Field of Science

Massively parallel sequencing meets the vaccine industry

Live attenuated vaccines (LAVS), such as those produced for measles, mumps and influenza viruses, must have both high safety and immunogenicity if we are ever going to prevent human infection. Those vaccines, which are deemed unsafe, will be withdrawn resulting in low uptake and increased pathogen transmission and those vaccines which are poorly immunogenic, will not be  protective and result in pathogen transmission and significant disease

The key to easily predicting how safe a vaccine is – and also how immunogenic -, may lie in our ability to infer the phenotype (safety in humans) from the genotype (nucleic acid sequence). One problem with this is the inherent genetic instability of  RNA viruses; viruses such as polio, measles and mumps which are responsible for considerable disease in humans and which we vaccinate millions of people worldwide each year. This genetic instability results in what is generally considered as a viral ‘qausipecies’; a cloud-like structure in viral genome sequence space that can have multiple phenotypic properties: one being the safety, or lack of in humans. One example is that of oral polio vaccine strains which during production in tissue culture can accumulate genomic changes resulting in neurovirulence in humans.

In order to assess the safety we must therefore assay the genetic consistency or the types and frequency of particular changes in our vaccines prior to human administration to avoid vaccine induced disease. As I mentioned previously, our ability to assess the safety relies on our means of predicting phenotype from genotype, something that for most viruses is particularly difficult and time consuming. We are therefore  in a position in which we do not know the genetic determinants of safety and so cannot predict it based on nucleic acid sequence.

[caption id="" align="aligncenter" width="257" caption="MPS analysis of two batches of type 3 OPV performed by pyrosequencing. (A) The number of times each nucleotide was read in forward (green) and reverse (red) orientations. (B and C) Mutational profiles for vaccine batches that failed and passed the MNVT, respectively. Here and in all other figures the contents of mutants is shown by colored bars: mutations to A shown in orange, mutations to C in red, mutations to G in blue, and mutations to U in green. Neverov & Chumakov.(2010)"][/caption]

Neverov and Chumakov, from the American Food and Drug association (FDA) recently published a method in which massively parallel sequencing (MPS) is used to accurately and rapidly quantify nucleotide changes across entire poliovirus vaccine genomes.  This method proved to be very sensitive at detecting low frequency changes, changes that may have led to disease in humans. The group put forward the view that we do not truly have to know the direct relationship between genome sequence and safety but what we can do is compare the genotype and frequency of each change with previous ‘safe’ vaccine sequences. Vaccines will be allowed for human use if they have similar viral populations as a previously used strain. They offer this method as a replacement to the slower and less accurate mutant analysis by PCR and restriction enzyme cleavage (MAPREC) method.

The authors admit that the wide-scale implementation of MPS will be inhibited by the high running cost of the equipment.; a cost that they say is much less than the previously used primate neuroviruelance assay. Investment in this technology is expected to lead to a rapid decrease in price and hence will result in increased uptake of this in LAV production worldwide. Neverov and Chumakov have applied this novel sequencing technology to an important area of the vaccine industry. This application will find use in not only polio vaccines but in other LAV production and may also be implemented in the discovery of new genetic determinants of viral safety and immunogenicity.

Neverov, Alexander, and Konstantin Chumakov. 2010. Massively parallel sequencing for monitoring genetic consistency and quality control of live viral vaccines. Proceedings of the National Academy of Sciences of the United States of America 107, no. 46 (November). doi:10.1073

A mothers love declines - a measles vaccine problem?

Worldwide, measles virus infection accounts for around 200,000 deaths annually; the importance of which is emphasized given the availability of a highly effective vaccine. Vaccine effectiveness, however, is a complex matter and is subject to many problems – a major one being transfer of maternal antibodies to children during early life, a form of natural passive immunity. Although these antibodies are there for a reason and do protect offspring from infections in early life, bridging the gap until they can synthesize their own antibodies, they have been shown to inhibit the activity to certain vaccines – measles vaccine is an example (see figure below).

[caption id="" align="aligncenter" width="381" caption="Antibody concentrations in child following birth showing decline in attenuation (infection attenuation)"][/caption]

During early childhood, maternal antibody concentrations begin to wane and eventually reach such a level as to offer little protection from microbial challenge. These antibodies however are able to dampen the ability of a child to develop protective immunity following vaccination; it is this ‘window of opportunity’ that is responsible for a great number of measles virus infections and fatalities every year. The development of an effective vaccination strategy to get around this blocking effect would therefore be of great medical interest.

Recently, Kim et al (2010) publish their investigations into understanding how and why measles virus infection, in the presence of specific antibody results in the inhibition of a protective response following vaccination. Prior to this study it was unknown whether in this situation, MeV-specific B cells were being generated at all or whether they simply failed to secrete neutralizing antibody. The group used a rat model of MeV infection and simulated maternal antibody effects by passively transferring MeV specific antibodies and measuring the immunological outcomes. They demonstrated that there is a specific failure of B cells to secrete protective antibody in the presence of transferred antibody.

B cells will only secrete antibody when 3 signals are triggered:


1.     B-cell receptor/antigen interactions


2.     B – cell/ T-cell interactions


3.     Action of soluble mediators (for example: cytokines like interferon)


Kim et al hypothesized that in this model, where both signals 1 and 2 were active, inhibition of antibody secretion may be accounted for by the interference with certain soluble mediators. This idea was attractive providing the great deal of evidence showing MeV obstruction of interferon production - a pathway that normally results in the robust development of innate and adaptive immune responses. This results in two major problems involving antibody-specific inhibition of protective immune responses (maternal antibody) combined with MeV’s natural ability to inhibit the development of immunity; cases which are shared during the ‘window of opportunity’.

To this effect, the group developed a novel vaccine vector to circumvent wild-type measles interferon inhibition. Using reverse-genetics technology, they incorporated a MeV antigen gene, the haemagglutinin (HN) glycoprotein into the Newcastle Disease virus (NDV) genome as an extra gene, generating NDV-HN. NDV is an avian virus that induces high concentrations of IFNs upon infection allowing for the possibility of an effective measles vaccine in the presence of measles antibody.

The investigation confirmed the group’s predictions in that NDV-HN induced much higher levels of IFN in rat tissues when compared to MeV and that this led to development of MeV-specific neutralizing antibodies in the presence of transferred antibody. This work further verified that role that the restoration signal 3, in the form of alpha-IFN allows for B-cell secretion of antibody in in vivo and in vitro systems.

Given how medically important vaccination has been in protecting populations from often fatal and serious infectious disease and the troubles that arise when maternal antibody concentrations drop, any work developing vaccine technology to avoid these difficulties should be welcomed. Kim et al. confirmed the basis for the immunological blocks in generating MeV antibodies and began the development of a novel vector system to rationally provide protection. The results in this rat model, although not specifically applicable to the human situation, are promising in that it provides a logical framework to advance vaccine technology and prevent thousands of childhood deaths worldwide.

Kim D, Martinez-Sobrido L, Choi C, Petroff N, GarcĂ­a-Sastre A, Niewiesk S, Carsillo T. (2011) Induction of Type I Interferon Secretion through Recombinant Newcastle Disease Virus Expressing Measles Virus Hemagglutinin Stimulates Antibody Secretion in the Presence of Maternal Antibodies. J Virol. 2011 Jan;85(1):200-7. PMID: 20962092