ArticlePDF Available

Origins, Divergence, and Contrasting Invasion History of the Sweet Potato Weevil Pests Cylas formicarius (Coleoptera: Brentidae) and Euscepes batatae (Coleoptera: Curculionidae) in the Asia-Pacific

Abstract and Figures

Cylas formicarius F. and Euscepes batatae Waterhouse are the most damaging sweet potato insect pests globally. Both weevils are thought to have invaded the Pacific alongside the movement of sweet potato (Ipomoea batatas (L.) Lam. Convolvulaceae), with C. formicarius having originated in India and E. batatae in Central or South America. Here we compare the genetic relationships between populations of the pests, primarily in the Asia-Pacific, to understand better their contemporary population structure and their historical movement relative to that of sweet potato. Cylas formicarius has divergent mitochondrial lineages that indicate a more complex biogeographic and invasive history than is presently assumed for this insect, suggesting it was widespread across the Asia-Pacific before the arrival of sweet potato. Cylas formicarius must have originally fed on Ipomoea species other than I. batatas but the identity of these species is presently unknown. Cylas formicarius was formerly designated as three species or subspecies and the genetic data presented here suggests that these designations should be reinvestigated. Euscepes batatae has very low genetic diversity which is consistent with its historical association with sweet potato and a recent introduction to the Asia-Pacific from the Americas. The distribution of E. batatae may be narrower than that of C. formicarius in the Asia-Pacific because it has relied relatively more on human-assisted movement. Consequently, E. batatae may become more widespread in the future. Investigating the invasion history of both species will help to understand the probability and nature of future invasions.
Content may be subject to copyright.
© The Author(s) 2019. Published by Oxford University Press on behalf of Entomological Society of America. All rights reserved. For permissions, please
Molecular Entomology
Origins, Divergence, and Contrasting Invasion History
of the Sweet Potato Weevil Pests Cylas formicarius
(Coleoptera: Brentidae) and Euscepes batatae
(Coleoptera: Curculionidae) in the Asia-Pacific
DeanR. Brookes,1 JamesP. Hereward, GimmeH. Walter, and MichaelJ. Furlong
School of Biological Sciences, The University of Queensland, Brisbane, 4072 Queensland, Australia and1Corresponding author,
Subject Editor: ScottGeib
Received 11 February 2019; Editorial decision 24 June 2019
Cylas formicarius F. and Euscepes batatae Waterhouse are the most damaging sweet potato insect pests glo-
bally. Both weevils are thought to have invaded the Pacific alongside the movement of sweet potato (Ipomoea
batatas (L.) Lam. Convolvulaceae), with C.formicarius having originated in India and E.batatae in Central or
South America. Here we compare the genetic relationships between populations of the pests, primarily in
the Asia-Pacific, to understand better their contemporary population structure and their historical movement
relative to that of sweet potato. Cylas formicarius has divergent mitochondrial lineages that indicate a more
complex biogeographic and invasive history than is presently assumed for this insect, suggesting it was wide-
spread across the Asia-Pacific before the arrival of sweet potato. Cylas formicarius must have originally fed on
Ipomoea species other than I.batatas but the identity of these species is presently unknown. Cylas formicarius
was formerly designated as three species or subspecies and the genetic data presented here suggests that
these designations should be reinvestigated. Euscepes batatae has very low genetic diversity which is con-
sistent with its historical association with sweet potato and a recent introduction to the Asia-Pacific from the
Americas. The distribution of E.batatae may be narrower than that of C.formicarius in the Asia-Pacific because
it has relied relatively more on human-assisted movement. Consequently, E.batatae may become more wide-
spread in the future. Investigating the invasion history of both species will help to understand the probability
and nature of future invasions.
Key words: Ipomoea, sweet potato weevil, West Indian sweet potato weevil, secondary contact, phylogeography
The most damaging weevil pests of sweet potato (Ipomoea batatas)
globally are the sweet potato weevil, Cylas formicarius F.and the
West Indian sweet potato weevil, Euscepes batatae Waterhouse
(Coleoptera: Curculionidae) (Chalfant et al. 1990). Both species
feed regularly on several species of Ipomoea (Chalfant etal. 1990),
and their close association with sweet potato has inuenced their
respective invasive distributions. These weevil species have different
evolutionary and geographic origins, with C.formicarius originating
from Asia (Wolfe 1991) and E.batatae from tropical America. It is
not clear, however, to what extent their current distribution and gen-
etic relationships have been facilitated by the multiple movements
of sweet potato by humans (Roullier etal. 2013). Cylas formicarius
is thought to have originated in India (Wolfe 1991) and invaded
much of the tropical and subtropical world from there, including
the Americas, Africa, Asia and much of the Pacic. Cylas formicarius
must, then, have evolved through interactions with Ipomoea spe-
cies other than the New World I. batatas. These other Ipomoea
species may well have inuenced the distribution and abundance of
C.formicarius before the arrival of sweet potato and may even do so
today. Euscepes batatae originates from tropical America, where the
sweet potato and its progenitors also originated less than one million
years ago (Mya) before their recent cultivation by humans (Roullier
et al. 2013, Munoz-Rodriguez et al. 2018a). Thus far, E. batatae
is thought to have invaded only parts of the Pacic (Papua New
Guinea [PNG], Fiji, Tonga, and Samoa) and parts ofJapan.
Sweet potato is a primary source of carbohydrates in much of the
Pacic and parts of Africa (McGregor etal. 2016, Low etal. 2017,
Iese etal. 2018). Cylas formicarius and E.batatae oviposit in both
the vines and roots of sweet potato plants (Talekar 1995, Yasuda
1997a) and larval feeding, particularly in the roots, can result in
applyparastyle "g//caption/p[1]" parastyle "FigCapt"
Journal of Economic Entomology, XX(XX), 2019, 1–9
doi: 10.1093/jee/toz198
Copyedited by: OUP
Downloaded from by University of Queensland user on 16 August 2019
signicant crop damage and economic loses (Jackson et al. 2012,
Yasuda 1997b). Crop losses are threatened by further invasions
and a lack of knowledge about the local ecology of each species.
Therefore, understanding the host use and dispersal of these weevil
pests is crucial for limiting their damage to sweet potato crops. The
distribution of E. batatae is much more restricted geographically
than that of C. formicarius and this pattern indicates that these
two weevil species have not been affected equally by the various
movements of sweet potato through Asia and the Pacic (described
below). Consequently, E.batatae may not have reached the full ex-
tent of its possible invasive range. The different distributions of these
weevils may be a consequence of the ecological differences between
them, the apparently more limited dispersal capabilities of E.batatae
(which are thought to be ightless; Sherman and Tamashiro 1954,
Alleyne 1982, Moriya and Miyatake 2001), the weevils having
had different invasion pathways, or some combination of these.
Addressing each of these possibilities rst requires characterization
of the genetic relationships across the populations of C.formicarius
and E.batatae.
Sweet potato has a complex history of human-assisted move-
ment that has undoubtedly also assisted in the movement of in-
dividuals of C. formicarius and E. batatae. Recently, it has been
suggested that sweet potato might have naturally dispersed from
the Americas to Polynesia in pre-human times (Munoz-Rodriguez
etal. 2018a,b) but, conventionally, sweet potato movement out of
the Americas is considered to have occurred through three primary
dispersal events, (the tripartite hypothesis; Barrau 1957, Yen 1974,
Denham 2013, Roullier etal. 2013). These are thought to have been
an initial introduction of sweet potato into the Pacic from South
America by Polynesian travelers about 1000 A.D., followed by two
separate introductions by Europeans about 500 yr later, one into the
Philippines from Central America and another into Indonesia from
the Caribbean and Central America. These introductions were then
followed by many secondary introductions (Roullier et al. 2013),
and the widespread introduction of many varieties in modern times,
including those perpetuated by colonial agricultural intensication
programs (Iese etal. 2018). The various introductions of sweet po-
tato into Asia and the Pacic (Roullier etal. 2013) have almost cer-
tainly affected C.formicarius and E.batatae differently, particularly
given their different geographic origins and different associations
with their host plants.
We used molecular methods to investigate the genetic diversity
of C.formicarius and E.batatae individuals from different localities
across Asia and the Pacic. Our intent was to assess the broad gen-
etic diversity of the species, rather than to comprehensively assess
that diversity for any one population. For C.formicarius, these data
were compared with publicly available sequence data. The results
presented here are a preliminary test of whether the genetic diversity
of these weevils reects their presumed invasion histories, and of
the relatedness of the populations in these regions. These results ex-
pand our understanding of the evolutionary and invasion histories of
the weevil genera Cylas and Euscepes relative to plants in the genus
Ipomoea, and in particular the cultivated sweet potato I. batatas.
Many questions remain, but the insights provided by this work allow
them to be addressed.
Materials and Methods
Sampling, DNA Extraction, and Amplification
Individuals of C. formicarius and E. batatae were collected by
manual inspection of I.batatae plants from several locations in the
Pacic, and from single locations in each of Florida and Australia
(Table 1). Adult male C. formicarius were also collected by placing
yellow sticky traps (20× 10cm) baited with synthetic female sex
pheromone (lure: P-299, Chemtica Internacional, Heredia, Costa
Rica) in I.batatas crops at sites across the Pacic. Individuals were
placed into 95% ethanol for transport and for freezer (−20°C)
storage before DNA extraction.
DNA was extracted from single whole adult insects (but larvae
from three sites: TOeu17–TOeu19, Table 1) after the insect was
Table 1. Sampling locations, sampling dates, latitude, longitude, number of individuals sequenced at the COI and CAD (rudimentary) gene
regions, and site codes (representing independently sampled regions) link individuals to GenBank accession numbers (Supp. Table S1
[online only]) for the Cylas formicarius and Euscepes batatae sequenced in thisstudy
Species and location Date Latitude Longitude COI CAD Site codes
Cylas formicarius
Bundaberg, Australia Mar. 2016 24°4341152°15408 6 BUNcy
Nadi Viti Levu, Fiji April 2015 17°459177°27536 - FJcy
Rewa Viti Levu, Fiji Nov. 2017 17°5538 178°25276 6 VLcy
Guadalcanal, Solomon Islands Nov. 2017 9°2559160°5106 6 GLcy
Makira, Solomon Islands July 2015 10°2433161°42504 - MAKcy
Foa, Tonga Jan. 2016 19°4450 174°17468 4 TO1cy
Lifuka, Tonga Jan. 2016 19°498 174°206015 7 TO2cy
TongaaDec. 2015 - - 4 2 TO3cy
Western Highlands, Papua
New Guinea
June 2017 5°4952 144°12393 5 WPcy
Lae, Papua New Guinea Oct. 2016 6°4018146°54402 - PGcy
Florida, United States June 2018 - - 6 6 FLcy
Euscepes batatae
FijiaMay 2016 - - 1 - FJeu
Foa, Tonga Jan. 2016 19°4450 174°17462 - TO1eu
TongaaDec. 2015 - - 25 - TO2eu
Western Highlands, Papua
New Guinea
June 2017 - - 6 - WPeu
Upolu, Samoa May 2016 5°4952144°12394 - SUeu
aMore precise localities are not known for locations.
Journal of Economic Entomology, 2019, Vol. XX, No. XX
Copyedited by: OUP
Downloaded from by University of Queensland user on 16 August 2019
suspended in 200 ul of lysis buffer and 5 ul of Proteinase K and
then crushed with a bead beater (Bullet Blender Storm 24, Next
Advance, United States) with three 3.2-mm stainless steel balls at the
highest speed for 5min for C.formicarius and 10min for E.batatae.
Crushed samples were lysed overnight at 55°C, before a spin column
extraction protocol was followed (Ridley et al. 2016) that used
lab-prepared buffers and 96-well column extraction plates (Epoch
Life Sciences, United States). Low initial polymerase chain reaction
(PCR) success (determined using a 2% agarose gel) with adult in-
sects, but not larvae, suggested PCR inhibitors were present in the
samples or that DNA quality was poor. Following DNA extraction
for adults, DNA was therefore cleaned to remove short DNA frag-
ments and any PCR inhibitors using PCRCleanDX magnetic beads
(Aline Biosciences, United States), following the manufacturer’s
protocol with a 1:1.7 ratio of DNA to beads. Optimal sampling
strategies need to be considered for future studies on these weevils
(Wadl etal. 2019).
Two gene regions, one mitochondrial (cytochrome c oxidase sub-
unit I[COI]) and one nuclear (rudimentary [CAD]), were PCR amp-
lied for C.formicarius individuals. The CAD gene was chosen from
among several primer combinations for nuclear genes that were
tested and had been used previously in beetle phylogenetics (Wild
and Maddison 2008). Following low initial amplication success
with COI primers (HCO2198 and LCO1490) for C. formicarius,
degeneracies were added to the HCO2198 primer (Folmer et al.
1994) (Table 2) from the LEP-R1 primer (Foottit etal. 2008). Only
the COI gene region was amplied for E.batatae individuals, and no
nuclear primers were optimized for E.batatae individuals because of
the low COI diversity found in this species. PCR conditions were 1×
MyTaq HS buffer, 0.2µm of forward and reverse primer, 1 unit my
MyTaq HS DNA polymerase (Bioline, United Kingdom), 3.0µl of
DNA template, and a total reaction volume of 20µl. PCR protocols
were one cycle of 90°C for 1min and then 40 cycles of the following;
95°C for 30s, annealing for 45s and 72°C for 1min 15s. Anal
cycle of 72°C for 10min was used. Annealing temperatures were
51°C for COI and 52°C for CAD. PCR primers and their source are
listed in Table 2.
All PCR products were cleaned using Exonuclease I and
Antarctic Phosphatase (New England Biolabs, United States) and
sequenced in both directions by Macrogen (Korea). Sequence data
was trimmed, aligned, and checked using CodonCode Aligner ver-
sion 4.1.1 (CodonCode Corporation, United States).
Publicly Available Sequences
Publicly available sequences matching the sequences obtained in this
study were searched for using species names for the GenBank and
BOLD databases and using BLASTn for the GenBank database as of
24/ix/2018. COI sequences for C.formicarius were obtained from
GenBank and BOLD (Table 3) but no such sequences were avail-
able for E. batatae. Cylas outgroup sequences were also obtained
from GenBank for the COI gene region from C.brunneus (Accession
No. MF510129), but no closely related outgroup could be found for
E.batatae. No publicly available sequences were available for the
CAD gene region for C.formicarius.
Genetic Analyses
All sequences were aligned using MAFFT (Katoh etal. 2002, Katoh
and Standley 2013). For the C.formicarius COI gene, the sequences
from this study and the publically available sequences were com-
bined into a single dataset comprised of a 371bp overlapping re-
gion, and this dataset was used for all further COI analyses for this
Table 2. PCR primers, the primer sequences and whether the primers were used with Cylas formicarius or Euscepes batatae (or both of
them; the source of each primer is also listed)
Primer Sequence (5–3) Species Source
HCO2198*Lep TAAACTTCWGGRTGWCCAAAAAATCA Both This study; modied from Folmer etal.
(1994) and (Foottit etal. 2008)
LCO1490 GGTCAACAAATCATAAAGATATTGG C.formicarius Folmer etal. (1994)
TYJ1460d TACAATYTATCGCCTAAACTTCAGCC E.batatae Simon etal. (1994)
CAD821F AGCACGAAAATHGGNAGYTCNATGAARAG C.formicarius Wild and Maddison (2008)
CAD1098R4 GCTATGTTGTTNGGNAGYTGDCCNCCCAT C.formicarius Wild and Maddison (2008)
Table 3. Sample locations and number of individuals for Cylas formicarius sequences obtained from BOLD and GenBank (the sequence
obtained for the outgroup species, C.brunneus, is also included)
Location nDatabase Accession number Authors
Townsville, Australia 1 BOLD COQT119-08 Unpublished (Cocks 2018)
Karnataka, India 3 GenBank KP233785, KU666424, KM459451 Unpublished (Rakshit etal.
(2014), (Sangeetha etal.
2016), (Rakshit 2018)
Okinawa, Japan 5 GenBank AB470606 - AB470608, AB470613 -
Unpublished (Sukenari etal.
Tokyo, Japan 2 GenBank AB470609, AB470610 Unpublished (Sukenari etal.
Myanmar 1 GenBank MF804620 Unpublished (Blaimer and
Mulcahy 2017)
Taiwan 2 GenBank AB470611 - AB47612 Unpublished (Sukenari etal.
United States/Unknown 1 GenBank FJ867849 McKenna etal. (2009)
Ghana (C.brunneus) 1 GenBank MF510129 Billah etal. (2017)
Journal of Economic Entomology, 2019, Vol. XX, No. XX
Copyedited by: OUP
Downloaded from by University of Queensland user on 16 August 2019
species. Ambiguous nucleotides were found at the CAD gene region
and so DNAsp 6.1.1 (Rozas etal. 2017) was used to estimate the
phased haplotype frequencies using the recombination model with
1× 104 iterations and 1 × 103 burn-in iterations. The sequence of
C.brunneus was the designated outgroup for all C.formicarius COI
phylogenetic analyses. Haplotype diversity (Hd) was calculated and
neutrality tests (Tajima’s D and Fu’s FS) performed for the COI gene
region of C.formicarius and E.batatae using DNAsp 6.12 (Rozas
etal. 2017). Localities with ve or fewer individuals were not ana-
lyzed independently of the total dataset. Haplotype networks were
created for each gene region for each species using PopArt (Leigh
and Bryant 2015) using the TCS method (Clement etal. 2000).
All phylogenetic analyses were performed using unique sequences.
The appropriateness of different evolutionary models for the phylo-
genetic analysis was determined using JModelTest 2.1.1 (Darriba etal.
2012). Maximum likelihood analysis was used to create individual
phylogenetic trees for the COI gene of C. formicarius using PhyML
3.0 (Guindon et al. 2010) with 1,000 bootstrap replicates as imple-
mented in Geneious 9.05 (Biomatters, New Zealand). The COI gene
of C. formicarius was also analyzed using Bayesian inference using
MrBayes 3.2.6 (Huelsenbeck and Ronquist 2001) with a chain length of
1× 106 and a molecular clock with a range of probable mutation rates
(1.15× 10−2 to 1.75× 10−2) (Brower 1994, Papadopoulou etal. 2010).
The trace and effective sample size (ESS) was examined for each MCMC
parameter to ensure they were adequately sampled. TreeAnnotator 2.4.8
(Bouckaert etal. 2014) was used to select a maximum clade credibility
tree with a burn-in of 10% and a posterior probability (PP) limit of 0.7.
The CAD gene for C.formicarius and the COI gene of E.batatae had
low complexity and were not used for phylogenetic analyses.
Genetic Diversity of C.formicarius and E.batatae
Individuals of C. formicarius had a total of 20 haplotypes over a
371bp common fragment of the COI mtDNA gene region (Fig.1)
with the highest pairwise divergence being 11.6%. Divergence
time estimates, from the Bayesian analysis of C. formicarius COI
sequences (Fig. 2) suggests that these lineages diverged beginning
about 6.2 to 13.8 Mya with the initial separation of the lineage
found in India from all others (Fig. 2). Cylas formicarius individ-
uals had a total of four haplotypes at the CAD nuclear gene region
(Fig. 1). COI sequences obtained from C. formicarius had much
higher genetic diversity than those from E.batatae, which had only
three COI haplotypes over a 669bp fragment of the gene (Fig. 3)
with the highest pairwise divergence being 1.2%. The large differ-
ence in diversity between the species was evident even though the
fragment common to all C.formicarius sequences (those generated
in this study and those publicly available) was shorter in length than
those from E.batatae. Populations with individuals belonging to di-
vergent clades, mostly Fiji and Tonga, had signicant results in the
neutrality tests (Table 4), indicating that these populations are not
at equilibrium, probably because of secondary contact between di-
vergent lineages. COI sequences were consistent with the evolution
of true mitochondrial genes, having no ambiguities, no stop codons,
and fewer transversions than transitions (Zhang and Hewitt 1996).
The CAD sequences also had no stop codons and fewer transversions
than transitions.
Cylas Phylogenetic Analyses
The substitution model with the most support for Cylas COI was
Hasegawa-Kishino-Yano (HKY, Hasegawa et al. 1985) with a
Gamma distribution (+G) based on Akaike information criterion
(AIC) with correction. This model was therefore used for each of
the Cylas COI phylogenetic analyses. The maximum likelihood and
Bayesian trees were not congruent (Supp. Fig. S1 [online only] and
Fig. 2). Maximum likelihood (Supp. Fig. S1 [online only]) performed
poorly when compared with Bayesian inference (Fig. 2), with the
former having lower bootstrap support for most nodes when com-
pared with the higher PP of the latter. Six well-supported (PP > 0.99)
mtDNA clades (A–F) were found in C. formicarius with varying
levels of genetic divergence in the Bayesian tree (Fig. 2). Clades D
and E had poor support (about 50%) in the maximum likelihood
tree, as did clades B and C.The Bayesian tree was used for interpret-
ation because of its better overall support.
Clade A, from India, was the most divergent lineage and none
of the haplotypes from these insects was shared with weevils from
other localities (Fig. 2). Clade B was represented by a single indi-
vidual from Myanmar. The remaining mtDNA clades showed some
geographic partitioning but individuals with haplotypes belonging
to different clades were also found in the same localities. Of the re-
maining individuals, weevils belong to one of two groups, the rst
comprising a single clade and the other a few closely related clades.
The rst group contains weevils from Makira (Solomon Islands), Lae
in Papua New Guinea (L-PNG) and the Western Highlands in PNG
Fig. 1. Haplotype networks for a 371 bp fragment of the COI gene region (top)
for 83 Cylas formicarius individuals and a 669bp fragment of the CAD (bottom)
gene region for 42 C. formicarius individuals. Each individual is colored
according to the location where it was sampled (Tables 1 and 3)and squares
group haplotypes together according to the clades shown in Fig. 2. The size
of each circle represents the number of sequences that were found with each
haplotype but each individual is represented twice in the CAD gene network
because phased haplotypes are used (as estimated by DNAsp, Rozas et al.
2017 ). Black circles represent single mutation steps that separate the haplotypes
that were found. CAD haplotypes are numbered for reference to Table 5.
Journal of Economic Entomology, 2019, Vol. XX, No. XX
Copyedited by: OUP
Downloaded from by University of Queensland user on 16 August 2019
(WH-PNG) which all belong to clade C.The second group combines
weevils from clades D, E, and F from Florida, Taiwan, Japan, and
Australia. Weevils from Fiji and Tonga had mtDNA haplotypes be-
longing to both groups. The geographical distribution of these clades
is mapped by sampled country in Fig. 4.
The CAD gene region had a similar pattern of diversity to that
found for the COI gene region, but a lower overall Hd (because of
the relatively slower mutation rate of nuclear genes when compared
with the COI gene). Individuals of C.formicarius from Tonga had
all four CAD haplotypes, those from WH-PNG had two haplotypes
separated by a single mutation and those from all other locations
(Solomon Islands, Fiji, and Australia) had a single haplotype (Fig.1).
All but ve weevils from Tonga were found to have a single CAD
haplotype. All ve individual weevils with more than one CAD
haplotype were found in Tonga (Table 5).
Differences in the genetic diversity found in C. formicarius com-
pared with E.batatae reect their different evolutionary and geo-
graphical histories in Asia, the Pacic and South America. The
limited distribution and low genetic diversity of E.batatae suggest
that it has invaded the Pacic only recently and that it probably
had a single invasive origin. The presence of multiple genetic lin-
eages within C. formicarius (Fig. 2) indicates that divergent and
geographically discrete lineages of this species were present in Asia,
and probably the Pacic, millions of years before the arrival of the
Fig. 2. Bayesian phylogenetic tree for Cylas formicarius based on a 371bp fragment of the COI gene region and using MrBayes 3.2.6 (Huelsenbeck and Ronquist
20 01). The species C.brunneus is included as an outgroup but is not shown. Node labels are PPs and are shown for nodes with great than 0.70 PP. Major mtDNA
clades that were well supported (PP > 0.95) are labeled Athrough F.Divergence estimates are shown in square brackets in millions of years (My) for major nodes
only. Where individuals associated with a mtDNA haplotype had more than one CAD haplotype (Table 5) the number of individuals is indicated with black dots
(all such individuals were found in Tonga).The location of individuals in each clade can also be determined by referring to this figure.
Fig. 3. Haplotype network for a 640 bp fragment of the COI gene region of
40 Euscepes batatae individuals. Each individual is colored according to the
location where it was sampled (Table 1). The size of each circle represents the
number of individuals that were found with each haplotype.
Journal of Economic Entomology, 2019, Vol. XX, No. XX
Copyedited by: OUP
Downloaded from by University of Queensland user on 16 August 2019
sweet potato (Fig. 2). Some C. formicarius lineages, such as that
from India (Fig.1), may still remain completely isolated from the
others. Clades C and F are estimated to have diverged 3.3 to 6.7 Mya
(Fig. 2) but secondary contact in evident from individuals from Fiji
and Tonga. The contemporary distribution of C.formicarius, then,
has not resulted solely from the recent introduction and movement
of sweet potato through Asia and the Pacic. Recent movement of
sweet potato (Roullier etal. 2013) has undoubtedly contributed to
the observed pattern in both species, but it has affected them dif-
ferently. Further, contrary to previous records which indicate that
Fig. 4. Map of the Asia-Pacific region showing where Cylas formicarius COI clades A–F (Fig. 2) are found. The United States is not shown but all individuals from
this region belong to clade F.Clade labels represent nations rather than specific sample sites. Asimplified version of the phylogenetic tree from Fig. 2 is shown
in the top right to indicate the relationship between the clades.
Table 4. Number of individuals (n), number of haplotypes (n hap.), haplotype diversity (H), Tajima’s D and Fu’s FS for the COI gene region
of Cylas formicarius and Euscepes batatae
Species Locality n n hap. H D FS
C.formicarius Australia 9 3 0.56 −1.72* 2.89
Fiji 12 4 0.56 −1.60 5.21*
Japan 7 4 0.86 0.57 2.88
Solomon Islands 10 2 0.20 −1.11 −0.34
Tonga 27 2 0.36 1.41 20.29**
United States 6 4 0.80 −1.23 −1.81
All 83 20 0.79 0.16 0.71**
E.batatae Tonga 27 2 0.36 0.62 0.98
PNG 6 1 - - -
All 38 3 0.38 −0.21 −0.12
Only populations with six or more individuals were analyzed independent of the ‘All’ dataset. Populations with genetic measures listed as ‘-’ showed no poly-
morphism. Calculated using DNAsp 6.12 (Rozas etal. 2017).
*P<0.05, **P ≤ 0.01.
Journal of Economic Entomology, 2019, Vol. XX, No. XX
Copyedited by: OUP
Downloaded from by University of Queensland user on 16 August 2019
C. formicarius is established in Samoa but that E. batatae is not
(Waterhouse 1997), we only recovered E.batatae there (Table 1). It
is likely that the older records are incorrect, possibly resulting from
misidentications based on the morphologically similar larvae of the
two species.
Cylas formicarius
The presence of divergent lineages of C. formicarius in the Asia-
Pacic indicates that this species was present in the region millions
of years before the arrival of sweet potato, and as such, has not
arrived in the region from a single location. The relatively recent
arrival of sweet potato means that C. formicarius must have per-
sisted on native host plants before this time. Each of the mtDNA
lineages uncovered here (Fig. 2) may have had different host associ-
ations and geographical distributions, and although the sweet potato
is economically the most signicant modern host for these weevils,
these native hosts may be important for understanding the ecology
and impact of these insects as pests within their respective ranges
today. Investigating whether there are meaningful biological differ-
ences across the clades (Fig. 2), whether they may in some cases be
distinct species or subspecies, and what their respective native and
invasive ranges and host plants are is, therefore, a priority, perhaps
mostly because it has implications for regional biosecurity.
Despite the general pattern in C.formicarius being one of geo-
graphically distinct mtDNA clades (or closely related clades being
found together), secondary contact is evident between individuals
from the most divergent non-Indian clades, in Tonga and Fiji, which
are estimated to have diverged between 3.3 and 6.7 Mya (Fig. 2).
Some heterozygous individuals from Tonga have nuclear DNA
haplotypes that are elsewhere associated only with single mtDNA
lineages (Table 5), and so together with the presence of divergent
mtDNA haplotypes, this is evidence for some amount of mating
having occurred across these clades. The genetic diversity of these
Tongan individuals at COI is low (two haplotypes from 27 individ-
uals, Table 4) even when compared with the known invasive popu-
lations from the United States (four haplotypes from six individuals,
Fig. 1) and so secondary contact may have occurred elsewhere before
C.formicarius invaded the region.
The extent to which invasion and endemism have shaped the
genetic structure of specic C.formicarius populations will be made
clearer in the future by using higher resolution molecular markers
alongside more comprehensive sampling of this taxon. This includes
the historical genetic relationship between the different clades un-
covered here (Fig. 2), whether these clades are truly as discrete as
they appear and the extent of admixture following secondary con-
tact. Some of the mtDNA lineages uncovered in C.formicarius (Fig.
2), however, probably represent divergent and independent popula-
tions that were present in Asia long before these insects encountered
sweet potato. Although sample numbers are small in some instances,
the extent of the divergence between the most divergent lineages in
C.formicarius necessitates the investigation of them being distinct
species or subspecies. Indeed, this weevil was formerly considered to
comprise three species or subspecies: C.formicarius, C.turcipennis,
and C.elegantulus. Pierce (1940) used the subspecies designations
C.formicarius formicarius for specimens from India, C.formicarius
elegantulus for those from the New World, Madagascar and the
Pacic and C.formicarius turcipennis for those from Indonesia and
the Philippines (Wolfe 1991).
A comprehensive comparison of the genetic and taxonomic data
is not yet possible, but some relationship between the three main
mtDNA clades discovered here and the historical species designa-
tions within C. formicarius appears likely. Secondary contact and
admixture of the C. formicarius included here have undoubtedly
impacted modern populations, and this needs to be considered
when interpreting historical taxonomic research. Large amounts of
within-population morphological and chromosomal variation have
been observed in C. formicarius by several authors (Pierce 1940,
Hung 1985, Wolfe 1991) and this suggests that mixed lineages may
have been used in previous taxonomic studies. Further sampling for
combined morphological and genetic analyses would help to re-
solve the relationship between historical and modern populations of
C.formicarius and their past species designations. Ipomoea species
native to the Asia-Pacic should also be surveyed, both in and out-
side of sweet potato cropping areas, to better understand the host
associations of this taxon in different localities.
Euscepes batatae
The low genetic diversity found in E.batatae (Fig. 3) is typical of
that found in invasive populations, demonstrating a recent and inva-
sive origin for this species. Sweet potato and E.batatae come from
the same broad region (tropical America and the Caribbean; Roullier
etal. 2013) and so invasion of the Pacic by E.batatae probably
occurred alongside one or more introductions of sweet potato. The
invasion of E.batatae into the Pacic is recent but not modern; it
was already present in Tahiti by 1849 (Fairmaire 1849, Zimmerman
1936), and in Hawaii by at least 1885 (Blackburn and Sharp 1885,
Zimmerman 1936). Euscepes batatae is not found through most of
Asia (only Japan) which suggests that it invaded the Pacic directly
from the Americas. Further, the dispersal capabilities of this weevil
are reportedly poor (Sherman and Tamashiro 1954, Alleyne 1982,
Moriya and Miyatake 2001), and this is one reason why E.batatae
was thought to have a more limited distribution than C.formicarius.
The results from both species indicate that the larger distribution of
C.formicarius may largely be a consequence of the species having
been more widespread before sweet potato arrived in the region.
Future invasions of E.batatae to new regions in the Asia-Pacic may,
therefore, be more likely than is presently assumed.
The contrasting genetic relationships across C. formicarius and
E. batatae populations has given insight into their evolutionary and
invasive histories, and may also have consequences for the manage-
ment of these pests. The results of this research are preliminary, given
the unexpected genetic diversity uncovered in C.formicarius, but they
provide direction for future research. Uncovering the host plants that
C.formicarius relied on before sweet potato arrived in Asia and the
Pacic, and perhaps upon which they still rely today, will have implica-
tions for understanding their local ecology and consequently their man-
agement. Extensive sampling across the entire range of C.formicarius,
and the native range of E.batatae is now necessary. Further invasions
Table 5. All individuals of Cylas formicarius that were sequenced
at the CAD gene region and which had two CAD haplotypes
Site code Location COI clade CAD haplotypes
TO2cy Lifuka, Tonga F1, 3
TO2cy Lifuka, Tonga F2, 3
TO2cy Lifuka, Tonga C3, 4
TO1cy Foa, Tonga F1, 2
TO1cy Foa, Tonga F1, 2
Sample site code (Table 1), COI clade (Fig. 2), and CAD haplotype number
(Fig. 1) are listed.
Journal of Economic Entomology, 2019, Vol. XX, No. XX
Copyedited by: OUP
Downloaded from by University of Queensland user on 16 August 2019
of E. batatae and secondary invasions of C. formicarius lineages
are likely and may well carry quarantine implications. Investigating
the recent invasion history of each species may help to predict po-
tential future invasion pathways for E. batatae and the lineages of
C.formicarius. Whether any of the divergent evolutionary lineages re-
corded in C.formicarius represent species or subspecies, and whether
there are biological differences between them, remains to be tested.
The genetic diversity uncovered in this study suggests strongly that the
species diversity of C.formicarius requires renewed investigation. Any
such investigation should consider the complex biogeographical and
evolutionary history of this pest, the historical connections among its
populations, and the genetic consequences of recent invasions.
Supplementary data are available at Journal of Economic
Entomology online.
We thank Sione Foliaki (Tonga), Nitesh Nand (Fiji), John Fasi, Pita
Tikai and Maria Gharuka (Solomon Islands), Aleni Uelese (Samoa),
and Thecla Guaf (Papua New Guinea) for help in collecting wee-
vils from the South Pacic, and Dakshina R.Seal and Catherine
Sabines for providing C.formicarius samples from Florida, United
States. We thank Tim Vance for technical assistance. We also thank
four anonymous reviewers for their helpful comments. This work
was funded by the Australian Centre for International Agricultural
Research (ACIAR) project HORT/2010/065.
Alleyne,E.H. 1982. Studies on the biology and behavior of the West Indian
sweet potato weevil, Euscepes postfasciatus (Fairmaire)(Coleoptera:
Curculionidae). In 18th Annual Meeting, 22–28 August 1982, Dover,
Barbados. Caribbean Food Crops Society.
Barrau,J. 1957. L’énigme de la patate douce en Océanie. Études d’Outre-Mer
40: 83–87.
Billah,M.K., A.Agbessenou, D.D.Wilson, W.Dekoninck, and C.Vangestel.
2017. Morphometric studies of the sweet potato weevil, Cylas species-
complex in southern Ghana. Science and Development 1: 2–18.
Blackburn,T., and D.Sharp. 1885. Memoirs of the Coleoptera of the Hawaiian
Islands. Scientic Transactions of the Royal Dublin Society 2: 253.
Bouckaert, R., J. Heled, D. Kühnert, T. Vaughan, C. H. Wu, D. Xie,
M. A. Suchard, A.Rambaut, and A. J. Drummond. 2014. BEAST 2: a
software platform for Bayesian evolutionary analysis. Plos Comput. Biol.
10: e1003537.
Brower,A. V. 1994. Rapid morphological radiation and convergence among
races of the buttery Heliconius erato inferred from patterns of mitochon-
drial DNA evolution. Proc. Natl. Acad. Sci. U.S.A. 91: 6491–6495.
Chalfant,R.B., R.K., Jansson, D.K.Deal, and J.M.Schalk 1990. Ecology and
management of sweet potato insects. Annu. Rev. Entomol. 35: 157–180.
Clement,M., D.Posada, and K.A.Crandall. 2000. TCS: a computer program
to estimate gene genealogies. Mol. Ecol. 9: 1657–1659.
Darriba,D., G.L. Taboada, R.Doallo, and D.Posada. 2012. jModelTest 2:
more models, new heuristics and parallel computing. Nat. Methods 9: 772.
Denham,T. 2013. Ancient and historic dispersals of sweet potato in Oceania.
Proc. Natl. Acad. Sci. U.S. A. 110: 1982–1983.
Fairmaire, L. 1849. Essai sur les Coléoptères de la Polynésie. Suite des
Curculionites. Rev. Mag. Zool. 2: 550–559.
Folmer,O., M.Black, W.Hoeh, R.Lutz, and R.Vrijenhoek. 1994. DNA primers
for amplication of mitochondrial cytochrome c oxidase subunit Ifrom di-
verse metazoan invertebrates. Mol. March Biol. Biotechnol. 3: 294–299.
Foottit,R.G., H.E.Maw, C.D.VONDohlen, and P.D.Hebert. 2008. Species
identication of aphids (Insecta: Hemiptera: Aphididae) through DNA
barcodes. Mol. Ecol. Resour. 8: 1189–1201.
Guindon, S., J. F. Dufayard, V. Lefort, M. Anisimova, W. Hordijk, and
O.Gascuel. 2010. New algorithms and methods to estimate maximum-
likelihood phylogenies: assessing the performance of PhyML 3.0. Syst.
Biol. 59: 307–321.
Hasegawa,M., H.Kishino, and T.Yano. 1985. Dating of the human-ape split-
ting by a molecular clock of mitochondrial DNA. J. Mol. Evol. 22: 160–174.
Huelsenbeck,J.P., and F.Ronquist. 2001. MRBAYES: Bayesian inference of
phylogenetic trees. Bioinformatics. 17: 754–755.
Hung, A. C. F. 1985. Chromosomal-polymorphism in sweet-potato weevil,
Cylas formicarius. Cytologia 50: 769–772.
Iese,V., E.Holland, M.Wairiu, R.Havea, S.Patolo, M.Nishi, T.Hoponoa,
R.M.Bourke, A.Dean, and L.Waqainabete. 2018. Facing food security
risks: The rise and rise of the sweet potato in the Pacic Islands. Glob.
Food Secur.-Agric.Policy 18: 48–56.
Jackson,D.M., H.F.Harrison, Jr, and J.R.Ryan-Bohac. 2012. Insect resist-
ance in sweetpotato plant introduction accessions. J. Econ. Entomol. 105:
Katoh,K., and D.M.Standley. 2013. MAFFT multiple sequence alignment
software version 7: improvements in performance and usability. Mol. Biol.
Evol. 30: 772–780.
Katoh,K., K.Misawa, K.I. Kuma, and M.Takashi. 2002. MAFFT: a novel
method for rapid multiple sequence alignment based on fast Fourier trans-
form. Nucleic Acids Res. 30: 3059–3066.
Leigh,J.W., and D.Bryant. 2015. popart: fullfeature software for haplotype
network construction. Methods Ecol. Evol. 6: 1110–1116.
Low,J., A.Ball, S. Magezi, J.Njoku, R.Mwanga, M.Andrade, K.Tomlins,
R.Dove, and T.vanMourik. 2017. Sweet potato development and delivery
in sub-Saharan Africa. Afr. J.Food Agric. Nutr. Dev. 17: 11955–11972.
McGregor,A., M.Taylor, R.M.Bourke, and V.Lebot. 2016. Vulnerability of
staple food crops to climate change. Pacic Community (SPC), Noumea,
New Caledonia.
McKenna, D. D., A. S. Sequeira, A. E. Marvaldi, and B. D. Farrell. 2009.
Temporal lags and overlap in the diversication of weevils and owering
plants. Proc. Natl. Acad. Sci. U.S. A. 106: 7083–7088.
Moriya,S., and T.Miyatake. 2001. Eradication programs of two sweetpotato
pests, Cylas formicarius and Euscepes postfasciatus, in Japan with special
reference to their dispersal ability. Japan Agricultural Research Quarterly:
JARQ 35: 227–234.
Muñoz-Rodríguez, P., T. Carruthers, J. R. I. Wood, B. R. M. Williams,
K. Weitemier, B. Kronmiller, D. Ellis, N. L. Anglin, L. Longway,
S. A.Harris, et al. 2018a. Reconciling conicting phylogenies in the
origin of sweet potato and dispersal to Polynesia. Curr. Biol. 28: 1246–
Munoz-Rodriguez,T. Carruthers, and R.W. P.Scotland. 2018b. Temporal dy-
namics of the origin and domestication of sweet potato and implications
for dispersal to Polynesia. bioRxiv: 309799.
Papadopoulou,A., I.Anastasiou, and A.P.Vogler. 2010. Revisiting the insect
mitochondrial molecular clock: the mid-Aegean trench calibration. Mol.
Biol. Evol. 27: 1659–1672.
Pierce, D. W. 1940. Studies of the sweet potato weevils of the subfamily
Cyladinae. Bull. South Calif. Acad. Sci. 39: 205–223.
Ridley, A. W., J.P. Hereward, G.J.Daglish, S.Raghu, G.A. McCulloch,
and G. H. Walter. 2016. Flight of Rhyzopertha dominica (Coleoptera:
Bostrichidae)-a spatio-temporal analysis with pheromone trapping and
population genetics. J. Econ. Entomol. 109: 2561–2571.
Roullier,C., L.Benoit, D.B.McKey, and V.Lebot. 2013. Historical collec-
tions reveal patterns of diffusion of sweet potato in Oceania obscured by
modern plant movements and recombination. Proc. Natl. Acad. Sci. U.S.
A. 110: 2205–2210.
Rozas,J., A.Ferrer-Mata, J.C.Sánchez-DelBarrio, S.Guirao-Rico, P.Librado,
S. E. Ramos-Onsins, and A.Sánchez-Gracia. 2017. DnaSP 6: DNA se-
quence polymorphism analysis of large data sets. Mol. Biol. Evol. 34:
Journal of Economic Entomology, 2019, Vol. XX, No. XX
Copyedited by: OUP
Downloaded from by University of Queensland user on 16 August 2019
Sherman,M., and M.Tamashiro. 1954. The sweetpotato weevils in Hawaii:
their biology and control, vol. 23. University of Hawaii, Honolulu,
Simon, C., F. Frati, A.Beckenbach, B.Crespi, H. Liu, and P.Flook. 1994.
Evolution, weighting, and phylogenetic utility of mitochondrial gene
sequences and a compilation of conserved polymerase chain reaction pri-
mers. Ann. Entomol. Soc. Am. 87: 651–701.
Talekar,N. S. 1995. Characteristics of infestation of sweet potato by sweet
potato weevil Cylas formicarius (Coleoptera: Apionidae), Int. J. Pest
Manage. 41: 238–242.
Wadl,P.A., L.H. Williams, K.R.Harris-Shultz, and G.T.McQuate. 2019.
Method for DNA isolation from sweetpotato weevil (Coleoptera:
Curculionidae) collected in pheromone-baited traps. J. Econ. Entomol.
112: 1001–1003.
Waterhouse,D.F. 1997. The major invertebrate pests and weeds of agricul-
ture and plantation forestry in the southern and western Pacic. AClAR,
Canberra, Australia. Monograph No. 44. 99 p.
Wild,A. L., and D. R. Maddison. 2008. Evaluating nuclear protein-coding
genes for phylogenetic utility in beetles. Mol. Phylogenet. Evol. 48:
Wolfe,G.W. 1991. The origin and dispersal of the pest species of Cylas with a
key to the pest species groups of the world. Westview Press, Boulder, CO.
Yasuda,K. 1997a. Occurrence of West Indian sweet potato weevil, Euscepes
postifasciatus (Fairmaire) (Coleoptera:Curculionidae) and damage to sweet-
potato (Ipomoea batatas (L) Lam) elds. Jpn. J.Appl. Entomol. Z. 41: 83–88.
Yasuda,K. 1997b. Control threshold on sweet potato, Ipomoea batatas L.,
damaged by the West Indian Sweet Potato Weevil, Euscepes postfasciatus
(Fairmaire) (Coleoptera: Curculionidae). Jpn. J.Appl. Entomol. Z. 41:
Yen,D. E. 1974. The sweet potato and Oceania: an essay in ethnobotany.
Bishop Museum Press, Honolulu, HI.
Zhang,D.X., and G.M.Hewitt. 1996. Nuclear integrations: challenges for
mitochondrial DNA markers. Trends Ecol. Evol. 11: 247–251.
Zimmerman,E.C. 1936. Cryptorhynchinae of the society islands. Occasional
Papers of the Bishop Museum. 12: 3–48.
Journal of Economic Entomology, 2019, Vol. XX, No. XX
Copyedited by: OUP
Downloaded from by University of Queensland user on 16 August 2019
... In recent years, the utility of sequences of mitochondrial DNA (mtDNA) to infer origin of invasive populations of some insects were reported (Corin et al. 2007;Downie 2002;Gariepy et al. 2014). Regarding C. formicarius and E. postfasciatus, Brookes et al. (2019) analyzed DNA barcode region, which was a part of cytochrome c oxidase subunit I (COI) gene of mtDNA, for populations in the Asia-Pacific region. Likewise, Sukenari et al. (2009) Descriptions of the primers used for PCR amplification and/or sequencing of Cylas formicarius and Euscepes postfasciatus a Single asterisk refers to the inner primers for sequencing. ...
... Recently, Brookes et al. (2019) showed that 83 individuals of C. formicarius from ten countries had 20 haplotypes, which were divided into six clades in the phylogenetic tree of the DNA barcode region of mtDNA. The results in the present study supported their findings and also revealed for the first time that haplotypes from Minamitorishima Island and the Philippine Islands formed a distinct clade (Clade F) in the ML tree (Fig. 1). ...
... Although the analyzed gene regions were different, we obtained same results after phylogenetic analysis of the DNA barcode region (Fig. 1). Brookes et al. (2019) showed that 40 individuals of E. postfasciatus collected from four countries had three haplotypes in the DNA barcode region. The two haplotypes detected from Japan were identical with two of those haplotypes, i.e., Hap 1 detected from the Nansei Islands, Fiji, Samoa, and Tonga, and Hap 2 detected from the Nansei Islands, and Samoa (Fig. 4, Supplementary Tables 2, 4). ...
Full-text available
We examined the genetic variation in the sequences of cytochrome c oxidase subunit I gene (DNA barcode region) and some other regions of mitochondrial DNA of sweet potato weevil, Cylas formicarius (Fabricius, 1798) (Coleoptera: Brentidae), and West Indian sweet potato weevil, Euscepes postfasciatus (Fairmaire, 1849) (Coleoptera: Curculionidae), in Japan. In the DNA barcode region of C. formicarius, 139 haplotypes were detected from 1705 individuals belonging to 46 geographical populations. In the maximum likelihood phylogenetic tree, haplotypes found in Japan were mainly divided into three clades. In the DNA barcode region of E. postfasciatus, two haplotypes were detected from 82 individuals belonging to eight geographical populations. Of those haplotypes, Hap 1 was detected from the Nansei Islands, Fiji, Samoa, and Tonga, while Hap 2 was detected from the Nansei Islands and Samoa. These results suggest that the Japanese populations of both the species were derived from several foreign countries. Based on the haplotype network analyses of some other gene regions, those regions may be useful for a more detailed estimation of the origin of an accidentally collected individual in non-distribution area in Japan.
... M. volkensii is a suitable dryland agroforestry tree and is a source of highly praised mahogany timber and termite-resistant poles [6]. M. volkensii seed kernel extracts have shown antifeedant and growth inhibition activity against several insect pests [7][8][9][10]. Insect antifeedant compounds such as salannin, volkensin, 1-tigloyl-trichilinin, 1-cinnamoyltrichilinin and 1-acetyltrichilinin have also been isolated from M. volkensii fruits [7,11]. ...
... Smith), are major insect pests of economic importance for sweet potato and maize farming in Africa, respectively. Significant sweet potato damage and economic losses are caused by C. puncticollis which oviposit in the vines and tubers [10]. These weevils cause up to 100% yield loss [12][13][14], and the damage from weevils continues to increase during storage [15]. ...
Full-text available
The African sweet potato weevil, Cylas puncticollis, and fall armyworm, Spodoptera frugiperda, are insect pests of economic importance that have a negative impact on sweet potato and maize production, respectively. In this study, we aimed to evaluate the potential of Melia volkensii extracts to protect sweet potato and maize plants against damage by both insect pests. We evaluated extracts from the bark, leaves, pulp and nuts of Melia volkensii for antifeedant activity against C. puncticollis and S. exigua (used as a substitute for S. frugiperda), under laboratory conditions. Interestingly, extracts of all plant parts showed antifeedant activity. These results led us to investigate the effectiveness of nut and pulp extracts to protect sweet potato and maize crops in greenhouse conditions. Against C. puncticollis, the sweet potato plants treated with nut extracts showed the lowest tuber damage (18%) when compared to pulp extracts (30%), positive control (33%) and negative control (76%). Nut extracts, pulp extracts and positive control reduced maize leaf and whorl damage by S. frugiperda compared to the negative control. Altogether, this study highlights the potential of M. volkensii extracts and their application in integrated insect pest management Keywords: botanical pesticides; crude extracts; integrated pest management; limonoids; maize leaf damage; sweet potato tuber damage
To conduct population genetics analyses on sweetpotato weevils (Cylas formicarius elegantulus) collected from three populations in the United States, microsatellite loci were developed from sweetpotato weevil transcriptome sequences obtained from a publicly available database. Nineteen of 27 microsatellite loci tested were usable for population analyses. Sweetpotato weevil individuals from Georgia (N = 17), Hawaii (N = 16) and South Carolina (N = 12) were analysed. Here, we present microsatellite primer sequences for 19 loci and population genetics statistics, including diversity, population differences and relatedness. Thirty‐nine alleles were detected in the 45 individuals, and four private alleles were observed in individuals from the Hawaii and South Carolina populations. Observed heterozygosities ranged from 0.00 to 1.00, and expected heterozygosities ranged from 0.00 to 0.44. The mean Shannon's information index ranged from 0.30 to 0.49 for the three populations. Pairwise differences among populations (FST estimates) from individuals ranged from a low of 0.022 between Georgia and Hawaii to a high of 0.036 between Georgia and South Carolina, and cluster analysis (PCA and Structure) indicated two populations independent of geographical location.
Full-text available
This study provides a protocol for the isolation of high-quality DNA from sweetpotato weevils (Cylas formicarius elegantulus (Summers)) collected from pheromone-baited aerial funnel traps. This study was based on our discovery that a 2-wk collection interval of sweetpotato weevils from pheromone traps did not permit isolation of intact high-quality genomic DNA. To test the effect of collection methods, i.e., sample collection interval and preservation method, on quality of isolated DNA, we placed freshly killed male sweetpotato weevils into aerial funnel traps in the field and removed subsamples at several times thereafter. DNA yield from freshly isolated (day = 0) samples was significantly greater than samples preserved in 70% ethanol or at -20°C, whereas there was no difference between 70% ethanol and -20°C storage. Likewise, DNA yield from freshly isolated (day = 0) samples was significantly greater than for later sampling times. Quality assessment of genomic DNA through gel electrophoresis and polymerase chain reaction (PCR) indicated isolation of high molecular weight DNA for all samples collected at t ≤ 7 d, but that DNA quality was degraded by 14 d. Our goal was to develop a reliable method for isolation of genomic DNA from field-collected sweetpotato weevil suitable for direct use in PCR. We discovered that it is critical to collect specimens from traps at an interval of 1 wk or less. Our findings allow for scheduling of sampling at reasonable intervals without the need for special materials. This has the added benefit of allowing individuals without special training to collect and prepare sweetpotato weevil specimens for genetic studies. Published by Oxford University Press on behalf of Entomological Society of America 2018.
Full-text available
Pacific Island communities are highly exposed to a range of hazards including extreme weather events and outbreaks of pests and diseases. These hazards can cause severe losses to yields of traditional food crops and increase the risks of famine and food insecurity in Pacific Island communities. Historically, the cultivation of sweet potato enabled communities to adjust their farming systems and reduce food security risks before, during and after disasters. The food security features of sweet potato contributed to its adoption as a staple crop by communities at the "edge" of agro-ecological limits for their traditional crops. Sweet potato was also adopted as a supplementary crop, adding nutrition and stability to communities' food systems. In present times, sweet potato is being cultivated as part of food security and climate change adaptation projects in Pacific communities. This has been facilitated by regional mechanisms for sustainable use of plant genetic resources. But as climate change continues to intensify extreme events and cause sea levels to rise, the resilience of current sweet potato varieties is not guaranteed. Sweet potato, like other Pacific staple crops, is not 100% "disaster-proof". There is a need for multi-partner, proactive agro-ecological based research on sweet potato and other staple crops to reduce both short-term and long-term food security risks faced by Pacific Island communities.
Full-text available
The Sweet potato weevil, Cylas species, is a key pest of sweet potato, and widely distributed on the African continent. The management of the pest is limited because its taxonomic status is not clear. Populations of the same species occupying distinct localities experience different ecological and climatic pressures that might result in differentiation in traits. This study sought to identify and compare body sizes of Cylas species from four regions in southern Ghana-Central, Eastern, Greater Accra and Volta. Of the 6,686 samples collected from the four regions, two species were identified: Cylas brunneus Fabricius, and Cylas puncticollis Boheman. Twelve morphometric characters were examined and measured, of which four traits-elytra and rostrum lengths, pronotum and head widths contributed most to the variations observed. In C. puncticollis, individuals with the longest body were recorded in Greater Accra Region (7.084 ± 0.089 mm), while those in the Central Region had the smallest body size (6.786 ± 0.086 mm). Our findings suggest that distinct localities may influence changes in body size.
Full-text available
We present version 6 of the DnaSP (DNA Sequence Polymorphism) software, a new version of the popular tool for performing exhaustive population genetic analyses on multiple sequence alignments. This major upgrade incorporates novel functionalities to analyse large datasets, such as those generated by high-throughput sequencing (HTS) technologies. Among other features, DnaSP 6 implements: i) modules for reading and analysing data from genomic partitioning methods, such as RADseq or hybrid enrichment approaches, ii) faster methods scalable for HTS data, and iii) summary statistics for the analysis of multi-locus population genetics data. Furthermore, DnaSP 6 includes novel modules to perform single- and multi-locus coalescent simulations under a wide range of demographic scenarios. The DnaSP 6 program, with extensive documentation, is freely available at
Full-text available
The flight of the lesser grain borer, Rhyzopertha dominica (F.), near grain storages and at distances from them, was investigated to assess the potential of these beetles to infest grain and spread insecticide resistance genes. We caught R. dominica in pheromone-baited flight traps (and blank controls) set at storages, in fields away from storages, and in native vegetation across a 12-mo period. A functional set of highly polymorphic microsat-ellite markers was developed, enabling population genetic analyses on the trapped beetles. Pheromone-baited traps caught just as many R. dominica adults at least 1 km from grain storages as were caught adjacent to grain storages. Samples of beetles caught were genetically homogeneous across the study area (over 7,000 km 2) in South Queensland, Australia. However, a change in genetic structure was detected at one bulk storage site. Subsequent analysis detected a heterozygous excess, which indicated a population bottleneck. Only a few beetles were caught during the winter months of June and July. To assess the mating status and potential fecundity of dispersing R. dominica females, we captured beetles as they left grain storages and quantified offspring production and life span in the laboratory. Nearly all (95%) of these dispersing females had mated and these produced an average of 242 offspring. We demonstrated that R. dominica populations in the study area display a high degree of connectivity and this is a result of the active dispersal of mated individuals of high potential fecundity.
Full-text available
Eradication programs are being implemented for 2 sweetpotato weevil pests, Cylas formicarius and Euscepes postfasciatus, in the Ryukyu Islands located in the southernmost part of Japan by the application of the sterile insect technique (SIT). As it is essential for the implementation of the programs to assess the dispersal ability of the weevils, recent studies were reviewed. Both the flight and walking ability of C. formicarius was much higher in males than in females when determined in the laboratory. Synthesized sex pheromone of C. formicarius has been used as a strong lure to capture the males in the field. Results of the mark-recapture experiments suggest that a distance of at least 2 km is needed for the range of the buffer zone to separate the SIT target area from others. Since E. postfasciatus is unable to fly, only the walking ability was evaluated in the laboratory, indicating that the females exhibited a fairly higher locomotion activity compared with the males. Since no effective attractant for E. postfasciatus has been identified yet, information on the dispersal activity in the field is limited. Fundamental studies should be conducted to determine the dispersal ability of both weevil pests to implement successfully the eradication programs.
The sweet potato is one of the world's most widely consumed crops, yet its evolutionary history is poorly understood. In this paper, we present a comprehensive phylogenetic study of all species closely related to the sweet potato and address several questions pertaining to the sweet potato that remained unanswered. Our research combined genome skimming and target DNA capture to sequence whole chloroplasts and 605 single-copy nuclear regions from 199 specimens representing the sweet potato and all of its crop wild relatives (CWRs). We present strongly supported nuclear and chloroplast phylogenies demonstrating that the sweet potato had an autopolyploid origin and that Ipomoea trifida is its closest relative, confirming that no other extant species were involved in its origin. Phylogenetic analysis of nuclear and chloroplast genomes shows conflicting topologies regarding the monophyly of the sweet potato. The process of chloroplast capture explains these conflicting patterns, showing that I. trifida had a dual role in the origin of the sweet potato, first as its progenitor and second as the species with which the sweet potato introgressed so one of its lineages could capture an I. trifida chloroplast. In addition, we provide evidence that the sweet potato was present in Polynesia in pre-human times. This, together with several other examples of long-distance dispersal in Ipomoea, negates the need to invoke ancient human-mediated transport as an explanation for its presence in Polynesia. These results have important implications for understanding the origin and evolution of a major global food crop and question the existence of pre-Columbian contacts between Polynesia and the American continent.
In sub-Saharan Africa, more than 40% of children under five years of age suffer from vitamin A deficiency. Among several interventions in place to address vitamin A deficiency is biofortification, breeding vitamin A into key staple crops. Staple crops biofortified with beta-carotene, the precursor to vitamin A, are orange in color. Given the natural occurrence of high levels of beta-carotene in many sweet potato varieties, breeding progress for biofortified orange sweet potato (OSP) has been much faster than for the other vitamin A enhanced staples. Nearly 3 million households have been reached with OSP. This paper reviews key factors influencing the uptake of OSP, the breeding investment, five key delivery approaches that have been tested in the region and efforts to broaden government and other stakeholder engagement.
Haplotype networks are an intuitive method for visualising relationships between individual genotypes at the population level. Here, we present popart, an integrated software package that provides a comprehensive implementation of haplotype network methods, phylogeographic visualisation tools and standard statistical tests, together with publication-ready figure production. popart also provides a platform for the implementation and distribution of new network-based methods – we describe one such new method, integer neighbour-joining. The software is open source and freely available for all major operating systems.