Biology Reference
In-Depth Information
Although a certain selective advantage is evident for a nuclear gene to induce such a phenotype,
certain conditions are necessary for a
Medea
-inducing gene when localized in the genome of a
maternally inherited symbiont. In the case of
A. tabida
,
Wolbachia
will only increase in frequency
when the sterilization of uninfected daughters increases the Ýtness of the infected ones. For example,
selective pressure could occur when high resource competition exists between the offspring of
closely related females. Such competition is likely to occur in
A. tabida
, where progeny can be
laid in the same local host resource and where cases of superparasitism (i.e., more than one parasitic
egg is deposited in each
Drosophila
larvae by several conspeciÝc females) have been described
(van Alphen and Nell, 1982).
Because this
Medea
effect resembles the modiÝcation-rescue model of CI, it can easily be
compared to
WolbachiaÏ
induced CI, and an evolutionary switch from CI
to
Medea
induction has
recently been proposed to explain the case of
A. tabida
(Charlat and Merot, 2001).
Direct Involvement of Bacterial Factors in Insect Oogenesis?
A second hypothesis that may explain how
Wolbachia
is involved in the reproductive physiology of
females is that bacterial factors actively participate in some metabolic pathways necessary to oogenesis.
As recently reviewed by McFall-Ngai (2002), many bacteria are involved in the development,
survival, and reproduction of their hosts, forming obligatory symbiosis, where both the host and
the symbiont depend on the association. In insects, such microorganisms are common, and they
form distinctive associations called Ñprimary symbiosisÒ (Moran and Baumann, 2000) or Ñbacte-
riocyte symbiosisÒ (Douglas, 1998). This type of symbiosis groups all cases where strictly intrac-
ellular microorganisms are restricted to one host cell type (called bacteriocyte), which is specialized
in housing them. In such primary symbiosis, it is widely accepted that host dependence has a
nutritional base, making exploitation of nutritionally poor or unbalanced diets possible through
providing essential nutrients such as amino acids, vitamins, or lipids (Douglas, 1994, 1998). In
these intimate associations, host and symbiont phylogenies are most often congruent, demonstrating
stable infection, cospeciation, and long coevolutionary history of partners (Moran and Baumann,
2000). This primary symbiosis opposes ÑsecondaryÒ (or facultative) symbiosis, where the associ-
ation is sporadic and bacteria occupy various host tissues. Effects on the host are usually not known,
and in all cases symbionts are not obligatory to host survival or reproduction.
Because infection is obligatory for
A. tabida
reproduction, we could consider their
Wolbachia
primary symbionts. However, they also share several features with secondary symbionts. First,
since they can be detected in various body parts (Dedeine et al., 2001), they are not restricted to a
precise cell type. Second, closely related
Asobara
species (
A. citri
,
A. persimilis
, and another
undescribed species from the United States) are naturally free of
Wolbachia
(F. Dedeine, unpub-
lished data), suggesting recent common evolutionary history of partners. Moreover, we would
expect a major difference in the evolutionary origin of bacterial dependence. In primary symbiosis,
host dependence probably originated in acquisition of beneÝcial bacterial metabolic functions that
permitted the insect to exploit poor diets and consequently to extend (or change) their previous
ecological niche. In this evolutionary scenario (the Ñevolutionary noveltyÒ), symbionts would have
become obligatory for the host by providing additional functions that are necessary for exploiting
speciÝc habitats. This scenario could apply to a number of symbiosis cases (Margulis and Fester,
1991) but certainly not to the
WolbachiaÏA. tabida
association, since autonomous egg production
by insects obviously preexisted
Wolbachia
acquisition.
We shall now consider a hypothetical ÑsubstitutionÒ evolutionary scenario. In an initial step, the
ancestral lineage of
A. tabida
would have become associated with a
Wolbachia
strain (
w
At3) that
spread into host populations thanks to CI induction. This
Wolbachia
may have had a particular genetic
determinant encoding for some factor directly involved in oogenesis of the wasp. In this situation,
both host and bacterial genes would have acted in the same insect function, and such nucleocytoplas-
mic redundancy could therefore have resulted in the loss of nuclear determinants by the insect, making
the bacteria totally obligatory for reproduction. A similar substitution process may have occurred in
