4.3 Attenuated pathogens

A major strategy for vaccine production has been the generation of attenuated organisms, which retain their antigenicity, but which have lost their pathogenicity. Generally speaking, attenuated vaccines containing killed organisms are less effective at inducing protective immunity than those using live attenuated strains.

In the earliest attempts, the method of producing attenuated strains was to grow the pathogen in vitro, or in laboratory animals over many generations, repeatedly testing to see whether the evolving strain had lost its pathogenicity. The first and most famous example was the strain of Mycobacterium bovis developed in France by Calmette and Guérin (Bacillus Calmette Guérin, BCG), which has been used since 1923 (see Table 1) as a vaccine against M. tuberculosis. Although data from BCG studies show highly variable levels of protection, this has been one of the most widely used of all vaccines – not least because it is cheap. However, it was only with the publication of the complete genome of M. tuberculosis in 1998 that it became clear exactly what the process of attenuation had done. Early in the development of the BCG strain, the bacteria lost a group of nine genes. Moreover, since the original preparation, different strains of BCG in laboratories in different parts of the world have continued to undergo further genetic diversification. Sequencing studies have also shown that M. bovis is quite closely related to M. tuberculosis over several areas of the genome, which explains why they share critical antigens recognised by the protective immune response.

The rationale for culturing a pathogen in vitro or in a non-human species is that it does not require some of its genes (perhaps for transmission, or spread within the body) and consequently these genes may be lost or mutated, (e.g. pox viruses only appear to require 70 per cent of their genes to grow in mammalian cells). However, this process of attenuation has been described as ‘genetic roulette’, since there is no way of knowing what combination of genes will be lost or mutated and the process will produce different strains each time it is carried out.

The Polio Case Study illustrated another problem inherent in using attenuated strains – pathogenic reversion. The oral vaccine contains all three live polio strains in attenuated forms: although the type 1 strain has 57 mutations and has never reverted to the wild type, type 2 and type 3 each have only two relevant mutations, so they require only two reversions to become pathogenic again – as indeed has occurred on a number of occasions.

One of the most powerful arguments in favour of genetic engineering in vaccine production is that it can deliberately ‘knock out’ the gene sequences responsible for an organism's pathogenicity. The ability to manipulate pathogen genomes and their products is increasingly important in vaccine design, as we discuss below.