z; VN3) with sporadic infections of clades 0, 3, and 5 [3–5]. Clade 2.3.4 (genotype Z and V) viruses gradually replaced clade 1 viruses in northern and central regions of the country between 2005 and 2010, when several introductions of clade 2.3.4 subgroups were identified in poultry [6, 7]. Despite incursion of new genetic groups in the northern and central regions of the country, including clade 2.3.2.1 in 2009–2010, clade 1 viruses remained dominant in southern regions during this time period [8].
One Health studies to capture the genetic relationships of avian influenza virus collected specifically at the point of circulation in poultry and exposure to humans, the animal–human interface (AHI), have been limited due to a lack of available data from both host populations over mutual time periods. When available, however, analysis of such data can identify both shared and disparate genetic signatures among the 2 hosts and help to recognize mutations that may result in changes associated with mammalian adaptation or other risk factors such as resistance to antiviral drugs. To gain a better understanding of H5N1 evolution at the AHI, we performed systematic phylogenetic and molecular analyses of viruses collected from both humans and poultry in Vietnam between 2003 and 2010.
By first examining the temporal and spatial distribution of human cases relative to poultry outbreaks, and then linking this information to full genome sequence data using molecular clock phylogenetic methods, we identified closely related poultry viruses that likely resulted in human infections in Vietnam. Amino acid changes were then identified between these viruses and characterized in the context of their emergence following human infection. The genetic diversity of both poultry and human viruses collected from 2003 to 2010 was also analyzed to understand how the rapid expansion of some groups of viruses may have influenced the prevalence of outbreaks in poultry and associated human cases of H5N1.
METHODS
Virus Isolation and Sequencing
H5N1 viruses were isolated from positive human clinical specimens and poultry outbreak samples by inoculation into the allantoic sac of 9- to 11-day-old embryonated chicken eggs. RNA was extracted from virus-infected allantoic fluid 24 hours postinoculation using RNeasy Kit, following the manufacturer’s protocol (Qiagen). Access Quick One-Step reverse-transcription polymerase chain reaction kit (Promega) and M13-tagged influenza specific primers were used to amplify each of the individual influenza segments. The amplicons were sequenced using dye-terminator chemistry and the Applied Biosystems 3730 system. Contigs were assembled using Sequencer 5.0 (Gene Codes). Specific viral isolation and sequencing protocols were described elsewhere [6, 8].
Bayesian Phylogenetic Analysis
The datasets comprised all animal and human H5N1 viral sequences collected in Vietnam between 2001 and 2010. To perform dated phylogenies, we used a Bayesian Markov chain Monte Carlo (MCMC) method implemented in Beast 1.7.4. Each gene was analyzed using the codon-based SRD06 nucleotide substitution model and uncorrelated log-normal relaxed clock [9]. In cases of viral isolates for which precise collection time (month or day) information was missing, missing collection time (month or day) values were assigned midyear (July 1) or midmonth (day 15) values. The Bayesian skyline coalescent model was used to characterize the viral population dynamics. For each viral gene segment, we performed at least 2 Bayesian MCMC runs with 150–250 million generations sampled to produce 10 000 trees. Convergence of 2 runs was confirmed in Tracer 1.5; effective sample size counts >300 were considered sufficient for downstream analyses. Maximum clade credibility (MCC) trees were generated using TreeAnnotator after removing 10% burnin and were analyzed with FigTree. The temporal and spatial association between clades of avian and human viruses was also correlated and mapped using ArcGIS 10 software (Esri) (Figure 1). The dated phylogeny of the HA gene was used to identify the clade of each virus and the MCC tree was plotted relative to the monthly distribution of human cases and avian outbreaks and the relative HA genetic diversity (Figure 2, Supplementary Tables 1–3).
Analysis of Amino Acid Differences at the Animal–Human Interface
Each codon complete amino acid sequence alignment was created by arranging sequences in the same order as in the MCC trees. These datasets were used to identify amino acid differences between human and closely related avian sequences of H5N1 viruses (ie, avian sequences belonging to the same genetic lineage and collected shortly before and after the human sequence). For each identified mutation, we recorded the residue position, the specific amino acid found in avian vs human sequences, and the number of avian vs human sequences in the respective group of the tree used for this analysis (Figure 3, Supplementary Table 4).
Neuraminidase Inhibition Assays
The drug concentration required to inhibit 50% of the neuraminidase (NA) activity (IC50) was assessed by the fluorescent assay, as previously described [10].The results were interpreted as outlined elsewhere [11]. Neuraminidase sequence alignments were created as described above to compare the presence of specific residues identified in closely related avian vs human viruses (Supplementary Table 5).
RESULTS
Epidemiology
Overall, 4489 H5N1 poultry outbreaks and 119 human cases were reported in Vietnam between 2003 and 2010 (Table 1, Figure 2A) [1, 12]. The first confirmed H5N1 infections in humans were recorded in 3 provinces in northern Vietnam (Nam Dinh, Bac Ninh, and Ha Nam) in December 2003. Forty-eight poultry outbreaks were reported in the same 3 provinces in January 2004. From 2004 to 2010, 112 of 119 human cases (94%; Table 1, Figure 1, and Supplementary Table 3) occurred in the same provinces where poultry outbreaks were also reported. Six of 7 human cases that occurred in provinces without reported outbreaks were in provinces adjacent to those with outbreaks. A single human case was observed in relative isolation of known animal outbreaks in Binh Duong province in southern Vietnam in March 2010 (Table 1, Figure 1, and Supplementary Table 3). Analysis of the temporal distribution of poultry outbreaks in Vietnam indicated that the annual number decreased from 2359 in 2004 to 48 in 2010, with the number of reported outbreaks stabilizing from 2008 to 2010 (Table 1, Figure 1, and Supplementary Table 2). The number of human cases varied considerably during the study period; the most cases (n = 61) occurred in 2005 but declined to an average of only 6 cases per year from 2008 to 2010, which appeared to coincide with the number of poultry outbreaks reported (Table 1, Supplementary Table 3). Interestingly, the annual number of poultry outbreaks and human cases were highly correlated (correlation coefficient = 0.76).
Year . | Total Human Cases . | Provinces with Human Cases . | Provinces with Both Human Cases and Avian Outbreaksa . | Provinces with Human Cases and Avian Outbreaks in the Adjacent Provinces . | Provinces with Isolated Human Cases (ie, w/o Avian Outbreaks in the Same or Adjacent Provinces) . | Total Avian Outbreaks . | Provinces with Avian Outbreaks . | Provinces with Isolated Avian Outbreaks (ie, w/o Human Cases in the Same or Adjacent Provinces) . |
---|---|---|---|---|---|---|---|---|
2003 | 3 | 3 | na | na | na | na | na | na |
2004 | 29 | 15 | 14 | 1 | 0 | 2359 | 58 | 13 |
2005 | 61 | 26 | 26 | 0 | 0 | 1628 | 48 | 7 |
2006 | 0 | 0 | 0 | 0 | 0 | 63 | 6 | 6 |
2007 | 8 | 6 | 4 | 2 | 0 | 252 | 29 | 16 |
2008 | 6 | 6 | 5 | 1 | 0 | 80 | 28 | 17 |
2009 | 5 | 5 | 5 | 1 | 0 | 59 | 18 | 8 |
2010 | 7 | 6 | 4 | 1 | 1b | 48 | 21 | 12 |
Year . | Total Human Cases . | Provinces with Human Cases . | Provinces with Both Human Cases and Avian Outbreaksa . | Provinces with Human Cases and Avian Outbreaks in the Adjacent Provinces . | Provinces with Isolated Human Cases (ie, w/o Avian Outbreaks in the Same or Adjacent Provinces) . | Total Avian Outbreaks . | Provinces with Avian Outbreaks . | Provinces with Isolated Avian Outbreaks (ie, w/o Human Cases in the Same or Adjacent Provinces) . |
---|---|---|---|---|---|---|---|---|
2003 | 3 | 3 | na | na | na | na | na | na |
2004 | 29 | 15 | 14 | 1 | 0 | 2359 | 58 | 13 |
2005 | 61 | 26 | 26 | 0 | 0 | 1628 | 48 | 7 |
2006 | 0 | 0 | 0 | 0 | 0 | 63 | 6 | 6 |
2007 | 8 | 6 | 4 | 2 | 0 | 252 | 29 | 16 |
2008 | 6 | 6 | 5 | 1 | 0 | 80 | 28 | 17 |
2009 | 5 | 5 | 5 | 1 | 0 | 59 | 18 | 8 |
2010 | 7 | 6 | 4 | 1 | 1b | 48 | 21 | 12 |
Abbreviation: na, not applicable.
aOutbreak information was based on http://empress-i.fao.org. For a comprehensive analysis of H5 poultry outbreak distribution and human cases, see Supplementary Tables 2 and 3 and Figure 1.
bA single human case of influenza A(H5) virus was recorded in isolation of an avian outbreak in Bing Duong province in southeastern Vietnam in 2010.
Year . | Total Human Cases . | Provinces with Human Cases . | Provinces with Both Human Cases and Avian Outbreaksa . | Provinces with Human Cases and Avian Outbreaks in the Adjacent Provinces . | Provinces with Isolated Human Cases (ie, w/o Avian Outbreaks in the Same or Adjacent Provinces) . | Total Avian Outbreaks . | Provinces with Avian Outbreaks . | Provinces with Isolated Avian Outbreaks (ie, w/o Human Cases in the Same or Adjacent Provinces) . |
---|---|---|---|---|---|---|---|---|
2003 | 3 | 3 | na | na | na | na | na | na |
2004 | 29 | 15 | 14 | 1 | 0 | 2359 | 58 | 13 |
2005 | 61 | 26 | 26 | 0 | 0 | 1628 | 48 | 7 |
2006 | 0 | 0 | 0 | 0 | 0 | 63 | 6 | 6 |
2007 | 8 | 6 | 4 | 2 | 0 | 252 | 29 | 16 |
2008 | 6 | 6 | 5 | 1 | 0 | 80 | 28 | 17 |
2009 | 5 | 5 | 5 | 1 | 0 | 59 | 18 | 8 |
2010 | 7 | 6 | 4 | 1 | 1b | 48 | 21 | 12 |
Year . | Total Human Cases . | Provinces with Human Cases . | Provinces with Both Human Cases and Avian Outbreaksa . | Provinces with Human Cases and Avian Outbreaks in the Adjacent Provinces . | Provinces with Isolated Human Cases (ie, w/o Avian Outbreaks in the Same or Adjacent Provinces) . | Total Avian Outbreaks . | Provinces with Avian Outbreaks . | Provinces with Isolated Avian Outbreaks (ie, w/o Human Cases in the Same or Adjacent Provinces) . |
---|---|---|---|---|---|---|---|---|
2003 | 3 | 3 | na | na | na | na | na | na |
2004 | 29 | 15 | 14 | 1 | 0 | 2359 | 58 | 13 |
2005 | 61 | 26 | 26 | 0 | 0 | 1628 | 48 | 7 |
2006 | 0 | 0 | 0 | 0 | 0 | 63 | 6 | 6 |
2007 | 8 | 6 | 4 | 2 | 0 | 252 | 29 | 16 |
2008 | 6 | 6 | 5 | 1 | 0 | 80 | 28 | 17 |
2009 | 5 | 5 | 5 | 1 | 0 | 59 | 18 | 8 |
2010 | 7 | 6 | 4 | 1 | 1b | 48 | 21 | 12 |
Abbreviation: na, not applicable.
aOutbreak information was based on http://empress-i.fao.org. For a comprehensive analysis of H5 poultry outbreak distribution and human cases, see Supplementary Tables 2 and 3 and Figure 1.
bA single human case of influenza A(H5) virus was recorded in isolation of an avian outbreak in Bing Duong province in southeastern Vietnam in 2010.
H5N1 HA Gene Diversity
Phylogenetic analysis of the HA gene of H5N1 viruses circulating in poultry that also resulted in human infections identified 6 clades (clades 1, 1.1, 2.3.4, 2.3.4.1, 2.3.4.2, and 2.3.4.3) that were detected in Vietnam between 2003 and 2010 (Figure 2B, Supplementary Figure 1D). In 2003 and 2004, clade 1 viruses were collected from both infected humans and poultry. Clade 1 viruses were identified initially in northern provinces in December 2003 and were detected throughout Vietnam in 2004. In 2005, clade 1 viruses continued to cause human and animal infections in southern provinces of Vietnam, primarily around the Mekong River delta, while clade 2.3.4 viruses emerged and were identified in both humans and poultry in northern and central regions of the country (Figure 1, Table 2, and Supplementary Figure 1D). Despite a reduction of outbreaks in Vietnam during most of 2006 (ie, only 3 outbreaks were recorded from January to November), the virus reemerged in poultry in December 2006 and was again detected in humans in May 2007. Clade 2.3.4 viruses from several subgroups continued to circulate in northern and central provinces of Vietnam between 2007 and 2010 and were detected in poultry causing sporadic human infections. Specifically, clade 2.3.4.3 viruses were found in poultry between 2007 and 2010 and caused 12 human infections between 2007 and 2009; clade 2.3.4.2 viruses were isolated from both poultry and humans (n = 3) between 2008 and 2010; clade 2.3.4.1 viruses emerged in poultry and caused human infections (n = 3) in 2009 and continued to circulate throughout 2010. In southern regions of the country, clade 1.1 viruses continued to evolve and caused numerous poultry outbreaks between 2007 and 2010, but only a single clade 1.1 virus was isolated from a human in Dong Thap province in southern Vietnam in 2009. Notably, there were several clades of H5N1 viruses that were not detected in humans between 2003 and 2010; clade 3 viruses were detected sporadically in poultry in Vietnam before 2005; clade 5 and clade 2.3.2 viruses were found in northern regions of Vietnam in 2003 and 2005, respectively; clade 7 viruses were sporadically detected in imported poultry; and clade 2.3.2.1 viruses emerged and spread to northern and central regions of the country during 2009 (Figure 1, Table 2, and Supplementary Figure 1D). We found that the H5 HA gene diversity did not change considerably between 2003 and 2007, increased sharply in mid-2007, and declined in 2010 (Figure 2C). This trend was paralleled by a decline in the number of human cases and animal outbreaks from 2003 to 2010 (Figure 2A and 2C). Similarly, analysis of internal genes of H5N1 viruses from Vietnam identified a significant increase in the genetic diversity in 2007 and a relative stabilization by 2009–2010 at levels higher than the 2003 to early 2007 time period (Supplementary Figure 2A–G).
Year . | HUMAN HPAI H5N1 viruses . | HA clade of HUMAN HPAI H5 virusesa . | HA clade of AVIAN HPAI H5 viruses circulating in the same or adjacent provinces to HUMAN cases . | Genotypebof HUMAN HPAI H5N1 viruses . | Genotype of AVIAN HPAI H5N1 viruses circulating in the same or adjacent provinces to HUMAN cases . |
---|---|---|---|---|---|
2003 | 3 | 1 | 1, 5 | VN3 | VN3, VN2 |
2004 | 27 | 1 | 1 | VN3 | VN3 |
2005 | 17 | 1, 2.3.4 | 1, 2.3.4, 2.3.2 | VN3, VN6 | VN3, VN6, VN4 |
2006 | NONE | na | na | na | na |
2007 | 8 | 2.3.4.3 | 1.1, 2.3.4.2, 2.3.4, (2.3.4.3) | VN7 | VN3, VN6, VN30, (VN7, VN29) |
2008 | 5 | 2.3.4.2, 2.3.4.3 | 1.1, 2.3.4.2, 2.3.4.3, 7.1c | VN21, VN7 | VN3, VN19, VN21, VN7, nac |
2009 | 4 | 1.1, 2.3.4.1, 2.3.4.2, 2.3.4.3 | 1.1, 2.3.4.1, 2.3.4.2, 2.3.4.3, 2.3.2.1 | VN3, VN10e, VN23d, VN7 | VN3, VN10e, VN11, VN7, VN12 |
2010 | 3 | 2.3.4.1, 2.3.4.2 | 1.1, (2.3.4.1), 2.3.4.2, (2.3.2.1) | VN10, VN11 | VN3, (VN14, VN15, VN10), VN11, (VN12, VN35, VN36) |
Year . | HUMAN HPAI H5N1 viruses . | HA clade of HUMAN HPAI H5 virusesa . | HA clade of AVIAN HPAI H5 viruses circulating in the same or adjacent provinces to HUMAN cases . | Genotypebof HUMAN HPAI H5N1 viruses . | Genotype of AVIAN HPAI H5N1 viruses circulating in the same or adjacent provinces to HUMAN cases . |
---|---|---|---|---|---|
2003 | 3 | 1 | 1, 5 | VN3 | VN3, VN2 |
2004 | 27 | 1 | 1 | VN3 | VN3 |
2005 | 17 | 1, 2.3.4 | 1, 2.3.4, 2.3.2 | VN3, VN6 | VN3, VN6, VN4 |
2006 | NONE | na | na | na | na |
2007 | 8 | 2.3.4.3 | 1.1, 2.3.4.2, 2.3.4, (2.3.4.3) | VN7 | VN3, VN6, VN30, (VN7, VN29) |
2008 | 5 | 2.3.4.2, 2.3.4.3 | 1.1, 2.3.4.2, 2.3.4.3, 7.1c | VN21, VN7 | VN3, VN19, VN21, VN7, nac |
2009 | 4 | 1.1, 2.3.4.1, 2.3.4.2, 2.3.4.3 | 1.1, 2.3.4.1, 2.3.4.2, 2.3.4.3, 2.3.2.1 | VN3, VN10e, VN23d, VN7 | VN3, VN10e, VN11, VN7, VN12 |
2010 | 3 | 2.3.4.1, 2.3.4.2 | 1.1, (2.3.4.1), 2.3.4.2, (2.3.2.1) | VN10, VN11 | VN3, (VN14, VN15, VN10), VN11, (VN12, VN35, VN36) |
aHA clades and genotypes of H5N1 viruses are colored to highlight the differences in their circulation: clades and genotypes circulating in both poultry and humans during a specific year are shown in bold; viruses circulating in humans but not detected in the same or adjacent provinces in poultry during that year are shown initalics; viruses circulating only in poultry are shown in regular characters.
bGenotypes are listed in the same order as HA clades. When multiple genotypes share the same HA clade due to reassortment, they are shown between parentheses. VN3 represents the typical genotype of clade 1/1.1 viruses; VN6 is the genotype of clade 2.3.4 viruses; VN7 is the genotype of clade 2.3.4.3; VN10 is the genotype of clade 2.3.4.1; VN11 is the genotype of clade 2.3.4.2 viruses; VN12 is the genotype of clade 2.3.2.1 viruses; VN7 and VN29 are both clade 2.3.4.3 viruses with distinct genotypes due to reassortment (NA of VN7 belongs to lineage A and NA of VN29 viruses belongs to lineage F); VN10, VN14, VN15 are clade 2.3.4.1 viruses with distinct genotypes due to reassortment (NA of VN10 belongs to lineage G, NA of VN11 belongs to lineage B and NA of VN15 belongs to lineage E); VN12, VN35 and VN36 are clade 2.3.2.1 viruses with distinct genotypes due to reassortment.
cGenotype of clade 7.1 viruses circulating in Hai Duong Province in 2008 was not characterized
dVN23 (clade 2.3.4.2) viruses were collected from human and poultry in Vietnam in 2009 but the collection location for the poultry viruses was not recorded.
eVN10 (clade 2.3.4.1) viruses circulated in both poultry in 2009, but in different, non-neighboring provinces (avian VN10 viruses were collected from Hanoi and Dien Bien, but human VN10 virus was collected from Quang Ninh).
Year . | HUMAN HPAI H5N1 viruses . | HA clade of HUMAN HPAI H5 virusesa . | HA clade of AVIAN HPAI H5 viruses circulating in the same or adjacent provinces to HUMAN cases . | Genotypebof HUMAN HPAI H5N1 viruses . | Genotype of AVIAN HPAI H5N1 viruses circulating in the same or adjacent provinces to HUMAN cases . |
---|---|---|---|---|---|
2003 | 3 | 1 | 1, 5 | VN3 | VN3, VN2 |
2004 | 27 | 1 | 1 | VN3 | VN3 |
2005 | 17 | 1, 2.3.4 | 1, 2.3.4, 2.3.2 | VN3, VN6 | VN3, VN6, VN4 |
2006 | NONE | na | na | na | na |
2007 | 8 | 2.3.4.3 | 1.1, 2.3.4.2, 2.3.4, (2.3.4.3) | VN7 | VN3, VN6, VN30, (VN7, VN29) |
2008 | 5 | 2.3.4.2, 2.3.4.3 | 1.1, 2.3.4.2, 2.3.4.3, 7.1c | VN21, VN7 | VN3, VN19, VN21, VN7, nac |
2009 | 4 | 1.1, 2.3.4.1, 2.3.4.2, 2.3.4.3 | 1.1, 2.3.4.1, 2.3.4.2, 2.3.4.3, 2.3.2.1 | VN3, VN10e, VN23d, VN7 | VN3, VN10e, VN11, VN7, VN12 |
2010 | 3 | 2.3.4.1, 2.3.4.2 | 1.1, (2.3.4.1), 2.3.4.2, (2.3.2.1) | VN10, VN11 | VN3, (VN14, VN15, VN10), VN11, (VN12, VN35, VN36) |
Year . | HUMAN HPAI H5N1 viruses . | HA clade of HUMAN HPAI H5 virusesa . | HA clade of AVIAN HPAI H5 viruses circulating in the same or adjacent provinces to HUMAN cases . | Genotypebof HUMAN HPAI H5N1 viruses . | Genotype of AVIAN HPAI H5N1 viruses circulating in the same or adjacent provinces to HUMAN cases . |
---|---|---|---|---|---|
2003 | 3 | 1 | 1, 5 | VN3 | VN3, VN2 |
2004 | 27 | 1 | 1 | VN3 | VN3 |
2005 | 17 | 1, 2.3.4 | 1, 2.3.4, 2.3.2 | VN3, VN6 | VN3, VN6, VN4 |
2006 | NONE | na | na | na | na |
2007 | 8 | 2.3.4.3 | 1.1, 2.3.4.2, 2.3.4, (2.3.4.3) | VN7 | VN3, VN6, VN30, (VN7, VN29) |
2008 | 5 | 2.3.4.2, 2.3.4.3 | 1.1, 2.3.4.2, 2.3.4.3, 7.1c | VN21, VN7 | VN3, VN19, VN21, VN7, nac |
2009 | 4 | 1.1, 2.3.4.1, 2.3.4.2, 2.3.4.3 | 1.1, 2.3.4.1, 2.3.4.2, 2.3.4.3, 2.3.2.1 | VN3, VN10e, VN23d, VN7 | VN3, VN10e, VN11, VN7, VN12 |
2010 | 3 | 2.3.4.1, 2.3.4.2 | 1.1, (2.3.4.1), 2.3.4.2, (2.3.2.1) | VN10, VN11 | VN3, (VN14, VN15, VN10), VN11, (VN12, VN35, VN36) |
aHA clades and genotypes of H5N1 viruses are colored to highlight the differences in their circulation: clades and genotypes circulating in both poultry and humans during a specific year are shown in bold; viruses circulating in humans but not detected in the same or adjacent provinces in poultry during that year are shown initalics; viruses circulating only in poultry are shown in regular characters.
bGenotypes are listed in the same order as HA clades. When multiple genotypes share the same HA clade due to reassortment, they are shown between parentheses. VN3 represents the typical genotype of clade 1/1.1 viruses; VN6 is the genotype of clade 2.3.4 viruses; VN7 is the genotype of clade 2.3.4.3; VN10 is the genotype of clade 2.3.4.1; VN11 is the genotype of clade 2.3.4.2 viruses; VN12 is the genotype of clade 2.3.2.1 viruses; VN7 and VN29 are both clade 2.3.4.3 viruses with distinct genotypes due to reassortment (NA of VN7 belongs to lineage A and NA of VN29 viruses belongs to lineage F); VN10, VN14, VN15 are clade 2.3.4.1 viruses with distinct genotypes due to reassortment (NA of VN10 belongs to lineage G, NA of VN11 belongs to lineage B and NA of VN15 belongs to lineage E); VN12, VN35 and VN36 are clade 2.3.2.1 viruses with distinct genotypes due to reassortment.
cGenotype of clade 7.1 viruses circulating in Hai Duong Province in 2008 was not characterized
dVN23 (clade 2.3.4.2) viruses were collected from human and poultry in Vietnam in 2009 but the collection location for the poultry viruses was not recorded.
eVN10 (clade 2.3.4.1) viruses circulated in both poultry in 2009, but in different, non-neighboring provinces (avian VN10 viruses were collected from Hanoi and Dien Bien, but human VN10 virus was collected from Quang Ninh).
H5N1 Genotypes Identified in Poultry and Human Viruses
Eight H5N1 genotypes were identified in humans in Vietnam between 2003 and 2010. Human viruses isolated in 2003 and 2004 had a VN3 genotype (clade 1 viruses), also the most prevalent genotype of animal H5N1 viruses isolated during this period (Table 2 and Supplementary Table 1). While VN6 viruses (clade 2.3.4) emerged in northern regions of Vietnam in both birds and humans in 2005, the VN3 genotype remained the most prevalent genotype in Vietnam. Between 2003 and 2006, both VN2 and VN4 H5N1 viruses were circulating in poultry, but were not isolated in humans. Five new H5N1 genotypes (VN7, VN10, VN11, VN21, and VN23) were isolated from humans in the northern provinces of Vietnam between 2007 and 2010. All these viruses also circulated in poultry, and based on the spatial and temporal distribution of these viruses, a correlation between the emergence of H5N1 viruses in animals and their transmission to humans was identified. For example, VN7 and VN11 viruses were found in both poultry and human samples collected in the same provinces during the same time period; VN10, VN16, VN21, and VN23 viruses were found in poultry during the same years when they were identified in humans (Table 2, Supplementary Table 1). During 2007–2010, the VN3 genotype was the most prevalent in poultry in the southern regions of Vietnam. Despite its continuous circulation and evolution in poultry, a single human VN3 virus was isolated from a patient in Dong Thap province in southern Vietnam in 2009. It should also be noted that clade 2.3.2.1 VN12 viruses circulated exclusively in poultry in 13 provinces in central and northern Vietnam in 2009 and 2010; this H5N1 virus was not detected in humans during this time interval (Table 2, Supplementary Table 1).
Mutations Between Poultry and Human H5N1 Viruses
An analysis of the HA amino acid changes at the AHI was performed to identify putative molecular determinants of human infections. Specifically, we examined amino acid changes between human and related animal sequences, and noted a mutation in humans if the human amino acid was different than the one found in the related animal sequences identified at the time of the human infection and belonging to the same monophyletic group in the phylogenetic tree. We next examined frequencies of these amino acid residues in humans relative to avian sequences (Figure 3, Supplementary Table 4). We found 34 HA amino acids that may have occurred after avian viruses infected humans. However, most of these mutations (n = 24) were detected in individual human viruses and may only represent sporadic changes. Interestingly, 18 of the 24 human amino acid residues were not present in any avian H5 HA protein sequences, and 4 other human amino acid residues were identified in only a single avian sequence (Figure 3). Of note, 3 amino acids occurred in >1 human virus, yet 2 of these 3 amino acids (residues 188 and 326) were found in a large number of avian H5 HA protein sequences as well. All human and avian isolates had Gln and Gly at position 222 and 224, respectively, at the receptor-binding pocket of HA1, indicating affinity for avian cell-surface receptors.
The analysis of amino acid differences between the internal proteins of human and animal H5N1 influenza viruses in Vietnam confirmed that the Glu to Lys mutation at residue 627 of the PB2 protein has a critical role in the ability of avian H5N1 viruses to replicate in humans [13, 14]. The Lys residue was found in 20 of the 37 human PB2 protein sequences isolated from humans in Vietnam. Asn at position 701 of PB2 was detected in 2 of the 17 viruses that lacked the 627 Lys residue, supporting its role in mammalian adaptation as well. Of note, the Lys at residue 627 was found in only 1 of 195 avian PB2 proteins and Asn at residue 701 was not found among the 195 avian PB2 proteins, suggesting that these mutations were most likely frequently selected following human infection. Combined, we found that 167 internal gene amino acid residues were detected in human viruses compared to avian counterparts, representing potential adaptive mutations. However, 121 of these substitutions were found in only a single human virus (Figure 3, Supplementary Table 4). Only 3 amino acid substitutions occurred in H5N1 influenza protein sequences belonging to distinct lineages. These were mutations at amino acid residue 86 found in 3 PA proteins belonging to lineages A (n = 2) and D (n = 1); amino acid residue 11 found in 2 PA-X proteins also of A and D lineages; and amino acid residue 429 found in 5 NA proteins belonging to lineages D (n = 4) and A (n = 1). We identified 9 other amino acid residues of potential public health interest based on the increased number of human sequences (between 2 and 4) with these residues and their lower prevalence in avian protein sequences (Figure 3, Supplementary Table 4).
Susceptibility to Antiviral Drugs
M2 sequences (n = 604) were inspected for established markers of resistance to adamantanes. Clade 1, 1.1, and 7 viruses from both poultry and humans were resistant to this drug class due to changes Ser31Asn and/or Val27Ala; however, the frequency of these mutations was low (0–5%) among other clades (data not shown). Viruses that caused human infections in Vietnam were resistant to adamantanes, if they belonged to clade 1.1 (n = 29); however, those not belonging to this clade were susceptible (n = 26).
NA sequences (n = 655) were analyzed to find the only established oseltamivir-resistance marker, His275Tyr; it was found only in the sequences of 2 previously reported viruses, a clade 1 virus recovered from an oseltamivir-treated patient [15] and a clade 2.3.2.1 virus collected from a bird [10]. Notably, 34 of the sequences analyzed in this study (5%) contained substitutions previously associated with potentially reduced susceptibility to NA inhibitor(s) (Supplementary Table 5). Of those, one had Val116Ala [16, 17], 5 contained Val149Ala or Ile [18], and 5 had Ser247Asn [16]. Among the most frequent substitutions were those at Ile223 [19] (n = 14/36); 3 viruses harbored substitutions at both Ile223 and Ser247. Ile117Lys was present in a single clade 1 virus while Ile117Val was present in 14 clade 2.3.4.3 viruses from 2007 to 2008. The latter substitution was present in viruses recovered from birds, a civet, and 2 humans. Notably, the 2 human Ile117Val viruses were collected from residents of 2 different provinces who were not treated with oseltamivir. One of the 2 cases was fatal [20]. Ile117Val was associated with an approximately 3-fold increase in oseltamivir IC50 (vs the median IC50 = 7.4 nM). However, the result of the NA inhibition assay can be interpreted as reduced inhibition (≥10-fold), if the IC50 value (2.4 nM) of the drug-susceptible A(H1N1) virus was used as the reference (Table 3).
Clade . | Virus . | NA Change . | IC50, nM (Fold Increase)a . | |||
---|---|---|---|---|---|---|
Oseltamivir . | Zanamivir . | Peramivir . | A-315675 . | |||
1 | A/duck/Vietnam/Ncvdcdc17/2005 | K432T | 1.9 (9) | 8.12 (12) | 2.21 (5) | 3.17 (7) |
Median (n = 41) | 0.21 | 0.67 | 0.45 | 0.48 | ||
2.3.2.1 | A/duck/Vietnam/NCVD-664/2010 | H275Y | 749.55 (1027) | 0.8 (1) | 130.01 (310) | Not tested |
Median (n = 18) | 0.73 | 0.84 | 0.42 | 0.37 | ||
2.3.4.2 | A/chicken/Vietnam/NCVD-44/2007 | I223T | 20.52 (3) | 0.79 (1) | 0.78 (1) | 1.21 (2) |
A/chicken/Vietnam/NCVD-45/2007 | I223T | 18.95 (3) | 0.7 (1) | 0.75 (1) | 1.07 (1) | |
A/chicken/Vietnam/NCVD-080/2008 | I223T | 16.04 (2) | 1.01 (1) | 0.93 (1) | 1.8 (3) | |
A/chicken/Vietnam/NCVD-283/2009 | I222T | 36.85 (5) | 1.76 (2) | 1.11 (2) | 1.59 (2) | |
A/chicken/Vietnam/NCVD-295/2009 | I223T | 45.78 (6) | 1.58 (1) | 1.09 (2) | 1.22 (2) | |
A/chicken/Vietnam/NCVD-296/2009 | I223T | 44.36 (6) | 1.73 (2) | 1.12 (2) | 1.51 (2) | |
A/chicken/Vietnam/NCVD-103/2007 | I223T and S247N | 50.28 (7) | 2.69 (3) | 4.11 (6) | 12.77 (18) | |
2.3.4 | A/duck/Vietnam/NCVD-31/2007 | I223L and S247N | 102.85 (14) | 1.25 (1) | 3.84 (5) | 110.47 (153) |
2.3.4.3 | A/Vietnam/HN31412/2008 | I117V | 24.44 (3) | 2.00 (2) | 0.9 (1) | 4.64 (6) |
A/duck/Vietnam/NCVD-94/2007 | I117V | 24.82 (3) | 3.19 (3) | 0.92 (1) | 6.17 (9) | |
A/duck/Vietnam/NCVD-100/2007 | I117V | 25.33 (3) | 3.11 (3) | 1.13 (2) | 6.52 (9) | |
A/chicken/Vietnam/NCVD-106/2007 | I117V | 22.79 (3) | 1.64 (2) | 0.75 (1) | 3.67 (5) | |
A/muscovy duck/Vietnam/NCVD-120/2007 | I117V | 28.75 (4) | 2.99 (3) | 1.4 (2) | 6.34 (9) | |
A/duck/Vietnam/NCVD-118/2007 | I117V | 24.14 (3) | 1.6 (2) | 0.8 (1) | 2.92 (4) | |
A/chicken/Vietnam/NCVD-138/2008 | I117V | 25.75 (3) | 1.72 (2) | 0.4 (1) | 6.88 (4) | |
A/Owston’s civet/Vietnam/NCVD-004/2008 | I117V | 22.86 (3) | 1.5 (1) | 0.84 (1) | 4.26 (6) | |
A/duck/Vietnam/NCVD107/2007 | I117V | 19.4 (3) | 1.4 (1) | 0.9 (1) | 3.4 (5) | |
A/duck/Vietnam/NCVD-56/2007 | A395T | 41.26 (6) | 3.47 (3) | 2.86 (4) | 5.46 (8) | |
A/chicken/Vietnam/NCVD-086/2008 | D402N | 30.37 (4) | 5.42 (5) | 3.11 (4) | 7.57 (11) | |
A/duck/Vietnam/NCVD-159/2008 | V424I | 40.2 (5) | 3.1 (3) | 1 (1) | 6.1 (8) | |
A/duck/Vietnam/NCVD-160/2008 | V424I | 48.71 (7) | 3.28 (3) | 1.36 (2) | 6.02 (8) | |
Median (n = 164) | 7.40 | 1.06 | 0.72 | 1.83 | ||
Reference A(H5N1) viruses | ||||||
1 | A/Vietnam/HN30408/2005 | N295S | 2.8 (13) | 2.1 (3) | 0.7 (2) | 0.9 (2) |
A/Vietnam/HN30408/2005 | H275Y | 863.1 (4110) | 1.6 (2) | 27.1 (60) | 3.1 (6) | |
A/Vietnam/1203/2004 | None | 0.3 (1) | 0.8 (1) | 0.6 (1) | 0.5 (1) | |
Reference A(H1N1) viruses | ||||||
Not applicable | A/Texas/36/1991, oseltamivir-resistant | H275Y | 1598 (666) | 3.7 (1) | 83.7 (60) | 12.8 (5) |
A/Texas/36/1991, oseltamivir-sensitive | None | 2.4 (1) | 3.2 | 1.4 | 2.4 |
Clade . | Virus . | NA Change . | IC50, nM (Fold Increase)a . | |||
---|---|---|---|---|---|---|
Oseltamivir . | Zanamivir . | Peramivir . | A-315675 . | |||
1 | A/duck/Vietnam/Ncvdcdc17/2005 | K432T | 1.9 (9) | 8.12 (12) | 2.21 (5) | 3.17 (7) |
Median (n = 41) | 0.21 | 0.67 | 0.45 | 0.48 | ||
2.3.2.1 | A/duck/Vietnam/NCVD-664/2010 | H275Y | 749.55 (1027) | 0.8 (1) | 130.01 (310) | Not tested |
Median (n = 18) | 0.73 | 0.84 | 0.42 | 0.37 | ||
2.3.4.2 | A/chicken/Vietnam/NCVD-44/2007 | I223T | 20.52 (3) | 0.79 (1) | 0.78 (1) | 1.21 (2) |
A/chicken/Vietnam/NCVD-45/2007 | I223T | 18.95 (3) | 0.7 (1) | 0.75 (1) | 1.07 (1) | |
A/chicken/Vietnam/NCVD-080/2008 | I223T | 16.04 (2) | 1.01 (1) | 0.93 (1) | 1.8 (3) | |
A/chicken/Vietnam/NCVD-283/2009 | I222T | 36.85 (5) | 1.76 (2) | 1.11 (2) | 1.59 (2) | |
A/chicken/Vietnam/NCVD-295/2009 | I223T | 45.78 (6) | 1.58 (1) | 1.09 (2) | 1.22 (2) | |
A/chicken/Vietnam/NCVD-296/2009 | I223T | 44.36 (6) | 1.73 (2) | 1.12 (2) | 1.51 (2) | |
A/chicken/Vietnam/NCVD-103/2007 | I223T and S247N | 50.28 (7) | 2.69 (3) | 4.11 (6) | 12.77 (18) | |
2.3.4 | A/duck/Vietnam/NCVD-31/2007 | I223L and S247N | 102.85 (14) | 1.25 (1) | 3.84 (5) | 110.47 (153) |
2.3.4.3 | A/Vietnam/HN31412/2008 | I117V | 24.44 (3) | 2.00 (2) | 0.9 (1) | 4.64 (6) |
A/duck/Vietnam/NCVD-94/2007 | I117V | 24.82 (3) | 3.19 (3) | 0.92 (1) | 6.17 (9) | |
A/duck/Vietnam/NCVD-100/2007 | I117V | 25.33 (3) | 3.11 (3) | 1.13 (2) | 6.52 (9) | |
A/chicken/Vietnam/NCVD-106/2007 | I117V | 22.79 (3) | 1.64 (2) | 0.75 (1) | 3.67 (5) | |
A/muscovy duck/Vietnam/NCVD-120/2007 | I117V | 28.75 (4) | 2.99 (3) | 1.4 (2) | 6.34 (9) | |
A/duck/Vietnam/NCVD-118/2007 | I117V | 24.14 (3) | 1.6 (2) | 0.8 (1) | 2.92 (4) | |
A/chicken/Vietnam/NCVD-138/2008 | I117V | 25.75 (3) | 1.72 (2) | 0.4 (1) | 6.88 (4) | |
A/Owston’s civet/Vietnam/NCVD-004/2008 | I117V | 22.86 (3) | 1.5 (1) | 0.84 (1) | 4.26 (6) | |
A/duck/Vietnam/NCVD107/2007 | I117V | 19.4 (3) | 1.4 (1) | 0.9 (1) | 3.4 (5) | |
A/duck/Vietnam/NCVD-56/2007 | A395T | 41.26 (6) | 3.47 (3) | 2.86 (4) | 5.46 (8) | |
A/chicken/Vietnam/NCVD-086/2008 | D402N | 30.37 (4) | 5.42 (5) | 3.11 (4) | 7.57 (11) | |
A/duck/Vietnam/NCVD-159/2008 | V424I | 40.2 (5) | 3.1 (3) | 1 (1) | 6.1 (8) | |
A/duck/Vietnam/NCVD-160/2008 | V424I | 48.71 (7) | 3.28 (3) | 1.36 (2) | 6.02 (8) | |
Median (n = 164) | 7.40 | 1.06 | 0.72 | 1.83 | ||
Reference A(H5N1) viruses | ||||||
1 | A/Vietnam/HN30408/2005 | N295S | 2.8 (13) | 2.1 (3) | 0.7 (2) | 0.9 (2) |
A/Vietnam/HN30408/2005 | H275Y | 863.1 (4110) | 1.6 (2) | 27.1 (60) | 3.1 (6) | |
A/Vietnam/1203/2004 | None | 0.3 (1) | 0.8 (1) | 0.6 (1) | 0.5 (1) | |
Reference A(H1N1) viruses | ||||||
Not applicable | A/Texas/36/1991, oseltamivir-resistant | H275Y | 1598 (666) | 3.7 (1) | 83.7 (60) | 12.8 (5) |
A/Texas/36/1991, oseltamivir-sensitive | None | 2.4 (1) | 3.2 | 1.4 | 2.4 |
Abbreviations: IC50, 50% inhibitory concentration; NA, neuraminidase.
aIf a change in IC50 is <10-fold, then the virus is concluded to have normal inhibition by the neuraminidase inhibition assay; 10- to 100-fold is reduced inhibition (shown in bold); and >100-fold is highly reduced inhibition relative to the IC50 of the median by clade-specific (shown in bold) [11].
Clade . | Virus . | NA Change . | IC50, nM (Fold Increase)a . | |||
---|---|---|---|---|---|---|
Oseltamivir . | Zanamivir . | Peramivir . | A-315675 . | |||
1 | A/duck/Vietnam/Ncvdcdc17/2005 | K432T | 1.9 (9) | 8.12 (12) | 2.21 (5) | 3.17 (7) |
Median (n = 41) | 0.21 | 0.67 | 0.45 | 0.48 | ||
2.3.2.1 | A/duck/Vietnam/NCVD-664/2010 | H275Y | 749.55 (1027) | 0.8 (1) | 130.01 (310) | Not tested |
Median (n = 18) | 0.73 | 0.84 | 0.42 | 0.37 | ||
2.3.4.2 | A/chicken/Vietnam/NCVD-44/2007 | I223T | 20.52 (3) | 0.79 (1) | 0.78 (1) | 1.21 (2) |
A/chicken/Vietnam/NCVD-45/2007 | I223T | 18.95 (3) | 0.7 (1) | 0.75 (1) | 1.07 (1) | |
A/chicken/Vietnam/NCVD-080/2008 | I223T | 16.04 (2) | 1.01 (1) | 0.93 (1) | 1.8 (3) | |
A/chicken/Vietnam/NCVD-283/2009 | I222T | 36.85 (5) | 1.76 (2) | 1.11 (2) | 1.59 (2) | |
A/chicken/Vietnam/NCVD-295/2009 | I223T | 45.78 (6) | 1.58 (1) | 1.09 (2) | 1.22 (2) | |
A/chicken/Vietnam/NCVD-296/2009 | I223T | 44.36 (6) | 1.73 (2) | 1.12 (2) | 1.51 (2) | |
A/chicken/Vietnam/NCVD-103/2007 | I223T and S247N | 50.28 (7) | 2.69 (3) | 4.11 (6) | 12.77 (18) | |
2.3.4 | A/duck/Vietnam/NCVD-31/2007 | I223L and S247N | 102.85 (14) | 1.25 (1) | 3.84 (5) | 110.47 (153) |
2.3.4.3 | A/Vietnam/HN31412/2008 | I117V | 24.44 (3) | 2.00 (2) | 0.9 (1) | 4.64 (6) |
A/duck/Vietnam/NCVD-94/2007 | I117V | 24.82 (3) | 3.19 (3) | 0.92 (1) | 6.17 (9) | |
A/duck/Vietnam/NCVD-100/2007 | I117V | 25.33 (3) | 3.11 (3) | 1.13 (2) | 6.52 (9) | |
A/chicken/Vietnam/NCVD-106/2007 | I117V | 22.79 (3) | 1.64 (2) | 0.75 (1) | 3.67 (5) | |
A/muscovy duck/Vietnam/NCVD-120/2007 | I117V | 28.75 (4) | 2.99 (3) | 1.4 (2) | 6.34 (9) | |
A/duck/Vietnam/NCVD-118/2007 | I117V | 24.14 (3) | 1.6 (2) | 0.8 (1) | 2.92 (4) | |
A/chicken/Vietnam/NCVD-138/2008 | I117V | 25.75 (3) | 1.72 (2) | 0.4 (1) | 6.88 (4) | |
A/Owston’s civet/Vietnam/NCVD-004/2008 | I117V | 22.86 (3) | 1.5 (1) | 0.84 (1) | 4.26 (6) | |
A/duck/Vietnam/NCVD107/2007 | I117V | 19.4 (3) | 1.4 (1) | 0.9 (1) | 3.4 (5) | |
A/duck/Vietnam/NCVD-56/2007 | A395T | 41.26 (6) | 3.47 (3) | 2.86 (4) | 5.46 (8) | |
A/chicken/Vietnam/NCVD-086/2008 | D402N | 30.37 (4) | 5.42 (5) | 3.11 (4) | 7.57 (11) | |
A/duck/Vietnam/NCVD-159/2008 | V424I | 40.2 (5) | 3.1 (3) | 1 (1) | 6.1 (8) | |
A/duck/Vietnam/NCVD-160/2008 | V424I | 48.71 (7) | 3.28 (3) | 1.36 (2) | 6.02 (8) | |
Median (n = 164) | 7.40 | 1.06 | 0.72 | 1.83 | ||
Reference A(H5N1) viruses | ||||||
1 | A/Vietnam/HN30408/2005 | N295S | 2.8 (13) | 2.1 (3) | 0.7 (2) | 0.9 (2) |
A/Vietnam/HN30408/2005 | H275Y | 863.1 (4110) | 1.6 (2) | 27.1 (60) | 3.1 (6) | |
A/Vietnam/1203/2004 | None | 0.3 (1) | 0.8 (1) | 0.6 (1) | 0.5 (1) | |
Reference A(H1N1) viruses | ||||||
Not applicable | A/Texas/36/1991, oseltamivir-resistant | H275Y | 1598 (666) | 3.7 (1) | 83.7 (60) | 12.8 (5) |
A/Texas/36/1991, oseltamivir-sensitive | None | 2.4 (1) | 3.2 | 1.4 | 2.4 |
Clade . | Virus . | NA Change . | IC50, nM (Fold Increase)a . | |||
---|---|---|---|---|---|---|
Oseltamivir . | Zanamivir . | Peramivir . | A-315675 . | |||
1 | A/duck/Vietnam/Ncvdcdc17/2005 | K432T | 1.9 (9) | 8.12 (12) | 2.21 (5) | 3.17 (7) |
Median (n = 41) | 0.21 | 0.67 | 0.45 | 0.48 | ||
2.3.2.1 | A/duck/Vietnam/NCVD-664/2010 | H275Y | 749.55 (1027) | 0.8 (1) | 130.01 (310) | Not tested |
Median (n = 18) | 0.73 | 0.84 | 0.42 | 0.37 | ||
2.3.4.2 | A/chicken/Vietnam/NCVD-44/2007 | I223T | 20.52 (3) | 0.79 (1) | 0.78 (1) | 1.21 (2) |
A/chicken/Vietnam/NCVD-45/2007 | I223T | 18.95 (3) | 0.7 (1) | 0.75 (1) | 1.07 (1) | |
A/chicken/Vietnam/NCVD-080/2008 | I223T | 16.04 (2) | 1.01 (1) | 0.93 (1) | 1.8 (3) | |
A/chicken/Vietnam/NCVD-283/2009 | I222T | 36.85 (5) | 1.76 (2) | 1.11 (2) | 1.59 (2) | |
A/chicken/Vietnam/NCVD-295/2009 | I223T | 45.78 (6) | 1.58 (1) | 1.09 (2) | 1.22 (2) | |
A/chicken/Vietnam/NCVD-296/2009 | I223T | 44.36 (6) | 1.73 (2) | 1.12 (2) | 1.51 (2) | |
A/chicken/Vietnam/NCVD-103/2007 | I223T and S247N | 50.28 (7) | 2.69 (3) | 4.11 (6) | 12.77 (18) | |
2.3.4 | A/duck/Vietnam/NCVD-31/2007 | I223L and S247N | 102.85 (14) | 1.25 (1) | 3.84 (5) | 110.47 (153) |
2.3.4.3 | A/Vietnam/HN31412/2008 | I117V | 24.44 (3) | 2.00 (2) | 0.9 (1) | 4.64 (6) |
A/duck/Vietnam/NCVD-94/2007 | I117V | 24.82 (3) | 3.19 (3) | 0.92 (1) | 6.17 (9) | |
A/duck/Vietnam/NCVD-100/2007 | I117V | 25.33 (3) | 3.11 (3) | 1.13 (2) | 6.52 (9) | |
A/chicken/Vietnam/NCVD-106/2007 | I117V | 22.79 (3) | 1.64 (2) | 0.75 (1) | 3.67 (5) | |
A/muscovy duck/Vietnam/NCVD-120/2007 | I117V | 28.75 (4) | 2.99 (3) | 1.4 (2) | 6.34 (9) | |
A/duck/Vietnam/NCVD-118/2007 | I117V | 24.14 (3) | 1.6 (2) | 0.8 (1) | 2.92 (4) | |
A/chicken/Vietnam/NCVD-138/2008 | I117V | 25.75 (3) | 1.72 (2) | 0.4 (1) | 6.88 (4) | |
A/Owston’s civet/Vietnam/NCVD-004/2008 | I117V | 22.86 (3) | 1.5 (1) | 0.84 (1) | 4.26 (6) | |
A/duck/Vietnam/NCVD107/2007 | I117V | 19.4 (3) | 1.4 (1) | 0.9 (1) | 3.4 (5) | |
A/duck/Vietnam/NCVD-56/2007 | A395T | 41.26 (6) | 3.47 (3) | 2.86 (4) | 5.46 (8) | |
A/chicken/Vietnam/NCVD-086/2008 | D402N | 30.37 (4) | 5.42 (5) | 3.11 (4) | 7.57 (11) | |
A/duck/Vietnam/NCVD-159/2008 | V424I | 40.2 (5) | 3.1 (3) | 1 (1) | 6.1 (8) | |
A/duck/Vietnam/NCVD-160/2008 | V424I | 48.71 (7) | 3.28 (3) | 1.36 (2) | 6.02 (8) | |
Median (n = 164) | 7.40 | 1.06 | 0.72 | 1.83 | ||
Reference A(H5N1) viruses | ||||||
1 | A/Vietnam/HN30408/2005 | N295S | 2.8 (13) | 2.1 (3) | 0.7 (2) | 0.9 (2) |
A/Vietnam/HN30408/2005 | H275Y | 863.1 (4110) | 1.6 (2) | 27.1 (60) | 3.1 (6) | |
A/Vietnam/1203/2004 | None | 0.3 (1) | 0.8 (1) | 0.6 (1) | 0.5 (1) | |
Reference A(H1N1) viruses | ||||||
Not applicable | A/Texas/36/1991, oseltamivir-resistant | H275Y | 1598 (666) | 3.7 (1) | 83.7 (60) | 12.8 (5) |
A/Texas/36/1991, oseltamivir-sensitive | None | 2.4 (1) | 3.2 | 1.4 | 2.4 |
Abbreviations: IC50, 50% inhibitory concentration; NA, neuraminidase.
aIf a change in IC50 is <10-fold, then the virus is concluded to have normal inhibition by the neuraminidase inhibition assay; 10- to 100-fold is reduced inhibition (shown in bold); and >100-fold is highly reduced inhibition relative to the IC50 of the median by clade-specific (shown in bold) [11].
DISCUSSION
A systematic analysis of H5N1 viruses isolated in Vietnam between 2003 and 2010 allowed us to examine the evolutionary changes of these viruses and identify differences between H5N1 viruses at the AHI over time. Our spatial and temporal analysis of H5 outbreaks in Vietnam indicated that human infections occurred in regions with increased circulation of the virus in poultry [3, 7, 21–24]. With one exception of a human H5N1 infection in 2009, all other H5N1 human cases diagnosed between 2003 and 2010 coincided with H5 poultry outbreaks reported in the same or an adjacent province. Although follow-up investigations were carried out after some human cases were detected, the surveillance activities that detected poultry outbreaks were performed independent of the human cases. Thus, the correlation between human case detection and poultry outbreaks did not appear to be a result of sampling bias in this respect. On the other hand, we cannot rule out the possibility that some human cases were detected as a result of reported poultry outbreaks that may have heightened awareness of clinical symptoms in people or led to differential testing for avian influenza. Despite efforts to contain H5 virus poultry outbreaks and human infections, including systematic vaccination and massive culling of infected poultry, the genetic diversity of H5N1 viruses increased after 2006, as demonstrated by the higher number of H5 HA clades and H5N1 genotypes circulating in Vietnam between 2007 and 2010 than before 2007.
Analysis of protein sequences of H5N1 viruses circulating in Vietnam identified amino acid mutations that occurred at the AHI. We found that amino acid residue 627 of the PB2 protein was associated with adaptation of H5N1 viruses in humans. This key finding validates our overall analytical approach given that the PB2 Glu627Lys mutation at the AHI is well established as a marker of mammalian adaptation. We also identified additional mutations that arose specifically at the AHI of H5N1 viruses. However, in only a few instances did the same mutation occur in multiple, genetically distinct viral proteins, and none were as prevalent as the PB2 Glu627Lys mutation. It is likely that H5N1 virus mutations that arise at the AHI depend on the genetic background of the virus. Due to the increasing variability of H5N1 viruses, it is expected that a broad spectrum of molecular changes can occur at the AHI, yet at a relatively low prevalence in the overall virus population. The small number of recurring amino acid mutations at the AHI may also be explained by the limited genetic fitness of H5N1 viruses in humans or the possibility that some mammalian adaptive mutations were lost following virus isolation in chicken eggs. Nevertheless, the study provided a catalogue of amino acid mutations identified at the AHI, which may help future molecular-based surveillance in Vietnam and elsewhere to quickly identify substitutions associated with adaptation to mammalian hosts. Results from laboratory-generated H5N1 strains showed that only a small number of amino acid mutations were sufficient for the virus to acquire aerosol transmission between ferrets. Thus, the identification of putative markers of mammalian adaptation in nature remains critical for virus risk assessment [25–27]. Similar observations (eg, spatial and temporal overlap of H5N1 poultry outbreaks and human cases and the close genetic relationships between the viruses) were made in other countries where a large number of outbreaks and human detections have occurred, but to date these studies have been limited to a small number of virus sequences from both avian and human hosts for direct comparison [28–31].
We also investigated the susceptibility of H5N1 viruses to influenza antiviral medications. A drastic, clade-specific difference was observed in the frequency of the molecular markers for adamantane resistance. Notably, all adamantane-resistant human infections in this study were caused by the viruses from the clades with the highest frequency of drug resistance (eg, clade 1.1). The mutations associated with adamantane resistance (Ser31Asn and/or Val27Ala) were detected irrespective of the host species. In addition to the NA His275Tyr substitution, the sole established marker of oseltamivir resistance, several other NA substitutions, with potential to alter susceptibility to NA inhibitor(s), were identified in avian viruses from various clades. Some mutations were shown to reduce the ability of drugs to inhibit the viral NA activity. Furthermore, this study found that poultry viruses carrying the Ile117Val substitution were able to transmit to humans and cause a fatal outcome. H5N1 viruses with reduced susceptibility to antiviral medications pose a serious public health threat because of the limited therapeutic options and because of the ability of drug-resistant viruses to gain evolutionary advantage and spread among both poultry and humans.
Our study presents a comprehensive analysis of H5N1 virus evolution at the AHI in Vietnam between 2003 and 2010. We found a strong temporal and spatial correlation between the emergence of H5N1 viruses in poultry and their transmission to humans. The limited molecular changes in H5N1 viral proteins detected at the AHI support the hypothesis that human cases in Vietnam were the result of direct transmission from poultry with little evidence of adaptive mutations being fixed in the virus population. Despite a significant decrease in the number of H5 influenza virus poultry outbreaks in Vietnam in recent years, sustained control measures remain critical to limit the spread in poultry and sporadic human infections in Vietnam, especially following the emergence of H5N6 clade 2.3.4.4 viruses in the country. Surveillance and characterization of avian influenza viruses remain imperative to One Health efforts of identifying and implementing effective public health and animal health control and prevention measures in Vietnam.
Supplementary Data
Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Notes
Acknowledgments. We gratefully acknowledge the authors, and the originating and submitting laboratories of the sequences from the Global Initiative on Sharing All Influenza Data (GISAID) EpiFlu database, which were used in this analysis.
Disclaimer. The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention (CDC) or the Agency for Toxic Substances and Disease Registry.
Financial support. This work was supported by the CDC, US Department of Health and Human Services.
Supplement sponsorship. This work is part of a supplement sponsored by the Centers for Disease Control and Prevention.
Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
References
Author notes
Present affiliations: bVaccine Research Center, National Institute for Allergy and Infectious Diseases, Bethesda, Maryland;
Center for Global Health, Centers for Disease Control and Prevention, Atlanta, Georgia.