52 – I.1. BACTERIA: PATHOGENICITY FACTORS
Phenotypic evidence: Within the genetic methods, two approaches are used:
1) inactivation of a virulence gene must result in loss of virulence; or 2) introduction of a
virulence gene into a non-virulent strain must add virulent properties. It should be noted
that both principles are heavily dependent on models to determine the virulent phenotype.
Models to determine virulence are ideally animal models that minutely mimic natural
disease, but these are not always available. More often, animal models have to be used
that display only some of the naturally occurring characteristics, or in vitro models that
only poorly resemble disease characteristics. Most processes leading to virulence are
multi-factorial. The complicated interaction of host and bacteria is often ignored when
in vitro models are applied. Even under simplified conditions of in vitro models,
a presumably straightforward process such as bacterial invasion can be driven and
regulated by multiple genes and gene loci, which work in concert or complementarity.
Inactivation of one link of the chain may eliminate invasiveness, but complementation in
a different setting may require several genetic loci. Alternatively, inactivation of a factor
may be overcome by alternative factors so that loss of virulence is not observed, but
complementation in a different environment may have strong phenotypic effects.
The relevance of the applied models to extrapolate their outcome as phenotypic evidence
of virulence is a point of debate, which is pragmatically ignored by lack of alternatives.
Immunological evidence: A second approach for identifying virulence genes is based
on the proposed immunogenicity of virulence factors. Knowing that acquired immunity
can protect against disease, it is assumed that protective antibodies are directed against
virulence-associated genes. When an infection results in an antigenic response directed
against one or more specific antigens, this is taken as a strong indication that these
antigens represent virulence-associated factors.
Comparative genetical evidence: Examples of the genetic approach to identification
of potential virulence-associated genes are the identification of: 1) genes with a degree of
homology to known virulence-associated genes that is considered significant in
bioinformatics surveys; 2) related genes that show variation that can be interpreted as
antigenic variation; 3) genes that are shown to be present in (more) virulent strains, while
absent in avirulent strains. Comparative genetic approaches are further discussed in the
section on trends in virulence gene identification.
In addition to these approaches, several techniques have been developed to identify
and characterise bacterial genes that are induced during the intracellular infection and
therefore, potentially, may play a role in pathogenesis. Examples are the search for genes
that are specifically induced in the host, and “signature-tagged mutagenesis” (STM),
involving comparative hybridisation to isolate mutants unable to survive the
environmental conditions in the host (Mahan, Slauch and Mekalanos, 1993; Chiang,
Mekalanos and Holden, 1999; Harb and Abu Kwaik, 1999). A very powerful approach to
isolate mutants that may be affected in a virulence gene is STM as discussed by Autret
and Charbit (2005). The general technique of STM can be applied to find specific genes
involved in survival persistence of a bacterium in a host; virulence genes would fall into
this class of genes (Wassenaar and Gaastra, 2001). The only prerequisite for a gene to be
found by STM is that its loss of function should not result in a lethal phenotype under
conditions of growth in vitro, in broth. This is probably not an impediment for most
virulence genes to be identified by this technique. The STM approach involves
transposon (usually) mutagenesis of a bacterial strain, followed by pooling of a number of
mutants that can be individually recognised by a polymerase chain reaction amplifiable
tag. The pooled mutants are inoculated in an animal model, and bacteria retrieved from
the animal are analysed for mutants that are present, as shown by the presence of their
