310 Invasive Stink Bugs and Related Species (Pentatomoidea)
retarded development, arrested growth, and abnormal body coloration, suggesting that the symbionts are
necessary for normal growth and development (Hosokawa et al. 2006). The symbionts also appear to
affect behavior of the first instars; those acquiring the symbionts rested on the egg mass whereas those
deprived of symbionts tended to wander off the mass (Hosokawa et al. 2008). Hosokawa et al. (2005)
determined the components of the symbiont capsules and where they are made, stored, and excreted.
They subsequently (Hosokawa et al. 2006) demonstrated that the symbiont was necessary for normal
development and proposed the name Candidatus Ishikawaella capsulata for it. Hosokawa et al. (2007a)
found that females produced one endosymbiont capsule for about 3.6 eggs irrespective of clutch size.
They manipulated symbiont titer in first instars and estimated that the titer necessary for successful
vertical transmission of the symbionts was 1.9 x 10^6 , which was only 1/10 of the actual titer detected
in newborn nymphs. Based on the actual titer detected in newborn nymphs, one capsule was sufficient
for symbiont transmission to six nymphs and females produce 1.7 times more symbiont capsules than
needed.
Hosokawa et al. (2007b) detailed differences between two populations of Megacopta in Japan, indicat-
ing that ‘M. punctatissima’ is found in an area from mainland Japan to Yakushima Island and feeds pri-
marily on kudzu but can invade soybean fields and become a pest. Bugs from the southwestern Japanese
islands (‘M. cribraria’) feed mainly on ‘Taiwan-kudzu’ and rarely reach pest status. Exchanging sym-
biont capsules resulted in the non-pest developing normally on soybean and the pest not developing
normally. Thus, the strain of endosymbiont appears to determine the developmental success or lack
thereof on various host plants. Nikoh et al. (2011) provided a complete genome sequence for Candidatus
Ishikawaella capsulata and stated that the genome suggests this symbiont provides essential amino acids
for the plant-feeding host.
A second endosymbiont, the alphaproteobacterial Wolbachia, also occurs commonly in populations
of Megacopta cribraria both in the United States and Japan (see Section 5.5.3 under Endosymbionts for
further discussion). Kikuchi and Fukatsu (2003) studied the diversity of these organisms and concluded
M. cribraria from different areas of Japan each had two strains of Wolbachia. One was shared by both
populations whereas each population had one unique strain.
5.5 Population Genetics
5.5.1 Mitochondrial DNA: Initial Studies
There were two objectives for the initial genetic studies conducted during the first two months following
the discovery and identification of Megacopta cribraria in Georgia: (1) provide a genetic marker that
would afford a one-to-one correlation between morphological taxonomy (Eger et al. 2010) and molecu-
lar taxonomy (Jenkins et al. 2010), and (2) standardize protocols for field collections obtained for DNA
extraction, sequencing, and statistical analyses (Jenkins et al. 2002, 2007, 2009). Because taxonomic
verification, genetic diversity, and insect origin were priorities, mtDNA was chosen because it is a non-
recombining haploid molecule that is inherited maternally, evolves four times faster than nuclear DNA
(nuDNA), and is relatively easy to amplify due to multiple copies in each cell (Zink and Barrowclough
2008). A 2336 bp mtDNA fragment, which included the 5’-tyrosine (Y) tRNA, cytochrome oxidase
subunit I (COI), leucine tRNA (L), and partial cytochrome oxidase subunit II (COII) genes (COI-L-
COII) (Table 5.2), was sequenced from all samples. By October, 2010, M. cribraria was confirmed in
80 counties in Georgia and collections processed for the 2336 bp mtDNA fragment showed only a single
haplotype, designated GA1 (GenBank No. HQ444175) (Table 5.2).
The mitochondrial genome (15,647 bp) of five adult bugs randomly collected from four Georgia
counties where Megacopta cribraria was found initially in 2009 (Suiter et al. 2010, Jenkins and Eaton
2011) showed consensus (GenBank No. JF288758). The COI-L-COII gene fragment within each of
these genomes was GA1 (Tracie M. Jenkins, unpublished data), a conclusion independently verified
by Amanda M. V. Brown. It appears, therefore, from these data that only one female lineage was
introduced into Georgia and was rapidly and predictably (Zhu et al. 2012) dispersing throughout the
southeast.