Biology Reference
In-Depth Information
TABLE 17.2
The Presence (+) or Absence (-) of Various Types of Reproductive Isolation in
Nasonia
Older Species Pair
(
N. giraulti/N. vitripennis
)
Younger Species Pair
(
N. giraulti/N. longicornis
)
Isolating Barrier
Wolbachia
-induced bidirectional CI
+
+
F1 hybrid inviability
+
Ï
F1 hybrid infertility
Ï
Ï
F2 hybrid inviability
+
Ï
F2 hybrid behavioral infertility
+
Ï
Sexual isolation
+
+
Data are based on laboratory characterizations of reproductive isolation by Breeuwer and Werren (1990, 1995),
Bordenstein et al. (2001), as well as unpublished data. Divergence times for the older and younger species pair
are approximately 0.800 and 0.250 million years ago, respectively (Campbell et al., 1993). The isolation proÝle
in the younger species pair indicates that
Wolbachia
-induced bidirectional CI and sexual isolation can evolve early
in speciation prior to the evolution of other isolating barriers.
cases were greeted with some skepticism. In the former case, it became apparent that CI was not
the only mechanism preventing gene Þow between
N. vitripennis
and
N. giraulti
. Severe levels of
hybrid inviability, hybrid sterility, and sexual isolation were also found (Breeuwer and Werren,
1995; Bordenstein and Werren, 1998; Drapeau and Werren, 1999; Table 17.2). Which form of
reproductive isolation played the primary causal role in speciation was a mystery. By not knowing
the temporal order in which these isolating barriers evolved (a difÝcult task in practice for any
speciation researcher), one could argue that
Wolbachia
came into these two species after speciation
was complete. In the
D. simulans
case, infections were still spreading through Þy populations, and
uninfected individuals were found to be regenerated each generation because of incomplete trans-
mission of the
Wolbachia
through females (Turelli and Hoffmann, 1995). Therefore, gene Þow
could still be Þuid through uninfected individuals in the two populations. Taken together, these
criticisms dampened the perceived plausibility of speciation via bidirectional CI.
The most recent study on speciation via bidirectional CI (Bordenstein et al., 2001) has to some
extent overcome these criticisms. Using the youngest
Nasonia
species pair,
N. giraulti
and
N.
longicornis
, the authors found that, in addition to different
Wolbachia
strains being Ýxed in their
respective species, bidirectional CI was the primary form of reproductive isolation: no other
postmating isolation was found, and premating isolation was weak and asymmetric (Table 17.2).
Therefore,
Wolbachia
-induced bidirectional CI in this hybridization preceded the evolution of other
forms of isolation. This study, as well as the
D. simulans
study, also show that multiple CI strains
can easily occur within a population or sister species. It is therefore plausible for
Wolbachia
to
play a causal role in speciation and act as the wedge that splits one species into two. If
Wolbachia
are speciation agents, we expect to Ýnd similar cases in other arthropod systems.
One caveat is that the species studied by Bordenstein et al. (2001) are not known to occur
sympatrically. They inhabit the eastern and western regions of North America (Darling and Werren,
1990). Therefore, bidirectional CI here will play no causal role in speciation, unless the species
come into contact or are found to already be hybridizing in geographic areas not surveyed. This
caveat outlines a task for future studies of CI and speciation Ð to discover systems of sympatric,
hybridizing species that are isolated because of CI. Here the conditions that restrict or promote CI-
assisted speciation in nature could be thoroughly investigated. However, such hybrid zones may
be transient and go unsampled, which in turn highlights the additional need for controlled, popu-
lation-cage experiments of these phenomena as well. Work by Bordenstein and Werren (unpub-
lished) indicates that CI alone can have strong effects on gene Þow in experimental populations of
N. vitripennis
. Using decay in linkage disequilibria between visible mutants as an estimate of gene
