Transovarial Transmission of Rickettsia
spp. and Organ-Specific Infection of the
Whitefly Bemisia tabaci
Marina Brumin, Maggie Levy and Murad Ghanim
Appl. Environ. Microbiol. 2012, 78(16):5565. DOI:
10.1128/AEM.01184-12.
Published Ahead of Print 1 June 2012.
These include:
REFERENCES
CONTENT ALERTS
This article cites 66 articles, 24 of which can be accessed free
at: http://aem.asm.org/content/78/16/5565#ref-list-1
Receive: RSS Feeds, eTOCs, free email alerts (when new
articles cite this article), more»
Information about commercial reprint orders: http://journals.asm.org/site/misc/reprints.xhtml
To subscribe to to another ASM Journal go to: http://journals.asm.org/site/subscriptions/
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
Updated information and services can be found at:
http://aem.asm.org/content/78/16/5565
Transovarial Transmission of Rickettsia spp. and Organ-Specific
Infection of the Whitefly Bemisia tabaci
Marina Brumin,a Maggie Levy,b and Murad Ghanima
Department of Entomology, the Volcani Center, Bet Dagan, Israel,a and Department of Plant Pathology and Microbiology, Robert H. Smith Faculty of Agriculture, Hebrew
University of Jerusalem, Rehovot, Israelb
T
he intimate interactions between endosymbionts and their arthropod hosts are well established for obligate primary symbionts such as Buchnera in aphids and Carsonella in psyllids (3, 28).
These bacteria are housed within specialized cells termed bacteriocytes and aggregate together to form an organ termed the bacteriome (3, 12, 56). Primary endosymbionts provide the host with
essential amino acids to complete their diet and are therefore necessary for host survival and development (12). In contrast, secondary symbionts are not essential, and until recently, their effect
on their host’s biology was underestimated. In the past few years,
however, their relevance has become a matter of interest, and accumulating data suggest that they can play a significant role in the
biology of their hosts (56). To date, secondary symbionts have
been reported to be involved in reproductive manipulations, host
plant utilization, and the ability to cope with environmental factors such as response to heat stress and chemical insecticides (8,
11, 43, 46, 49, 57, 58). The reported localization patterns of secondary symbionts are diverse and vary within the host body. The
bacteria can be either diffusely distributed throughout the host
body or restricted to specific tissues. For example, secondary symbionts have been reported from the hemolymph of many insect
taxa (15, 26, 34, 67), primary bacteriocytes in whiteflies (39, 63),
secondary bacteriocytes and sheath cells mainly in aphids (34, 55,
61, 67), salivary glands (54, 59, 60), Malpighian tubules (13), and
reproductive organs (26, 32, 41, 60, 68).
Like other phloem feeders, the sweet potato whitefly, Bemisia
tabaci (Gennadius) (Hemiptera: Aleyrodidae), harbors a diverse
array of endosymbionts, including the primary endosymbiont
Portiera aleyrodidarum (3), and several other facultative secondary symbionts, including Rickettsia, Hamiltonella, Wolbachia, Arsenophonus, Cardinium, and Fritschea (3, 38). B. tabaci is a complex of several cryptic species, also termed biotypes, which differ
genetically and biologically. The most widespread and damaging
August 2012 Volume 78 Number 16
biotypes are B and Q (recently termed Middle East Asia Minor 1
[MEAM1] and Mediterranean [MED], respectively) (24). The B
biotype from Israel has been associated with Hamiltonella, while
the Q biotype is associated with Wolbachia and Arsenophonus.
Both biotypes harbor Rickettsia, with infection rates ranging from
22 to 100% (17). Neither biotype harbors Cardinium or Fritschea
(17). A molecular phylogenetic analysis based on 16S rRNA gene
and gltA sequences revealed that the Rickettsia sp. from B. tabaci
belongs to the group consisting of Rickettsia bellii from ticks and
Rickettsia spp. from herbivorous arthropods such as aphid and
leafhopper. This Rickettsia shares 99 and 97% similarity with the
R. bellii 16S rRNA and gltA sequences, respectively (38). Rickettsia
symbionts have been further identified from a wide variety of
invertebrates, such as aphids (15), leafhoppers (22), lady bird beetles (69), bruchid beetles (33), leeches (47), and ticks (6).
All secondary symbionts in B. tabaci colocalize with the primary symbiont inside the bacteriocytes, ensuring their vertical
transmission. However, only Rickettsia localizes outside the bacteriocytes and appears in most of the body cavity except the bacteriocytes (39). In only one case, Rickettsia was described to be
colocalizing with the primary symbiont inside the bacteriocyte
(14, 39). It is still unclear how Rickettsia that localizes outside the
bacteriosome is transferred and spread in B. tabaci populations,
especially how it can reach high prevalence and near fixation in
Received 12 April 2012 Accepted 22 May 2012
Published ahead of print 1 June 2012
Address correspondence to Murad Ghanim, [email protected].
This article is contribution 501/12 from ARO publications.
Copyright © 2012, American Society for Microbiology. All Rights Reserved.
doi:10.1128/AEM.01184-12
Applied and Environmental Microbiology
p. 5565–5574
aem.asm.org
5565
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
The whitefly Bemisia tabaci is a cosmopolitan insect pest that harbors Portiera aleyrodidarum, the primary obligatory symbiotic
bacterium, and several facultative secondary symbionts. Secondary symbionts in B. tabaci are generally associated with the bacteriome, ensuring their vertical transmission; however, Rickettsia is an exception and occupies most of the body cavity, except
the bacteriome. The mode of Rickettsia transfer between generations and its subcellular localization in insect organs have not
been investigated. Using electron and fluorescence microscopy, we show that Rickettsia infects the digestive, salivary, and reproductive organs of the insect; however, it was not observed in the bacteriome. Rickettsia invades the oocytes during early developmental stages and resides in follicular cells and cytoplasm; it is mostly excluded when the egg matures; however, some bacterial
cells remain in the egg, ensuring their transfer to subsequent generations. Rickettsia was localized to testicles and the spermatheca, suggesting a horizontal transfer between males and females during mating. The bacterium was further observed at
large amounts in midgut cells, concentrating in vacuole-like structures, and was located in the hemolymph, specifically at exceptionally large amounts around bacteriocytes and in fat bodies. Organs further infected by Rickettsia included the primary salivary glands and stylets, sites of possible secretion of the bacterium outside the whitefly body. The close association between Rickettsia and the B. tabaci digestive system might be important for digestive purposes. The vertical transmission of Rickettsia to
subsequent generations occurs via the oocyte and not, like other secondary symbionts, the bacteriome.
Brumin et al.
MATERIALS AND METHODS
Insects and rearing conditions. Rickettsia-free and Rickettsia-containing
B-biotype B. tabaci populations were reared on cotton seedlings (Gossypium hirsutum L. cv. Acala) maintained inside insect-proof cages and
growth rooms under standard conditions of 25 ⫾ 2°C, 60% relative humidity, and a 14-h light/10-h dark photoperiod. Identification of the B
biotype was based on microsatellite markers (23). The two strains of B.
tabaci biotype B used in this study were established by selection of isofemale strains as previously described (11). The original B biotype population used to establish the isofemale strains was collected from Ayalon
Valley in Israel (31°52=3⬙N, 34°57=34⬙E) in 1996 and cultured under laboratory conditions. This population tested positive for Portiera and Hamiltonella.
DNA extraction and PCR amplification. To confirm the presence of
Rickettsia in each adult B. tabaci individual, 20 whiteflies were individually
homogenized in lysis buffer as previously described (31). The lysate was
then incubated at 65°C for 15 min, followed by incubation at 95°C for 10
min. The samples were tested for the presence of Rickettsia by PCR using
Rickettsia-specific primers for amplification of the 16S rRNA gene fragment: primer Rb-F (5=-GCTCAGAACGAACGCTATC-3=) and primer
Rb-R (5=-GAAGGAAAGCATCTCTGC-3=) (38). The reaction was carried out in a 20-␮l volume containing 2 ␮l template DNA lysate, 20 pmol
of each primer, 10 mM deoxynucleoside triphosphates, 1⫻ DreamTaq
buffer, and 1 U DreamTaq DNA polymerase (Fermentas). PCR-amplified
products were visualized on a 1% agarose gel containing ethidium bromide. To detect Rickettsia in specific organs using PCR, B. tabaci midguts,
5566
aem.asm.org
stylets, primary salivary glands, and hemolymph were dissected under a
binocular in 1⫻ phosphate-buffered saline (PBS), washed, and subjected
to the above-described lysis-PCR amplification. The reaction mixture
contained 10 ␮l of template DNA lysate.
FISH. Fluorescent in situ hybridization (FISH) was performed as previously described (38). Briefly, specimens were fixed overnight in Carnoy’s fixative (chloroform-ethanol-glacial acetic acid, 6:3:1, vol/vol) and
then decolorized in 6% H2O2 in ethanol for 2 h and hybridized overnight
in hybridization buffer (20 mM Tris-HCl, pH 8.0, 0.9 M NaCl, 0.01%
[wt/vol] sodium dodecyl sulfate, 30% [vol/vol] formamide) containing 10
pmol fluorescent probes per ml. Dissected midguts, salivary glands, and
ovaries were fixed for 5 min in Carnoy’s fixative and hybridized overnight.
For specific targeting of Portiera and Rickettsia, probes BTP1-Cy3 (5=-C
y3-TGTCAGTGTCAGCCCAGAAG-3=) and Rb1-Cy5 (5=-Cy5-TCCAC
GTCGCCGTCTTGC-3=), respectively, were used. Nuclei were stained
with 4=,6=-diamidino-2-phenylindole (DAPI; 0.1 mg ml⫺1). The stained
samples were mounted whole in hybridization buffer and viewed under
an IX81 Olympus FluoView500 confocal microscope. For each stage, at
least 15 specimens were viewed under the microscope to confirm reproducibility. Optical sections (0.7 to 1.0 ␮m thick) were prepared from each
specimen. Specificity of detection was confirmed using no-probe and
Rickettsia-free whitefly controls.
Electron microscopy. Detached adult whitefly abdomens and heads
were fixed overnight in 2.5% glutaraldehyde in 1⫻ PBS at 4°C and processed using a standard method for transmission electron microscopy
(TEM): rinsing in 1⫻ PBS buffer, osmification, another rinse in 1⫻ PBS
buffer, dehydration in an ascending ethanol series, acetone incubation,
and embedment in epoxy resin agar 100 (Agar Scientific, Essex, England).
Thin sections (60 to 90 nm) were cut using an ultramicrotome, stained
with aqueous uranyl acetate and lead citrate, and examined in a Tecnai 12
electron microscope (Philips/FEI, Eindhoven, The Netherlands).
RESULTS
General distribution of Rickettsia in the adult hemolymph. The
distribution of Rickettsia was monitored in whole insects and dissected organs by means of FISH and TEM. The bacterium was
generally distributed throughout the body cavity, appearing in the
head, thorax, and abdomen (Fig. 1A). In some cases, the bacterium was heavily concentrated around bacteriocytes and was less
observable in other locations (Fig. 1B to F). The latter phenotype
was not significantly correlated with whitefly sex or age. Many
serial confocal microscopy and TEM sections indicated that Rickettsia is not present in bacteriocytes in the adult female stage, even
when the bacterium cells are highly abundant around bacteriocytes. In only one case, Rickettsia-like cells were observed attached
to the outer surface of the bacteriocyte and potentially internalized by the bacteriocyte (Fig. 1F). These observations suggest that
the mode of vertical transmission to the next generation via bacteriocytes cannot explain the high abundance and perfect vertical
transmission of Rickettsia from the female to its offspring.
Oocyte-mediated vertical transmission of Rickettsia. We
tested the hypothesis that Rickettsia hitchhikes with the bacteriocyte for vertical transmission via the oocyte, as was previously
described for the other endosymbionts in whiteflies (12, 20, 39, 40,
65). The results indicated that Rickettsia is the exception. The possibility for vertical transmission via free bacteriocytes was excluded on the basis of TEM and confocal serial sections (Fig. 1).
Rickettsia was not seen penetrating bacteriocytes that were already
contained in the oocyte or free in the hemolymph, suggesting that
this is not a route of oocyte entry. The path of Rickettsia entrance
into developing oocytes was followed, focusing on different developmental stages of the oocyte. Indeed, Rickettsia invaded the
oocytes in the very early stages of development, and high levels of
Applied and Environmental Microbiology
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
natural populations, as was recently shown in Arizona (45). The
latter study has further shown that Rickettsia-infected B. tabaci
females exhibit high fitness benefits, such as increased fecundity
and a greater rate of survival, and produce a higher proportion of
daughters via Rickettsia manipulation of host reproduction (45).
Additional studies have shown that the presence of Rickettsia in B.
tabaci populations influenced the whitefly’s response to heat stress
by benefitting its host under high temperatures (11) and has also
been shown to increase the whitefly’s susceptibility to chemical
insecticides (49). Studies on Rickettsia from other invertebrates
have revealed diverse effects of the bacteria on their hosts. For
example, in the pea aphid Acyrthosiphon pisum, Rickettsia-infected
individuals showed lower fresh body weight, reduced fecundity,
and significantly suppressed densities of Buchnera, suggesting a
negative effect of Rickettsia (16, 61). Reproductive manipulation
and Rickettsia-associated parthenogenesis have been shown in the
endoparasitoid Neochrysocharis formosa (42) and in the parasitoid
wasp Pnigalio soemius (37), as have Rickettsia-associated male killing in beetles (71, 69) and Rickettsia involvement in oogenesis of
booklice (72).
To unravel additional biological functions for Rickettsia in B.
tabaci, including possible transmission routes, the spatial subcellular localization of the bacterium in internal organs and cells
must be determined. Little is known about the specific subcellular
localization of Rickettsia within arthropod hosts. To hypothesize
possible biological functions and interactions with the insect host,
fluorescence microscopy and electron microscopy were used here
to show that Rickettsia occupies specific, previously undescribed
niches and organs within B. tabaci’s digestive, salivary, and reproductive organs and cells. The presence of the bacterium in these
organs gives some clue as to the specific undescribed interactions
and possible transmission routes, suggests possible horizontal
transfer of the bacterium during mating, and confirms a recent
report suggesting that horizontal transmission of Rickettsia might
be plant mediated (14).
Rickettsia-Whitefly Interactions
occupies most of the body cavity, while Portiera is located inside bacteriocytes in the abdomen; (B) Rickettsia is sometimes found at exceptionally high
concentrations around the bacteriocytes; (C) details of the box in panel B; (D) TEM image showing three bacteriocytes (B) in the hemolymph surrounded by high
concentrations of Rickettsia (R) cells; (E) Rickettsia (R) cells surrounding a bacteriocyte that harbors Portiera (P) and Hamiltonella (H) in the hemolymph of B.
tabaci biotype B; (F) box in panel E showing a bacteriocyte surrounded by Rickettsia (R) and some Rickettsia-like cells attached to or inside the bacteriocyte. Bars,
2.5 ␮m (D), 2 ␮m (E), and 0.5 ␮m (F). N, nucleus.
bacterial cells were concentrated inside the oocytes at all developmental stages, especially the younger stages, stages 1 and 2 (Fig. 2A
and C to F). The concentration of bacterial cells decreased in the
more mature oocytes, especially in their central part (Fig. 2A to E),
while they were not observed in free bacteriocyte cells surrounding the developing oocytes (Fig. 2F). The location of Rickettsia in
the developing oocytes was further explored using serial TEM sections. Young oocytes in stages 1 to 3 contained larger amounts of
Rickettsia cells in the cytoplasm and in the surrounding follicular
cells (Fig. 3A to C). However, oocytes in later stages 4 and 5 of
development contained bacterial cells mostly in the surrounding
follicular cells and less Rickettsia inside the oocyte cytoplasm (Fig.
3D and E). At stage 5 of development, the oocyte is almost completely filled with fat vacuoles and droplets, and bacterial cells were
not observed in the cytoplasm (Fig. 3E).
Organ- and tissue-specific infection. The general distribution
of Rickettsia in B. tabaci as observed by FISH led us to monitor its
presence in other whitefly organs and subcellularly, using both
TEM and FISH analyses. The organs and tissues analyzed were the
digestive and salivary systems, including the midgut, salivary
glands, and stylet. Furthermore, the male and female reproductive
organs were analyzed for infection by Rickettsia, and the possible
sexual transfer of the bacterium between infected and healthy individuals was tested. Other tissues that were tested for infection
were flight muscles in the thorax and fat bodies in the abdomen.
Experiments employing dissected midguts and Rickettsia-specific probe showed that the midgut contains very large amounts of
bacterial cells, unlike many of the other organs in B. tabaci (Fig. 4).
August 2012 Volume 78 Number 16
Rickettsia cells sometimes clustered in groups inside vacuolar-like
structures (Fig. 4A and 5C and D) and were easily distinguished by
TEM and FISH (Fig. 4A and C and 5A and B). The bacterium cells
were never observed in the midgut lumen (Fig. 4B and C and 5A
and C). Midguts dissected from younger 1- to 2-day-old males
and females, as well as older 7- to 14-day-old individuals, contained similar patterns of Rickettsia localization.
FISH analysis detected Rickettsia in the primary salivary glands
(Fig. 6B), and this result was confirmed using TEM (Fig. 6C and
D). The bacterial cells were located in cells 1 and 2 described by
Ghanim et al. (36). These cells contain exceptionally large nuclei,
electron-dense granules, and lipid-like accumulations, but their
cytoplasm is less dense than that of the other cells in the gland (36),
and they were hypothesized to act in producing the gelling saliva
required for stylet penetration in the plant leaf. Using FISH, we
were able to confirm the presence of Rickettsia in the whitefly
stylet, along the interlocked maxillae through which the food and
salivary canals pass (Fig. 7). Interestingly, a FISH signal was also
observed on the mandibles that surround these two canals (Fig. 7).
At this resolution, we were not able to determine whether the
signal exists only in the food canal, salivary canal, or both canals.
The reproductive organs of both B. tabaci males and females
were monitored for the presence of Rickettsia. The bacterium was
detected by FISH in dissected testes from B. tabaci males (Fig. 8A
to D) and dissected spermathecae from fertilized females (Fig. 8E
and F). Interestingly, the labeling was observed in all stages of
spermatogenesis in the primary cells of the primary glands in the
testes and spermathecae; however, fewer Rickettsia cells were as-
aem.asm.org 5567
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
FIG 1 Rickettsia localization in Bemisia tabaci adults by TEM and FISH analysis with Rickettsia-specific probe (blue) and a probe for Portiera (red). (A) Rickettsia
Brumin et al.
sociated with the mature spermatids (Fig. 8D). This observation
suggested that Rickettsia might be sexually transmitted between
individuals, one mechanism for horizontal transfer, thus affording a potential route for Rickettsia maintenance in B. tabaci populations. Indeed, copulation experiments that we conducted on
artificial diet between 30 Rickettsia-infected females with 30 uninfected males and vice versa showed that Rickettsia could be detected by PCR in some of the uninfected counterparts (data not
shown).
Further tissues in which Rickettsia cells were detected using
TEM ultrathin sections were muscle band cells in the flight muscles located in the thorax of adult males and females (Fig. 9A) and
fat bodies in the hemolymph of both sexes (Fig. 9B and C).
DISCUSSION
In the present study, we obtained new data regarding the infection of B. tabaci by Rickettsia. We further showed the organspecific distribution and possible relevance to the high abundance of Rickettsia in the insect populations (45). The high
abundance of Rickettsia in the hemolymph, especially around
bacteriocytes, first suggested possible vertical transmission via
the oocyte by hitchhiking with bacteriocytes, as was previously
suggested (38). Although it is widely accepted that many bacterial endosymbionts are vertically transmitted into newly developed oocytes with bacteriocytes, especially in whiteflies (12,
20, 38, 39, 40, 65), the results presented here refute this hypoth-
5568
aem.asm.org
esis and show that Rickettsia is the exception (Fig. 1). TEM
serial sections did not detect Rickettsia cells inside bacteriocytes, and only the abundant primary symbiont Portiera and
the secondary symbiont Hamiltonella were always associated
with bacteriocytes (Fig. 2 and 3). TEM sections through the
developing oocytes and bacteriocytes showed that younger
oocytes are heavily infected with Rickettsia cells, which are
mostly excluded and concentrate in follicular cells, with development. It is thus likely that Rickettsia cells are transferred to
the next generation via the egg by residing in the follicular cells.
The observation of younger oocytes being filled with bacterial
cells corresponds with the fact that young oocytes are more
permeable to materials in the hemolymph than developed
ones, which exhibit more selective barriers. Follicular cells facilitate the uptake of various materials from the hemolymph to
the oocyte and are involved in the synthesis and transport of
precursors of both the internal vitelline and external chorion
envelopes that surround the oocyte (53). Rickettsia has been
shown to invade the follicular cells of parasitic wasps of B.
tabaci (18), but bacterial cells were not observed beyond the
follicular cells. Other bacterial symbionts, such as Wolbachia in
Drosophila melanogaster (32) and Cardinium in the leafhopper
Scaphoideus titanus (60), have been shown to invade the reproductive system, including follicular cells, ensuring their own
transmission to the next generation. Our observation of Rickettsia invading the follicular cells and the cytoplasm of devel-
Applied and Environmental Microbiology
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
FIG 2 FISH of Bemisia tabaci ovaries using a Rickettsia-specific probe (green), a probe specific for Portiera (red), and DAPI staining of the nuclei (blue). (A)
Stages 1 to 4 of ovary development showing very high concentrations of Rickettsia cells invading younger ovaries; (B) one ovary at stage 4 of development showing
the distribution of Rickettsia cells in and around it; (C) a focal plane showing the presence of Rickettsia cells inside the ovary and at the center of the oocyte (O);
(D) ovary at stage 5 of development showing the included Rickettsia-free bacteriocyte (B); (E) same as panel D but with only one focal plane showing Rickettsia
cells inside the oocyte; (F) dissected ovaries and Rickettsia-free bacteriocytes (B) in the hemolymph of Bemisia tabaci.
Rickettsia-Whitefly Interactions
Rickettsia (R and arrows); (C) stage 3 ovary with follicular cells (FC) and Rickettsia (arrows) mainly in the oocyte (O); (D) follicular cells (FC) filled with and
surrounded by Rickettsia (R and arrows); (E) mature egg showing the oocyte (O) with Rickettsia excluded and only in the surrounding follicular cells (arrows).
H, hemolymph; FV, fat vacuole; FD, fat-dense granules. Bars, 4 ␮m (A), 2.5 ␮m (B), 1 ␮m (C), 1 ␮m (D), and 2 ␮m (E).
oping oocytes suggests that this bacterium uses the oocyte, not
the bacteriocyte route, to ensure its transmission to the next
generation.
For the first time, we show that the B. tabaci midgut is heavily
infected with Rickettsia. It is unclear how or when Rickettsia enters
and becomes established in the midgut cells. One possibility is the
oocyte and egg cells that give rise to intestinal tissue in later developmental stages. The high concentration of Rickettsia in the
midgut and its specific localization in the vacuoles (Fig. 5B and D)
suggest a possible role in food digestion. Midgut-associated sym-
bionts have been reported from many arthropods, such as the
Mediterranean fruit fly, tsetse fly, silkworms, termites, and stinkbugs, among many others (46). The diversity of midgut microbes
is enormous: they belong to various genera and exhibit different
localization patterns, modes of transmission, and functional roles,
depending on both the bacterium and the host (25, 29). Intestinal
microbes may contribute to food digestion, provide essential
amino acids and vitamins to the host, fix nitrogen, and keep out
potentially harmful microbes (2, 4, 5, 10, 66, 35). Gut microorganisms also possess metabolic properties that are absent in in-
FIG 4 FISH of dissected midgut using Rickettsia-specific probe (red) and DAPI staining. (A) Whole midgut showing Rickettsia concentrated in clusters in some
midgut cells (arrows); (B) DAPI staining showing high concentrations of Rickettsia in the cytoplasm of midgut cells; (C) same as panel B, but with FISH using a
Rickettsia-specific probe showing that the cells stained with DAPI are now labeled with the probe. C, ceca; FC, filter chamber; DM, descending midgut; AM,
ascending midgut; H, hindgut; L, lumen; N, nucleus.
August 2012 Volume 78 Number 16
aem.asm.org 5569
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
FIG 3 TEM sections of Bemisia tabaci abdomen. (A) Stage 1 ovary filled with Rickettsia in follicular cells (FC); (B) stage 2 ovary filled with and surrounded by
Brumin et al.
sects, thus enabling the phytophagous insects to overcome
biochemical barriers to herbivory by detoxifying plant allelochemicals such as flavonoids, tannins, and alkaloids (7, 25, 27).
However, the gut bacteria explored to date are phylogenetically
distant from the alphaproteobacterial group of Rickettsia, and
nothing is known about the latter’s functional role in the digestive
system.
Interestingly, Rickettsia cells were found in the primary salivary
FIG 6 TEM section and FISH using Rickettsia-specific probe (green) of Bemisia tabaci primary salivary glands. (A) Two dissected primary salivary glands (PSG)
stained with DAPI (blue); (B) same as in panel A, but with Rickettsia-specific FISH; (C) TEM section in one primary salivary gland (PSG) in the thorax, showing
a giant cell and the nucleus (N); (D) details of the box in panel C showing Rickettsia-specific FISH in the gland and lipid-like accumulations (arrows). T, thorax;
MB, muscle band; R, Rickettsia. Bars, 2.5 ␮m (C) and 1 ␮m (D).
5570
aem.asm.org
Applied and Environmental Microbiology
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
FIG 5 TEM sections of Bemisia tabaci midgut. (A) Concentration of Rickettsia in midgut epithelial cells (EC); (B) details of the box in panel A; (C) concentration of
Rickettsia (R) in vacuoles in midgut epithelial cells (EC); (D) details of the box in panel C. L, lumen; N, nucleus. Bars, 3 ␮m (A), 1 ␮m (B), 2.5 ␮m (C), and 1 ␮m (D).
Rickettsia-Whitefly Interactions
FIG 7 FISH of Bemisia tabaci stylet using Rickettsia-specific probe (red). Specific FISH signal seen under bright field (A) and dark field (B) in the maxillae that
form the food and salivary canals (MX) and in the mandibles (MD), which were detached from the maxilla in this preparation. LR, labrum.
Rickettsia in the whitefly stylet is restricted to the interlocked maxillae through which the food and salivary canals pass (Fig. 7).
Interestingly, a FISH signal was also observed on the mandibles
that surround these two canals (Fig. 7). At this resolution, we were
not able to determine whether the signal exists in the food canal,
salivary canal, or both canals; however, the presence of the bacterium in either one of these canals suggests its passage into or out of
the whitefly body with plant sap. It has recently been suggested
that Rickettsia is horizontally transmitted through the plant host
(14), where it was observed in the plant sieve elements following
feeding of Rickettsia-infected whiteflies and was also acquired
FIG 8 FISH of Bemisia tabaci testes and spermathecae dissected from males and females, with Rickettsia-specific probe (green) and DAPI staining (blue). (A)
Two testicles (T) and two accessory glands (AG) in the male genitalia dissected and stained with DAPI. The ducts connecting the testicles to the common duct
(CD) are also seen (arrows). (B) One testicle under bright field showing the major locations of sperm production where the spermatogonia (sg), spermatocytes
(sy), and spermatids (st) or mature sperm are located; (C) the same testicle in panel B with Rickettsia-specific FISH (green) and DAPI staining; (D) the same
testicle in panels B and C with Rickettsia-specific FISH only; (E) spermatheca dissected from fertilized female showing the sperm and Rickettsia-specific FISH
(green); (F) spermatheca seen in panel E with Rickettsia-specific FISH (green) only.
August 2012 Volume 78 Number 16
aem.asm.org 5571
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
glands (Fig. 6). The occurrence of bacteria in the salivary glands of
different insect species has been reported and has been suggested
to serve as a route for their horizontal transmission to plants or
other intermediate hosts (26, 54, 59, 60). The first and strongest
evidence of horizontal transmission of symbiotic bacteria via an
insect’s plant host was reported for the leafhopper Euscelidius variegates (59) and the pathogenic symbiont BEV. Similar to other
sap-sucking arthropods, the horizontal transmission of B. tabaci
symbionts requires the passage of the bacterial cells through several barriers, most importantly, from salivary gland cells through
the salivary duct to the stylet. Our results show that the presence of
Brumin et al.
FIG 9 TEM section in Bemisia tabaci thorax and fat body in the hemolymph. (A) Rickettsia cells (R) are located in one muscle cell close to the nucleus; (B) several
fat body cells (FB) filled with fat vacuoles; (C) details of the box in panel B showing Rickettsia cells in fat body cells (arrows). MB, muscle bands; N, nucleus; FV,
fat vacuole. Bars, 1 ␮m (A), 2.5 ␮m (B), and 1 ␮m (C).
5572
aem.asm.org
ginale (Rickettsiales: Anaplasmataceae) in the male Rocky Mountain wood tick Dermacentor andersoni (48).
The association of Rickettsia with B. tabaci fat bodies was not
expected, as the presence and functional association of endosymbionts in fat bodies are not common in insects. Only endosymbionts from the blattabacteria that coevolved with termites and
cockroaches were reported from the fat body, but no specific functional role for these endosymbionts has been assigned (19, 50, 52).
In whiteflies, vitellogenin and possibly other important proteins
are produced in the fat body, and thus, the presence of Rickettsia
might indicate a role in these processes.
In conclusion, FISH and TEM analyses showed the presence of
a Rickettsia endosymbiont in B. tabaci whiteflies. Rickettsia was
localized to major organs involved in the host’s reproductive,
feeding, circulation, and secretion systems. Interestingly, the
broad spectrum of Rickettsia-infected tissues in B. tabaci documented in this study resembles that found for the pathogenic Rickettsia felis in cat fleas (1, 9, 51). Further investigation is warranted
to understand the mechanism by which Rickettsia reaches and
becomes established in these organs and its possible roles there.
ACKNOWLEDGMENTS
We thank Eduard Belausov, Vered Holdengreber, and Svetlana Kontsedalov for technical assistance.
This research was supported by Binational Science Foundation (BSF)
grant 2007045 and Israel Science Foundation (ISF) grant 884/07.
REFERENCES
1. Adams JR, Schmidtmann ET, Azad AF. 1990. Infection of colonized cat
fleas, Ctenocephalides felis (Bouche), with a rickettsia-like microorganism. Am. J. Trop. Med. Hyg. 43:400 – 409.
2. Anand AA, et al. 2010. Isolation and characterization of bacteria from the
gut of Bombyx mori that degrade cellulose, xylan, pectin and starch and
their impact on digestion. J. Insect Sci. 10:107.
3. Baumann P. 2005. Biology of bacteriocyte-associated endosymbionts of
plant sap-sucking insects. Annu. Rev. Microbiol. 59:155–189.
4. Behar A, Jurkevitch E, Yuval B. 2008. Bringing back the fruit into fruit
fly-bacteria interactions. Mol. Ecol. 7:1375–1386.
5. Behar A, Yuval B, Jurkevitch E. 2008. Gut bacterial communities in the
Mediterranean fruit fly (Ceratitis capitata) and their impact on host longevity. J. Insect Physiol. 54:1377–1383.
6. Benson MJ, Gawronski JD, Eveleigh DE, Benson DR. 2004. Intracellular
symbionts and other bacteria associated with deer ticks (Ixodes scapularis)
from Nantucket and Wellfleet, Cape Cod, Massachusetts. Appl. Environ.
Microbiol. 70:616 – 620.
7. Berenbaum MR. 1988. Micro-organisms as mediators of intertrophic and
intratrophic interactions, p 91–123. In Barbosa P, Letourneau DK (ed),
Applied and Environmental Microbiology
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
from the plant by Rickettsia-free whiteflies. Combining our microscopy findings with the recent report of Caspi-Fluger et al. (14)
provides stronger evidence for the horizontal transfer of Rickettsia
in B. tabaci through the plant host.
Our results indicated the infection of male and female reproductive tissues by Rickettsia, and preliminary copulation experiments that we performed, using artificial diet, suggest possible
transfer of Rickettsia during mating (Fig. 8 and unpublished data).
These results, in addition to a recent study suggesting horizontal
transmission through the plant (14), might explain the exceptionally high prevalence of Rickettsia in natural populations, such as its
establishment and near fixation in populations of B. tabaci in Arizona in just 6 years (45). In the latter study, Rickettsia-infected
whiteflies produced more offspring, had higher survival rates, developed faster, and produced a higher proportion of daughters.
The observed sex ratio bias might be attributed in part to possible
effects of Rickettsia on the Rickettsia-infected sperm in the female
spermatheca, which is used by the female to fertilize eggs. In this
arrhenotokous mode of reproduction, fertilized eggs develop to
females, while unfertilized eggs develop to males.
Wolbachia is the most studied microorganism with respect to
associations with insect reproductive organs, which induce cytoplasmic incompatibility (CI) and reproductive manipulations in
several insect species, and was associated with breakdown of spermatogenesis (30). Several other bacterial symbionts which reside
in male and female reproductive organs and induce reproductive
manipulations were further described, including Cardinium (21,
73, 74), “Candidatus Blochmannia floridanus” (62), and Arsenophonus (70). A spotted fever group Rickettsia was previously described in the spermatogonia, spermatocytes, and maturing spermatids of its male tick vector Ixodes ricinus, and it could be
sexually transmitted to females, but it did not have negative effects
on the developing eggs (44).
The presence of Rickettsia in muscle cells was somewhat surprising, and it is unclear whether this has any effect on the musclerelated biology of the whitefly, such as flight. Several bacterial
symbionts that have been previously localized to thoracic and
muscle tissues of their hosts were associated with inducing negative and even pathogenic effects on the host. The best-studied
examples of Rickettsia species that infect muscle tissues stem from
pathogenic Rickettsia species and their interactions with their arthropod vectors. A few examples include the two major groups of
pathogenic Rickettsia, the spotted fever and typhus groups, and
their interactions with their tick vectors (64) and Anaplasma mar-
Rickettsia-Whitefly Interactions
8.
9.
10.
11.
12.
13.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
August 2012 Volume 78 Number 16
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
in a bruchid beetle Kytorhinus sharpianus (Coleoptera: Bruchidae). Appl.
Entomol. Zool. 34:391–397.
Fukatsu T, Nikoh N, Kawai R, Koga R. 2000. The secondary endosymbiotic bacterium of the pea aphid Acyrthosiphon pisum (Insecta: Homoptera). Appl. Environ. Microbiol. 66:2748 –2758.
Fukatsu T, Hosokawa T. 2009. Capsule-transmitted obligate gut bacterium of plataspid stinkbugs: a novel model system for insect symbiosis
studies, p 95–122. In Bourtzis K, Miller TA (ed), Insect symbiosis, vol 3.
CRC Press LLC, Boca Raton, FL.
Ghanim M, et al. 2001. Digestive, salivary, and reproductive organs of
Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) B type. J. Morphol.
248:22– 40.
Giorgini M, Bernardo U, Monti MM, Nappo AG, Gebiola M. 2010.
Rickettsia symbionts cause parthenogenetic reproduction in the parasitoid
wasp Pnigalio soemius (Hymenoptera: Eulophidae). Appl. Environ. Microbiol. 76:2589 –2599.
Gottlieb Y, et al. 2006. Identification and localization of a Rickettsia sp. in
Bemisia tabaci (Homoptera: Aleyrodidae). Appl. Environ. Microbiol. 72:
3646 –3652.
Gottlieb Y, et al. 2008. Inherited intracellular ecosystem: symbiotic bacteria share bacteriocytes in whiteflies. FASEB J. 22:1–9.
Gueguen G, et al. 2010. Endosymbiont metacommunities, mtDNA diversity and the evolution of the Bemisia tabaci (Hemiptera: Aleyrodidae)
species complex. Mol. Ecol. 19:4365– 4376.
Hadfield SJ, Axton JM. 1999. Germ cells colonized by endosymbiotic
bacteria. Nature 402:482.
Hagimori T, Abe Y, Date S, Miura K. 2006. The first finding of a
Rickettsia bacterium associated with parthenogenesis induction among
insects. Curr. Microbiol. 52:97–101.
Haine ER. 2008. Symbiont-mediated protection. Proc. Biol. Sci. 275:353–
361.
Hayes SF, Burgdorfer W, Aeschlimann A. 1980. Sexual transmission of
spotted fever group Rickettsiae by infected male ticks: detection of Rickettsiae in immature spermatozoa of Ixodes ricinus. Infect. Immun. 27:638 –
642.
Himler AG, et al. 2011. Rapid spread of a bacterial symbiont in an invasive whitefly is driven by fitness benefits and female bias. Science 332:254 –
256.
Jurkevitch E. 2011. Insect “symbiology” is coming of age, bridging between bench and field, p 27– 41. In Rosenberg E, Gophna U (ed), Beneficial microorganisms in multicellular life forms. Springer-Verlag, Berlin,
Germany.
Kikuchi Y, Sameshima S, Kitade O, Kojima J, Fukatsu T. 2002. Novel
clade of Rickettsia spp. from leeches. Appl. Environ. Microbiol. 68:999 –
1004.
Kocan KM, et al. 2009. Silencing of genes involved in Anaplasma marginale-tick interactions affects the pathogen developmental cycle in Dermacentor variabilis. BMC Dev. Biol. 9:42.
Kontsedalov S, et al. 2008. The presence of Rickettsia is associated with
increased susceptibility of Bemisia tabaci (Homoptera: Aleyrodidae) to
insecticides. Pest Manage. Sci. 64:789 –792.
López-Sánchez MJ, et al. 2008. Blattabacteria, the endosymbionts of
cockroaches, have small genome sizes and high genome copy numbers.
Environ. Microbiol. 10:3417–3422.
Macaluso KR, Pornwiroon W, Popov VL, Foil LD. 2008. Identification
of Rickettsia felis in the salivary glands of cat fleas. Vector Borne Zoonotic
Dis. 8:391–396.
Maekawa K, Kon M, Matsumoto T, Araya K, Lo N. 2005. Phylogenetic
analyses of fat body endosymbionts reveal differences in invasion times of
blaberid wood-feeding cockroaches (Blaberidae: Panesthiinae) into the
Japanese archipelago. Zool. Sci. 22:1061–1067.
Margaritis LH. 1985. Structure and physiology of the insect egg-shell,
p 153–230. In Kerkut GA, Gilbert LI (ed), Comprehensive insect physiology, biochemistry and pharmacology. Pergamon Press, Oxford,
United Kingdom.
Mitsuhashi W, Saiki T, Wei W, Kawakita H, Sato M. 2002. Two novel
strains of Wolbachia coexisting in both species of mulberry leafhoppers.
Insect Mol. Biol. 11:577–584.
Moran NA, Russell JA, Koga R, Fukatsu T. 2005. Evolutionary relationships of three new species of Enterobacteriaceae living as symbionts of
aphids and other insects. Appl. Environ. Microbiol. 71:3302–3310.
Moran NA, McCutcheon JP, Nakabachi A. 2008. Genomics and evolution of heritable bacterial symbionts. Annu. Rev. Genet. 42:165–190.
aem.asm.org 5573
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
14.
Novel aspects of insect-plant interactions. John Wiley & Sons, Inc, New
York, NY.
Bourtzis K, Miller T. 2006. Insect symbiosis, vol 2. CRC Press LLC, Boca
Raton, FL.
Bouyer DH, et al. 2001. Rickettsia felis: molecular characterization of a
new member of the spotted fever group. Int. J. Syst. Evol. Microbiol.
51:339 –347.
Breznak JA. 2000. Ecology of prokaryotic microbes in the guts of woodand litterfeeding termites, p 209 –231. In Abe T, Bignell DE, Higashi M
(ed), Termites: evolution, sociality, symbiosis, ecology. Kluwer Academic
Publishers, Dordrecht, The Netherlands.
Brumin M, Kontsedalov S, Ghanim M. 2011. Rickettsia influences thermotolerance in the whitefly Bemisia tabaci B biotype. Insect Sci. 18:57– 66.
Buchner P. 1965. Endosymbiosis of animals with plant microorganisms,
p 210 – 432. Interscience, New York, NY.
Bution ML, Caetano FH, Zara, FJ. 2008. Contribution of the Malpighian
tubules for the maintenance of symbiotic microorganisms in Cephalotes
ants. Micron 39:1179 –1183.
Caspi-Fluger A, et al. 2012. Horizontal transmission of the insect symbiont Rickettsia is plant-mediated. Proc. Biol. Sci. 279:1791–1796.
Chen DQ, Campbell BC, Purcell AH. 1996. A new Rickettsia from a
herbivorous insect, the pea aphid Acyrthosiphon pisum (Harris). Curr.
Microbiol. 33:123–128.
Chen DQ, Montllor CB, Purcell AH. 2000. Fitness effects of two facultative endosymbiotic bacteria on the pea aphid, Acyrthosiphon pisum,
and the blue alfalfa aphid, A. kondoi. Entomol. Exp. Appl. 95:315–323.
Chiel E, Gottlieb Y, Inbar M, Zchori-Fein E, Ghanim M. 2007. Distribution of secondary symbionts in Israeli populations of Bemisia tabaci.
Bull. Entomol. Res. 97:407– 413.
Chiel E, et al. 2009. Almost there: transmission routes of bacterial symbionts between trophic levels. PLoS One 4:e4767. doi:10.1371/
journal.pone.0004767.
Clark J, Kambhampati S. 2003. Phylogenetic analysis of Blattabacterium,
endosymbiotic bacteria from the wood roach, Cryptocercus (Blattodea:
Cryptocercidae), including a description of three new species. Mol. Phylogenet. Evol. 26:82– 88.
Costa HS, Toscano NC, Henneberry TJ. 1996. Mycetocyte inclusion in
the oocytes of Bemisia argentifolii (Homoptera: Aleyrodidae). Ann. Entomol. Soc. Am. 89:694 – 699.
Dallai R, Mercati D, Giusti F, Gottardo M, Carapelli A. 2011. A Cardinium-like symbiont in the proturan Acerella muscorum (Hexapoda).
Tissue Cell 43:151–156.
Davis MJ, Ying Z, Brunner BR, Pantoja A, Ferwerda FH. 1998. Rickettsial relative associated with papaya bunchy top disease. Curr. Microbiol. 36:80 – 84.
De Barro PJ. 2005. Genetic structure of the whitefly Bemisia tabaci in the
Asia-Pacific region revealed using microsatellite markers. Mol. Ecol. 14:
3695–3718.
De Barro PJ, Liu SS, Boykin LM, Dinsdale AB. 2011. Bemisia tabaci: a
statement of species status. Annu. Rev. Entomol. 56:1–19.
Dillon RJ, Dillon VM. 2004. The gut bacteria of insects: nonpathogenic
interactions. Annu. Rev. Entomol. 49:71–92.
Dobson SL, et al. 1999. Wolbachia infections are distributed throughout
insect somatic and germ line tissues. Insect Biochem. Mol. Biol. 29:153–
160.
Douglas AE. 1992. Microbial brokers of insect-plant interactions, p 329 –
336. In Menken SBJ, Visser JH, Harrewijn P (ed), Proceedings of the 8th
International Symposium on Insect-Plant Relationships. Kluwer Academic Publishers, Dordrecht, The Netherlands.
Douglas AE. 1998. Nutritional interactions in insect-microbial symbiosis:
aphids and their symbiotic bacteria Buchnera. Annu. Rev. Entomol. 43:
17–37.
Douglas AE. 2011. Lessons from studying insect symbioses. Cell Host
Microbe 10:359 –367.
Ehrman L, Kernaghan RP. 1972. Infectious heredity in Drosophila paulistorum, p 227–250. In Pathogenic mycoplasmas: a Ciba Foundation symposium. Associate Scientific Publishers, Amsterdam, The Netherlands.
Frohlich DR, Torres-Jerez I, Bedford ID, Markham PG, Brown JK.
1999. A phylogeographical analysis of the Bemisia tabaci species complex
based on mitochondrial DNA markers. Mol. Ecol. 8:1683–1691.
Frydman HM, Li JM, Robson DN, Wieschaus E. 2006. Somatic stem cell
niche tropism in Wolbachia. Nature 441:509 –512.
Fukatsu T, Shimada M. 1999. Molecular characterization of Rickettsia sp.
Brumin et al.
5574
aem.asm.org
67.
68.
69.
70.
71.
72.
73.
74.
cellulase activity against crystalline- vs carboxymethyl-cellulose in the digestive systems of diverse, wood-feeding termites. Physiol. Entomol. 30:
372–380.
Tsuchida T, Koga R, Meng XY, Matsumoto T, Fukatsu T. 2005. Characterization of a facultative endosymbiotic bacterium of the pea aphid
Acyrthosiphon pisum. Microb. Ecol. 49:126 –133.
Veneti Z, Clark ME, Karr TL, Savakis C, Bourtzis K. 2004. Heads or tails:
host-parasite interactions in the Drosophila-Wolbachia system. Appl. Environ. Microbiol. 70:5366 –5372.
von der Schulenburg JHG, et al. 2001. Incidence of male-killing Rickettsia spp. (␣-proteobacteria) in the ten-spot ladybird beetle Adalia decempunctata L. (Coleoptera: Coccinellidae). Appl. Environ. Microbiol. 67:
270 –277.
Werren JH, Skinner SW, Huger AM. 1986. Male-killing bacteria in a
parasitic wasp. Science 231:990 –992.
Werren JH, et al. 1994. Rickettsial relative associated with male killing in
the ladybird beetle (Adalia bipunctata). J. Bacteriol. 176:388 –394.
Yusuf M, Turner B. 2004. Characterisation of Wolbachia-like bacteria
isolated from the parthenogenetic stored-product pest psocid Liposcelis bostrychophila (Badonnel) (Psocoptera). J. Stored Prod. Res. 40:
207–225.
Zchori-Fein E, et al. 2001. A newly discovered bacterium associated with
parthenogenesis and a change in host selection behavior in parasitoid
wasps. Proc. Natl. Acad. Sci. U. S. A. 98:12555–12560.
Zchori-Fein E, Perlman SJ, Kelly SE, Katzir N, Hunter MS. 2004.
Characterization of a “Bacteroidetes” symbiont in Encarsia wasps (Hymenoptera: Aphelinidae): proposal of “Candidatus cardinium hertigii.” Int. J.
Syst. Evol. Microbiol. 54:961–968.
Applied and Environmental Microbiology
Downloaded from http://aem.asm.org/ on July 27, 2012 by guest
57. Oliver KM, Russell JA, Moran NA, Hunter MS. 2003. Facultative bacterial symbionts in aphids confer resistance to parasitic wasps. Proc. Natl.
Acad. Sci. U. S. A. 100:1803–1807.
58. Perlman SJ, Hunter MS, Zchori-Fein E. 2006. The emerging diversity of
Rickettsia. Proc. R. Soc. Lond. B Biol. Sci. 273:2097–2106.
59. Purcell AH, Suslow KG, Klein M. 1994. Transmission via plants of an
insect pathogenic bacterium that does not multiply or move in plants.
Microb. Ecol. 27:19 –26.
60. Sacchi L, et al. 2008. Multiple symbiosis in the leafhopper Scaphoideus
titanus (Hemiptera: Cicadellidae): details of transovarial transmission of
Cardinium sp. and yeast-like endosymbionts. Tissue Cell 40:231–242.
61. Sakurai M, Koga R, Tsuchida T, Meng XY, Fukatsu T. 2005. Rickettsia
symbiont of the pea aphid Acyrthosiphon pisum: novel cellular tropism,
effect on the host fitness, and interaction with the essential symbiont
Buchnera. Appl. Environ. Microbiol. 71:4069 – 4075.
62. Sauer C, Dudaczek D, Hölldobler B, Gross R. 2002. Tissue localization
of the endosymbiotic bacterium “Candidatus Blochmannia floridanus” in
adults and larvae of the carpenter ant Camponotus floridanus. Appl. Environ. Microbiol. 68:4187– 4193.
63. Skaljac M, Zanic K, Ban S, Kontsedalov S, Ghanim M. 2010. Coinfection and localization of secondary symbionts in two whitefly species.
BMC Microbiol. 10:142. doi:10.1186/1471-2180-10-142.
64. Socolovschi C, Mediannikov O, Raoult D, Parola P. 2009. The relationship between spotted fever group Rickettsiae and Ixodid ticks. Vet. Res.
40:34.
65. Szklarzewicz T, Moska A. 2001. Ultrastructure, distribution, and transmission of endosymbionts in the whitefly Aleurochiton aceris Modeer (Insecta, Hemiptera, Aleyrodinea). Protoplasma 218:45–53.
66. Tokuda G, Lo N, Watanabe H. 2005. Marked variations in patterns of