Paradigm Shift Intervention Monitoring
twitter
Live feed of underlying pandemic map data here
Commentary
CDC On D225G In Egg
Isolates Raises H1N1 Vaccine Concerns
Recombinomics
Commentary 14:58
March 7, 2010
a study done by the WHO Collaborating Centres for Reference and Research on Influenza (WHOCC) in Atlanta located in the Centers for Disease Control and Prevention (CDC) found the D222G substitution in 14 virus isolates but not in viruses in the original clinical specimens indicating the D222G substitution in these 14 virus isolates occurred after growth in the laboratory.
The above comments from a response to the recent report from Norway on the significant association between D225G and severe/fatal H1N1 cases, and a WHO preliminary report of D225G, respectively, clearly indicate that the CDC has concluded that samples that yield sequences with D225G in virus grown in chicken embryos are artifact and do not represent D225G in the clinical sample. The CDC has published HA sequences from at least 15 patients with D225G (see list below). Most of these sequences are from initial pandemic H1N1 isolates, collected in April last spring, including the current pandemic vaccine target, A/California/7/2009. These initial isolates were generally from students in California (San Diego County), Texas (San Antonio area), and New York (Queens), and were generally mild. Although D225G was detected (alone or as a mixture with wild type) in virus grown in eggs, it was not detected in direct sequencing or in virus grown in mammalian cells. This pattern has been interpreted by the CDC as a lab generated artifact, although the data would also support an alternative interpretation of a mixture, with D225G representing a minor component.
D225G is a receptor binding domain change that recognizes gal 2,6 and gal 2,3 receptors. Gal 2.3 receptors are on chicken embryo cells in eggs, as well as many cell types in the human lung. Human receptors in the upper respiratory tract as well as mammalian cells used to grow virus in the lab has largely gal 2,6 receptors. Thus, in direct sequencing of the clinical sample, the major strain will be detected, which would be wild type in samples with D225G as a minor component. The same result would be found in virus grown in mammalian cells because wild type would maintain its numerical advantage. However, virus grown in eggs would have D225G, because the gal 2,3 specificity would allow D225G to out compete wild type, changing the ratio and allowing for D225G detection either alone or as a mixture with wild type, which was found in the April isolates listed below. After April, CDC and most labs used mammalian cells to grow virus in the lab, selecting against D225G and allowing detection only in samples with high levels of D225G.
These ratios would explain why the April cases with D225G were mild. The D225G was at low levels and consequently not detected in direct sequencing or virus grown in mammalian cells, but detected in virus grown in eggs. More recent isolates, including those in Norway, which involved detection of D225G in mammalian cells, but frequently as mixtures with wild type, or via direct sequencing of autopsy lung samples from Ukraine, where D225G, D225N, or both were detected because D225G/N was at high levels, which strongly correlated with fatal cases.
The conclusion that the D225G in the April samples, including A/California/7/2009, may have influenced the decision to select X-181A, which lacks D225G, as the killed vaccine. target. This absence is likely linked to the low yields that led to vaccine shipment delays resulting in the delivery of most vaccine after the fall peak in cases. In contrast, the live attenuated vaccine made by MedImmune had D225G and grew well in eggs, leading to shipment weeks ahead of the killed vaccine.
Thus, the CDC conclusion that D225G identified in eggs was a lab artifact may have had significant consequences regarding the selection of the vaccine target in April, as well as the recent recommendation that the vaccine can be used in the 2010/2011 season, when D225G may be much more prevalent and killing many more people.
D225G (CDC generated sequences)
name age/gender date
A/Texas/05/2009 16M 4/15/2009
A/Texas/11/2009 9M 4/23/2009
A/Georgia/01/2009 12F 4/27/2009
A/New York/04/2009 17F 4/2009
A/Illinois/10/2009 55M# 7/31/2009
A/North Carolina/53/2009 53M 10/18/2009
mixture with wild type
A/New York/31/2009 14F 4/24/2009
A/New York/11/2009 16F 4/25/2009
A/California/13/2009 4/21/2009
A/Texas/10/2009 6F 4/23/2009
A/California/07/2009 54M 4/9/2009
A/Nebraska/02/2009 48M 4/2009
A/Texas/04/2009 16M 4/14/2009
A/Utah/42/2009 28F$* 7/24/2009
A/North Carolina/39/2009#* 43F 10/15/2009
# = H274Y
* = known or likely fatal
$ = D225N mixture
Media Links
Recombinomics
Presentations
Recombinomics
Publications
Recombinomics
Paper
at Nature Precedings