Wolbachia Induces Density-Dependent Inhibition to

Dengue Virus in Mosquito Cells

Peng Lu1, Guowu Bian2, Xiaoling Pan2, Zhiyong Xi1,2*
1 Department of Microbiology and Molecular Genetics, Michigan State University, East Lansing, Michigan, United States of America, 2 Department of Entomology,
Michigan State University, East Lansing, Michigan, United States of America

Abstract
Wolbachia is a maternal transmitted endosymbiotic bacterium that is estimated to infect up to 65% of insect species. The
ability of Wolbachia to both induce viral interference and spread into mosquito vector population makes it possible to
develop Wolbachia as a biological control agent for dengue control. While Wolbachia induces resistance to dengue virus in
the transinfected Aedes aegypti mosquitoes, a similar effect was not observed in Aedes albopictus, which naturally carries
Wolbachia infection but still serves as a dengue vector. In order to understand the mechanism of this lack of Wolbachia-
mediated viral interference, we used both Ae. albopictus cell line (Aa23) and mosquitoes to characterize the impact of
Wolbachia on dengue infection. A serial of sub-lethal doses of antibiotic treatment was used to partially remove Wolbachia
in Aa23 cells and generate cell cultures with Wolbachia at different densities. We show that there is a strong negative linear
correlation between the genome copy of Wolbachia and dengue virus with a dengue infection completely removed when
Wolbacha density reaches a certain level. We then compared Wolbachia density between transinfected Ae. aegypti and
naturally infected Ae. albopictus. The results show that Wolbachia density in midgut, fatbody and salivary gland of Ae.
albopictus is 80-, 18-, and 24-fold less than that of Ae. aegypti, respectively. We provide evidence that Wolbachia density in
somatic tissues of Ae. albopictus is too low to induce resistance to dengue virus. Our results will aid in understanding the
mechanism of Wolbachia-mediated pathogen interference and developing novel methods to block disease transmission by
mosquitoes carrying native Wolbachia infections.

Citation: Lu P, Bian G, Pan X, Xi Z (2012) Wolbachia Induces Density-Dependent Inhibition to Dengue Virus in Mosquito Cells. PLoS Negl Trop Dis 6(7): e1754.
doi:10.1371/journal.pntd.0001754
Editor: Scott L. O’Neill, Monash University, Australia
Received February 17, 2012; Accepted June 12, 2012; Published July 24, 2012
Copyright: ß 2012 Lu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the National Institutes of Health/National Institute of Allergy and Infectious Disease grant R01AI080597 (http://www.niaid.
nih.gov). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: [email protected]

Introduction cytoplasmic incompatibility (CI) mechanism, Wolbachia can induce

early embryo death when a Wolbachia-infected male mates with an
Dengue fever and dengue hemorrhagic fever are emerging uninfected female [5]. Since infected females can successfully mate
globally as a major public health problem in more than 100 with uninfected males, Wolbachia can spread quickly in a
countries. 2.5 to 3.0 billion people, or two fifths of the world’s population. This has been observed to occur naturally in Drosophila
populations, are currently living in dengue-endemic areas. Each simulans [6], and demonstrated in Ae. aegypti through both
year there are an estimated 100 million new dengue cases and laboratory cage studies and a recent field trial [7,8].
22,000 deaths [1]. The number of cases reported to WHO has Another important feature of Wolbachia is its ability to induce
increased nearly nine times in the past 4 decades, and it continues resistance to a variety of pathogens, including DENV, in its insect
to expand into temperate climates. Its fatal form, dengue hosts [9,10,11,12,13,14,15,16]. In the transinfected Ae. aegypti, all
hemorrhagic fever, has expended from Southeast Asia to 28 the three different types of Wolbachia, wAlbB, wMelPop-CLA, and
countries in the Western Hemisphere. Dengue virus (DENV) is wMel, show a significant inhibition to replication and dissemina-
transmitted to humans by the mosquitoes Aedes aegypti and Aedes tion of DENV, resulting in either complete or partial block of viral
albopictus. Currently no drug therapy or vaccines are available for transmission [9,12,13]. Recent studies further show that Wolbachia
dengue fever. Vector control is the primary intervention tool, but induces production of reactive-oxygen species (ROS) which then
this barrier is weakened by increased pesticide resistance. This activate Toll-pathway to induce expression of antiviral effectors
situation justifies seeking alternative and innovative approaches for [17]. In the Drosophila host, native Wolbachia can also confer
the development of new prevention options. Wolbachia-based resistance to DENV and the other pathogens [15]. This resistance
dengue control strategy is showing great potential because it could appears to be induced by the non-immune related mechanisms
provide a solution that will be more sustainable, economical and because the tested immune genes do not show differential
environmental friendly than the other methods [2]. expression in response to Wolbachia [18,19]. Inhibition of dengue
Wolbachia spp. are intracellular alpha-proteobacteria closely virus replication was also observed in cell lines, and the extent of
related to Rickettsia. Maternally inherited infections with Wolbachia inhibition is related to bacterial density [20].
occur in more than 65% of all the insect species and approxi- Ae. albopictus is generally considered as the secondary vector of
mately 28% of the surveyed mosquito species [3,4]. Through the DENV. This mosquito species naturally carries two types of

www.plosntds.org 1 July 2012 | Volume 6 | Issue 7 | e1754

 Wolbachia Density-Dependent Viral Interference

Author Summary 75-cm2 flasks of confluent Aa23 cells were shaken, centrifuged at
1,0006g for 10 min, re-suspended in 1.5 ml Schneider’s Drosophila
Transmitted by mosquitoes, dengue virus causes the most Medium in a 50 ml conical tube, and lysed by votexing with 3-mm-
important arbovirus disease in humans. Increasing prob- diameter glass beads. The lysate was centrifuged at 2,5006 g for
lems in insecticide resistance and the lack of drugs and 10 min, and the supernatant was filtered through a 5 mm syringe
vaccines make it urgent to develop novel strategies for filter (Millipore). 500 ml of the filtrate was overlaid on Aag2 cells
dengue control. Wolbachia is a maternally transmitted grown in 12-well plate to 80% confluence. The plate was
Gram-negative endosymbiotic bacterium that infects centrifuged at 2,0006 g at 15uC for 1 h and then incubated at
approximately 28% of mosquito species. It can not only 26uC overnight. After that, the cells were transferred to a 25 ml cell
spread within mosquito populations through its unique flask with fresh medium. Wolbachia was stably maintained in w-Aag2
ability to manipulate mosquito reproduction but can also
for more than 17 passages before being used for dengue infection.
induce resistance to dengue virus in mosquito vectors. This
leads to a genetic control strategy in which mosquito
hosts are made to be inhospitable to dengue virus Density-dependent assay
through population replacement. However, it is chal- To generate Aa23 cells with different Wolbachia densities, the
lenged by the fact that Ae. albopictus naturally carries cell culture was treated with rifampicin to make a final
Wolbachia infections but still services as a dengue vector. concentration at 5 mg/ml, 0.5 mg/ml and 0.05 mg/ml in 24-well
In this study we show the native Wolbachia induces a plates. For each dose, cells were treated for four time periods: 5 h,
resistance to dengue virus in Wolbachia density-depen- 10 h, 40 h and 70 h. After treatment, the old medium was
dent manner in Ae. albopictus. With a decrease in removed and cells were washed with fresh medium for one time.
Wolbachia density within the host cells, dengue infection After adding fresh medium, cells were grown for another three
increases dramatically. We provide evidence that a very days. After that, cells were passaged to another 24 well-plate with
low Wolbachia density in mosquito tissues where dengue new medium. Two days after the passage, cells were infected with
virus will reside and travel could contribute to absence of DENV-2 at multiplicity of infection (MOI) of 0.1. At Day 5 post
Wolbachia-mediated resistance to dengue virus in Ae. infection, cells were collected to measure the genome copy of
albopictus.
Wolbachia and DENV-2. To study Wolbachia-density dependent
expression of Defensin D (DEFD), cells were grown for 7 days in
Wolbachia, wAlbA and wAlbB, which distribute throughout both fresh medium after rifampicin treatment. Then, cells were
germ line and somatic tissues in mosquitoes [21]. However, collected to measure both the genome copy of Wolbachia and the
Wolbachia-mediated viral interference was not observed in Ae. expression of DEFD [27].
albopictus [12], consistent with the fact that it is a competent vector
for at least 22 arboviruses [22]. The recent study shows a
DENV-2 Infections in Mosquitoes
transinfected Ae. albopictus line that carries wMel is resistant to
The New Guinea C strain of DENV serotype 2 (DENV-2) was
DENV [23], which excludes the possibility that Ae. albopictus lacks
propagated in C6/36 cells as previously described [28]. Virus was
the genetic background for Wolbachia to induce viral interference.
harvested 7 days post infection by collection of supernatants and
Thus, although both wAlbB and Ae. albopictus own the machinery
centrifugation at 3,0006 g. The virus suspension was mixed in 1:1
to induce viral interference, the resistance to DENV does not
with commercial ox defibrinated blood (Colorado Serum Com-
occur for unknown reasons.
pany). The blood meal was warmed at 37uC in water bath for
In this work, we used an Aa23 cell line, initially established from
30 min and then used to feed 7-day-old mosquitoes as described
eggs of Ae. albopictus [24], to study the mechanism underlying the lack
previously [29].
of Wolbachia-mediated resistance to DENV. We confirmed that
wAlbB induces a strong resistance to DENV in Aa23 cell line.
Moreover, the levels of resistance strongly correlate with density of RNA Extraction, cDNA Synthesis and Quantitative Reverse
Wolbachia in Aa23 cells. Further comparison of genome copy of Transcription Polymerase Chain Reaction (qRT-PCR)
wAlbB in both somatic and germ line tissues between dengue The total RNA or viral genomic RNA were extracted from
resistant transinfected Ae. aegypti and susceptible Ae. albopictus suggests either cell lines or mosquitoes tissues by RNeasy Mini Kit
that Wolbachia density is too low to induce resistance to DENV in Ae. (QIAGEN Sciences, Germantown, MD, USA) and then the
albopictus. cDNA were transcript using QuantiTect reverse Transcription Kit
(QIAGEN Sciences, Germantown, MD, USA). Real-time PCR
Methods was performed using the QuantiTect SYBR Green PCR Kit
(QIAGEN Sciences, Germantown, MD, USA) and ABI Prism
Mosquito rearing and cell culture maintenance 7900HT Sequence Detection System (Applied Biosystems, Foster
Houston (HOU) and HT1 strain of Ae. albopictus and WB1 strain City, CA, USA). DENV-2 genomic RNA was measured by qRT-
of Ae. aegypti were maintained at 27uC and RH 85% with a 12-hr PCR using primers directed to NS5 gene [30]. The dengue copy
light/dark cycle. Ae. albopictus Aa23, Aa23T cell lines (derived from was normalized with host actin gene [31]. A standard curve was
Aa23 cells through tetracycline treatment) and Ae. aegypti Aag2 cell generated for each of NS5 and actin by analyzing 101 to 108
line were cultured as previously described [25,26]. Those cells copies/reaction of the two plasmids containing each individual
were maintained in Schneider’s Drosophila Medium (Invitrogen) fragment [31]. DENV-2 in mosquito head was diagnosed by RT-
supplemented with 10% (v/v) heat-inactivated fetal bovine serum PCR with the same NS5 primers. DEFD was amplified with the
(FBS). primers: For (59-GTCTGTTGCCAACTCTCTTT -39) and Rev
(59- CACAAGCACTGTCACCAAC -39).
Introduction of Wolbachia into Aag2 cells
w-Aag2 cell line was generated by introducing Wolbachia from Wolbachia Quantitative PCR (q-PCR)
Aa23 cells into Aag2 cell line using shell vial technique as qPCR was performed to measure the Wolbachia density in the
previously described [25] with a slight modification. In brief, two cell lines and mosquitoes [31]. Genomic DNA was extracted from

www.plosntds.org 2 July 2012 | Volume 6 | Issue 7 | e1754

 Wolbachia Density-Dependent Viral Interference

either Aa23 cells (two to six biological replicates) or mosquito samples were incubated with DAPI for 10 min. Immunostaining
tissues (ten biological replicates). The primers specifically directed was examined under with an Olymupus Fluoview 1000 confocal
to Wolbachia surface protein (wsp) of wAlbA and wAlbB were used microscope.
in PCR to measure the Wolbachia genome copy, which was
normalized with the host actin for Aa23 cells [31] or ribosomal Results
protein S6 (RPS6) gene for both Ae. albopictus and Ae. aegypti
mosquitoes. RPS6 was amplified with the primers: For (59- The Wolbachia wAlbB induce strong resistance to DENV-
GAAGTTGAACGTATCGTTTC-39) and Rev (59-GAGATG- 2 in Ae. albopictus Aa23 cells
GTCAGCGGTGATTT-39). A standard curve was generated for We previously reported that the Wolbachia wAlbB does not
each of wAlbA, wAlbB and actin by analyzing 101 to 108 copies/ inhibit dissemination of DENV-2 to the mosquito head in the
reaction of the pQuantAlb plasmid [31]. Another plasmid naturally infected Ae. albopictus [12]. This could indicate that
containing the RPS6 fragment was cloned to generate another wAlbB does not induce resistance to DENV in the host genetic
standard curve for RPS6. background of Ae. albopictus. To test that possibility, we compared
susceptibility for DENV-2 between wAlbB-infected Aa23 cell line
Susceptibility of the host cells to DENV infection and its aposymbiotic Aa23T. Seven day after viral infection, the
The susceptibility of Aa23, Aa23T, w-Aag2 and Aag2 cell lines viral titer in Aa23T reached to 1.16107 PFU/ml while no virus
for DENV-2 was measured by plaque assay with a slight particle was detected in Aa23 cell line (Fig. 1A). A parallel assay
modification [29]. DENV-2 virus stock (26107 PFU/ml; deter- was conducted to measure the Wolbachia-mediated viral interfer-
mined previously in C6/36 cells) was serially diluted in ten-fold ence in Ae. aegypti cell line Aag2. Like Ae. aegypti mosquito, Aag2 cell
increments (from 101 to 106) using cell medium, and then line is uninfected by Wolbachia. We introduced wAlbB from Aa23
inoculated into each of the four mosquito cells in 24-well plates. into Aag2 through a shell-vial technique, generating a Wolbachia-
After incubation at 26uC for 5 days, the plates were assayed for infected cell line w-Aag2. As expected, viral titer in the Aag2 cell
plaque formation by peroxidase immunostaining, using mouse line (1.36106 PFU/ml) is significantly higher than in the w-Aag2
hyperimmune ascitic fluid and a goat anti-mouse HRP conjugate cell line (12.5 PFU/ml; P,0.01, Student’s t-Test; Fig. 1B). These
as the primary and secondary antibody. At least five biological results confirm that wAlbB can induce resistance to DENV-2 even
replicates were used for each treatment. in Ae. albopictus host background.

Immunofluorescence Generation of Aa23 cell lines infected with Wolbachia at

The viral antigen and Wolbachia wsp protein in Aa23 cells were different densities
detected simultaneously using an indirect immunofluorescence To study how the density of wAlbB can influence its induced
assay (IFA). Confluent rifampicin -treated Aa23 cells were grown resistance to DENV in Aa23 cell line, we used sub-lethal doses of
on chamber slides (Thermo), and then were infected with DENV- rifampicin to partially remove Wolbachia and generated cell
2 at 0.1 MOI. After incubation for 2 days, cells were fixed for cultures with Wolbachia at different densities. Three dosages
30 min at 4uC in 4% (w/v) paraformaldehyde in PBS, containing (0.05 mg/ml, 0.5 mg/ml and 5 mg/ml) with four treatment times
0.5% (v/v) Triton X-100. Then, samples were incubated in PBS (4 h, 10 h, 40 h and 70 h), resulting in a total of 12 treatments,
and 10% (v/v) goat serum at room temperature for 1 h. were designed to make Wolbachia density cover a broad range. A
Subsequently, the slides were incubated with a mouse anti-dengue mock treatment (control) was included to treat cells with only
complex monoclonal antibody at a 1:300 dilution and a rabbit methanol solvent for the above four time periods. As a result
anti-wsp polyclonal antibody (GenScript) at a 1:500 dilution in (Fig. 2), we generated a serial of cell cultures with different
PBS at room temperature for 1 h. After wash, the slides were Wolbachia densities, with a highest at 953.2 wsp/actin (from
incubated with Alexa-488 anti-mouse and Alexa-594 anti-rabbit 0.05 mg/ml and 4 h) and a lowest at 18.0 wsp/actin (from 5 mg/
antibodies (Molecular Probes, Invitrogen) at a 1:1000 dilution in ml and 70 h). Wolbachia density significantly decreased with an
PBS at room temperature for 1 h. To stain cell nucleus, the increase in treatment dose. Within each dosage, no significant

Figure 1. wAlbB induces a strong resistance to DENV-2 in mosquito cells. wAlbB is a native infection in Ae. albopictus Aa23 cells (A) while
Aa23T cells were initially generated by tetracycline treatment of Aa23 cell to remove Wolbachia infection. There is no Wolbachia in Ae. aegypti Aag2
cells (B). w-Aag2 was generated from Aag2 cells by introducing wAlbB from Aa23 using a shell-vial technique. Five days after inoculated with DENV-2,
the cells were tested for dengue infection by plaque assay. Error bars are standard errors of the mean of at least three biological replicates. **, P,0.01;
***, P,0.001in Student’s t-Test.
doi:10.1371/journal.pntd.0001754.g001

www.plosntds.org 3 July 2012 | Volume 6 | Issue 7 | e1754

 Wolbachia Density-Dependent Viral Interference

Figure 2. Generation of Aa23 cells with different Wolbachia

density. Cells were treated using sub-lethal doses of rifampicin for a
different time periods. Three dosages (0.05 mg/ml, 0.5 mg/ml and 5 mg/ Figure 3. Wolbachia induces density-dependent inhibition to
ml) and four time periods (4 h, 10 h, 40 h and 70 h) were used. The DENV-2 in Aa23 cell lines. Cell cultures derived from twelve different
genome copies of wsp were measure by q-PCR, normalized by host treatments in Fig. 2 were used in the assay. Five days after these cells
gene actin. Error bars are standard errors of the mean of at least three were infected with DENV-2, viral genome copies were measured by
biological replicates. Statistical significance is represented by letters qRT-PCR. Actin was used as a host gene to normalize the data. There is
above each column, with different letters signifying distinct statistical a negative linear correlation between Wolbachia density and DENV
groups. Student’s t test: a vs. b, P,0.001; b vs. c, P,0.001; d vs. a, copy. Each point is the mean of at least three biological replicates.
P,0.05. doi:10.1371/journal.pntd.0001754.g003
doi:10.1371/journal.pntd.0001754.g002

difference in Wolbachia density is observed among four treatment

times. This indicates a treatment for 4 h can effectively remove
Wolbachia and a further increase in treatment time will not
significantly enhance the inhibitory effect of rifampicin on
Wolbachia. For example, a treatment at dosage of 5 mg/ml for
4 h can dramatically reduce Wolbachia to a very low level (75.2
wsp/actin), compared to the mock treatment (1,888.3 wsp/actin).
Further increasing treatment time at this dose will not significantly
change the Wolbachia density, resulting in 74.1 wsp/actin (10 h),
52.4 wsp/actin (40 h), and 18.0 wsp/actin (70 h, Fig. 2).

The Wolbachia wAlbB induces resistance to DENV in a

density-dependent manner
To study the relationship between Wolbachia density and dengue
infection level, cell cultures derived from the above 12 treatments
were inoculated with DENV-2. Five days after infection, we
measured genome copies of DENV-2 in the cells. Wolbachia density
in the cells and its corresponding viral infection level were
recorded in parallel to examine their interactions. We observed
that genome copy of DENV-2 in cells increased with a decrease in
Wolbachia density. Moreover, there is a negative correlation
between the genome copy of Wolbachia and DENV-2 (r = 20.82;
P,0.001). The level of DENV-2 (y) can be predicted by the
genome copy of Wolbachia (x) via a model: y = 20.004x+3.847
(Fig. 3). These results confirm that Wolbachia induces a density-
dependent viral inhibition in Aa23 cells.

Wolbachia density in somatic tissues of Ae. albopictus is

too low to induce viral interference Figure 4. Wolbachia density in somatic tissues of Ae. albopictus is
Within mosquito vectors, DENV needs replicate and pass too low to induce resistance to DENV. (A). The density of wAlbA is
through midguts and salivary glands before transmission occurs. significantly lower than wAlbB in midgut, fatbody, salivary gland and
ovary of Ae. albopictus. The copy number of the Wolbachia wsp was
Wolbachia-mediated density-dependent viral inhibition in mosquito normalized by Ae. albopictus actin; (B). Wolbachia density in somatic
cells leads to a hypothesis that the Wolbachia density in the somatic tissues is significantly lower in the Ae. albopictus HOU strain than in the
tissues of Ae. albopictus is too low to induce viral interference in this transinfected Ae. aegypti WB1 strain. The copy number of the Wolbachia
mosquito species. Thus, we measured Wolbachia density in both wsp was normalized by one conserved RPS6 in both Ae. albopictus and
somatic tissues (midgut, salivary gland, fatbody) and germ line Ae. aegypti. In all the assays, midguts, salivary glands, fatbodies and
ovaries of 7-day-old non blood fed females were dissected and used for
tissue (ovary) of the HOU strain of Ae. albopictus (carrying wAlbA
extraction of total genomic DNA. ***, P,0.001; **, P,0.01 in Student’s
and wAlbB). As shown in Fig. 4A, wAlbB is significantly higher t-Test. Error bars are standard errors of the mean of ten biological
than wAlbA in all the four tissues. While none of wAlbA could be replicates.
detected in midgut, salivary gland and fatbody, only 5.4 wsp/actin doi:10.1371/journal.pntd.0001754.g004

www.plosntds.org 4 July 2012 | Volume 6 | Issue 7 | e1754

 Wolbachia Density-Dependent Viral Interference

of wAlbA presents in ovary. There are 0.3, 5.3, and 12.3 wsp/ dependent viral inhibition in Aa23 cells relates to host immunity,
actin of wAlbB in midgut, salivary gland and fatbody, respectively, we measured both Wolbachia density and the expression of DEFD
compared to 57.9 wsp/actin of wAlbB in ovary. in the antibiotic treated Aa23 cells. As shown in Fig. 5, there is a
To better compare Wolbachia density between the HOU strain positive correlation between Wolbachia density and the amount of
of Ae. albopictus and the transinfected Ae. aegypti WB1 strain DEFD transcript (r = 0.92; P = 0.0001). This supports that host
(carrying wAlbB), we designed one set of primers that can amplify immune level could contribute to Wolbachia-density dependent
a conserved region of Ribosomal protein S6 (RPS6) in both viral inhibition.
mosquito species, which was used to normalize the Wolbachia copy
in qRT-PCR. We observed that wAlbB density is significantly Cells host both Wolbachia and DENV when Wolbachia
lower in the midgut than in the ovary and salivary gland for both density is low
mosquito species (P,0.05, Student’s t-Test). However, wAlbB To study whether the above antibiotic treatment leads to a
density in fatbody is similar to that in salivary gland. Importantly, uniform change in both Wolbachia and DENV at the cellular level,
a significant lower density of wAlbB presents in midgut, fatbody cells at two days post-DENV infection were assayed by double
and salivary gland (P,0.001, Student’s t-Test) of HOU mosqui- immunofluorescent staining to visualize the distribution of
toes as compared with that of WB1 mosquitoes. Specifically, the Wolbachia and DENV-2 using antibodies against the wsp of
Wolbachia genome copies in midgut, fatbody and salivary gland of Wolbachia and envelop protein of DENV-2. As expected, we
HOU are 80-, 18-, and 24-fold less than that of WB1, respectively observed Wolbachia was largely removed by rifampicin treatment,
(Fig. 4B). In ovary, wAlbB density is also 5.6-fold lower in HOU resulting in more cells infected by DENV-2 (Fig. 6A–C). However,
than in WB1 (P,0.001, Mann-Whiteney U test). such a reduction in Wolbachia density appears not to be even,
To confirm a lack of Wolbachia-mediated viral interference in Ae. resulting in that certain cells host much more Wolbachia than the
albopictus, we compared mosquito susceptibility for DENV-2 at a others (Fig. 6A–C). Importantly, we observed that colocalization of
serial diluted viral titers between the HOU and its aposymbiotic DENV-2 and Wolbachia can occur in the same cells when Wolbachia
strain HT1 (initially derived from Houston strain through density is low in those cells (Fig. 6D–E). For those cells that were
tetracycline treatment). We have previously observed that heavily infected by DENV-2, however, they typically do not
Wolbachia suppresses viral dissemination in the HOU mosquitoes contain Wolbachia (Fig. 6G–I).
after mosquitoes took infectious blood meal at a viral titer of 107
PFU/ml [12]. However, it is possible that Wolbachia induces only a Discussion
weak resistance to DENV in Ae. albopictus which could be
overcome when a very high viral titer was used in the infection Wolbachia can induce a resistance to DENV and other
assay. As shown in Table 1, Wolbachia does not inhibit viral arboviruses in Ae. aegypti [12,13]. This raises a puzzle as to why
dissemination even mosquitoes took blood with a low viral titers. mosquitoes that naturally carry Wolbachia, such as Ae. albopictus, can
The head infection rate of both HOU and HT1 strain was still serve as vectors for the arboviruses. Here, we demonstrate that
dropped dramatically when mosquitoes fed with dengue-infected wAlbB is able to induce resistance to DENV in Ae. albopictus Aa23
blood at diluted viral titers. No significant difference in head cell line, and such viral interference occurs in Wolbachia density-
infection rate was observed between HOU and HT1 in all the dependent manner. We also show that Wolbachia density in
three diluted viral titers (P.0.05, Fisher’s exact test). These results somatic tissues of Ae. albopictus is significantly lower than the
confirm that Wolbachia does not suppress viral dissemination in the transinfected Ae. aegypti WB1. Such a low density of Wolbachia is
naturally infected Ae. albopictus. predicted not to induce viral interference in Ae. albopictus.
Consistently, Wolbachia does not suppress viral dissemination in
Wolbachia induces expression of antimicrobial peptide Ae. albopictus even when mosquitoes take an infectious blood meal
at a very low viral titer. Moreover, we observed a positive
DEFD in a density-dependent manner in Ae. albopictus
correlation between Wolbachia density and the expression of the
cells
We previously show that the boosted host immunity boosted
plays important roles in Wolbachia-mediated viral inference and
antimicrobial peptide defensin is involved in this anti-dengue
resistance [17]. To test whether the above Wolbachia-density

Table 1. The native Wolbachia does not inhibit dissemination

of DENV-2 to mosquito heads in Ae. albopictus.

Head infection frequency, %

PFU/ml Houston HT1

6
10 61.1 (11/18) 50.0 (9/18)
Figure 5. The expression of DEFD was induced by Wolbachia in
105 38.9 (7/18) 38.9 (7/18) density-dependent manner in Aa23 cell lines. Seven days after
104 22.2 (4/18) 16.7 (3/18) rifampicin treatment, cells were collected to measure Wolbachia
densities and DEFD expression. Because treatment with 5 mg/ml of
The HOU strain of Ae. albopictus and its aposymbiotic strain HT1 were infected rifampicin for 70 h resulted in both Wolbachia infection and DEFD
with the blood containing DENV-2 at titers of 106, 105, and 104 PFU/ml. At Day expression at lowest level, it serves as a reference for all the other
14 post infection, heads of ten mosquitoes were collected and used for treatments to calculate a fold increase in both Wolbachia density and
diagnosis of DENV-2 by RT-PCR. Data from two experiments were pooled DEFD expression. Each point is the mean of at least three biological
together. replicates.
doi:10.1371/journal.pntd.0001754.t001 doi:10.1371/journal.pntd.0001754.g005

www.plosntds.org 5 July 2012 | Volume 6 | Issue 7 | e1754

 Wolbachia Density-Dependent Viral Interference

Figure 6. Localization of Wolbachia and DENV-2 in Aa23 cells. Double immunofluorescence staining of cells showing the localization of
dengue virus (green) and Wolbachia (red). Cells were probed simultaneously with polyclonal anti-wsp antibody (Wolbachia) and monoclonal anti-
DENV-2 antibody, followed by Alexa 594 (red) and Alexa 488 (green) conjugated antibodies, respectively. DNA (blue) is stained with DAPI. In panels
(A, B, and C), the red, green and blue channels are merged. A and B show Aa23 cells with mock treatment and treatment with 5 mg/ml of rifampicin
for 5 hr, respectively, followed by dengue infection. C is Aa23T cells without dengue infection (negative control). D to F or G to I is the same sample
with different channel merged: D and G show only red and blue channel merged, E and H show only green and blue channel merged, F and I show all
the red, green and blue merged. Aa23 cells treated with 5 mg/ml of rifampicin for 10 hr (D to F) and 40 hr (G to I), followed by dengue infection, are
shown.
doi:10.1371/journal.pntd.0001754.g006

antimicrobial peptide DEFD. When Wolbachia density is low, are the two major sites for DENV to replicate and migrate through
DENV can colocalize with Wolbachia in the same cell hosts. in order for mosquitoes to be infectious, and fatbody is the main
Consistent with previous studies, Wolbachia distributes in both immune organ to defend against foreign invaders [32]. Our results
somatic and germ line tissues in Ae. albopictus [21]. We further show indicate that the Wolbachia infection in midgut and fatbody of Ae.
that only wAlbB presents in midgut, salivary gland and fatbody albopictus may be too low to suppress viral replication in midgut
while wAlbA was detected in only ovary. In this work, we focus on and the subsequent dissemination into the other parts of mosquito
the above three somatic tissues because midgut and salivary gland body. This is supported by the lack of difference in the head

www.plosntds.org 6 July 2012 | Volume 6 | Issue 7 | e1754

 Wolbachia Density-Dependent Viral Interference

infection rate between the HOU and HT1 strain. Comparison of proteins that are required for DENV-2 propagation in cell lines [38].
wAlB infection between the dengue resistant WB1 and the Evidence indicates that these host proteins form dengue-associated
susceptible HOU strain shows that there is a general reduction protein interaction networks, consisting of highly interconnected
in Wolbachia density in HOU strain and the magnitude of proteins with closely related functions in each of replication/
reduction in midgut is more than the other tissues. transcription/translation, immunity, transport and metabolism [39].
We observe a linear negative correlation between Wolbachia Wolbachia could perturb the above network to influence dengue
density and genome copy of DENV-2 in Aa23 cell line. Based on the infection. The cell culture provides us an ideal tool to study the
model (y = 20.004x+3.847), Wolbachia density at 961.8 wsp/actin is interaction of Wolbachia with these dengue host proteins/networks.
required to completely clean DENV in Aa23 cells. With the In the field, a difference in Wolbachia density in mosquito vectors
observed Wolbachia copy in midgut (0.3 wsp/actin) and salivary gland may directly influence the disease transmission. Wolbachia density
(5.3 wsp/actin) of Ae. albopictus HOU strain, Wolbachia would have was reported to differ largely among Ae. albopictus collected from the
no inhibition to DENV-2 in midgut and reduced DENV-2 by 0.6% field in different locations in Thailand [40]. It is still unknown to
in salivary gland in this mosquito species. However, it is possible that what extent this will affect the vector status and vector competence
our model will be more appropriate to be used in the in vitro system. of Ae. albopictus. However, conflicting results were reported on the
Wolbachia strains, host cells or tissues and environment may influence relative susceptibility of Ae. albopictus versus Ae. aegypti to oral DENV
the viral inhibition, resulting in a modified linear correlation. Even infection [22,41]. Some studies show that Ae. albopictus is more
that, a general pattern of Wolbachia density dependent viral inhibition susceptible to DENV than Ae. aegypti, while the others show opposite
should present in different systems. When wMelPop-CLA was results [41,42,43,44,45]. It was also reported a significant increase of
introduced into two Ae. albopictus cell lines, RML12 and C6/36, the susceptibility to DENV in Ae. albopictus with increasing generations
cell line with a high Wolbachia density (C6/36) shows a strong in the laboratory [41,45]. Although generally considered a
inhibition to DENV-2 [20]. In D. simulans, Wolbachia strains that secondary vector of dengue, Ae. albopictus was the primary cause of
grow to high density provide the highest protection from virus dengue epidemics in a number of areas [22,46,47,48,49]. In China,
infection [33]. All these indicate that Wolbachia density-dependent Ae. albopictus has long been considered the primary dengue vector.
viral inhibition occurs for both native and non-native Wolbachia and Dengue epidemics frequently occurred without Ae. aegypti, with
cross different host cell types. 22,122 new infections, including the fatal dengue hemorrhagic
The linear negative correlation between Wolbachia density and fever, recorded in single year [22,50,51,52]. All the above suggest
dengue copy could lead to a simple hypothesis, in which Wolbachia vector competence and vector status of Ae. albopictus can vary in
secretes antimicrobial effector molecules into a host cell to directly different locations. Future work should study the impacts of
inhibit pathogens. Such effectors should have a broad-spectrum Wolbachia on vector status and vector competence of Ae. albopictus.
activity against a variety of pathogens, including virus, plasmodi- Our results could provide important implications for the future
um and worm. Consistently, a bacterium was recently reported to vector-borne disease control. First, novel control strategies can be
produce ROS and induce resistance to malaria parasites in developed to manually increase Wolbachia density in mosquitoes
Anopheles gambiae mosquitoes [34]. The Type IV Secretion System, naturally carrying this bacterium. This could lead to blocking
which presents within Wolbachia [35,36], was reported to secret
pathogen transmission to human but without a need to eradicate
different factors into host to mediate interactions between the
these mosquito species. Second, it is still unknown whether
other intracellular bacteria and their hosts [37]. Our preliminary
Wolbachia can persistently maintain infection at a level that
studies show that induction of DEFD expression occurs in
effectively induces complete resistance to DENV in the transin-
Wolbachia-density dependent manner, providing us support to
fected Ae. aegypti. We could not exclude the possibility that
further test this hypothesis.
adaption of a recent Wolbachia to a new host will finally lead to a
Currently there are two major modes to explain the mecha-
low Wolbachia density, resulting in that what we see today in Ae.
nisms of Wolbachia-mediated viral interference [16,19]. The first is
albopictus will be what happens tomorrow in those transinfected Ae.
Wolbachia boosts mosquito basal immunity against viral infection,
aegypti. Understanding how Wolbachia density is regulated by
while the second is that Wolbachia and virus compete for the same
mosquito hosts and how the Wolbachia machinery controls its
host resource. It appears that Wolbachia density dependent viral
replication will facilitate the current effort to eliminate dengue
inhibition could be explained by both modes. In the first mode, a
through Wolbachia-based population replacement [53].
high Wolbachia density could cause high oxidative stress in host,
leading to production of more ROS for a local antiviral immune
response. The low Wolbachia density in somatic tissues of Ae. Acknowledgments
albopictus is consistent with the previous observation that Wolbachia We thank S. L. Dobson for providing us Ae. albopictus HOU and HT1
neither induces nor suppresses transcripts encoding antimicrobial strains, and cell lines Aa23 and Aa23T, R. Palli for providing us Aag-2 cell
peptides in Ae. albopictus [18]. In the second mode, more Wolbachia line, M. Weill for providing the pQuantAlb plasmid, and the Arbovirus
could lead to less host resource for virus. While it appears to be Diseases Branch at the Centers for Disease Control for providing the anti-
straightforward for the second mode to fit the density-dependent DENV-2 antibodies. We also thank F. Zhang, G. Zhou and M. Mcfadden
data, it will be challenging to identify the common host for their assistance with experiments.
components needed by both Wolbachia and a variety of pathogens.
Our results indicate cell lines could be used as a simple in vitro Author Contributions
system to study the mechanism of Wolbachia-mediated viral Conceived and designed the experiments: PL ZX. Performed the
interference. DENV was strongly inhibited in Wolbachia-infected experiments: PL GB XP. Analyzed the data: PL ZX. Contributed
Aa23 and Aag-2 cells. Recent studies identified hundreds of host reagents/materials/analysis tools: PL ZX. Wrote the paper: PL ZX.

References
1. Gubler DJ, Kuno G (1997) Dengue and Dengue Hemorrhagic Fever. New York 2. Iturbe-Ormaetxe I, Walker T, O’Neill SL (2011) Wolbachia and the biological
Cab International. control of mosquito-borne disease. EMBO Rep 12: 508–518.

www.plosntds.org 7 July 2012 | Volume 6 | Issue 7 | e1754

 Wolbachia Density-Dependent Viral Interference

3. Werren JH, Baldo L, Clark ME (2008) Wolbachia: master manipulators of 29. Das S, Garver L, Ramirez JR, Xi Z, Dimopoulos G (2007) Protocol for Dengue
invertebrate biology. Nat Rev Microbiol 6: 741–751. Infections in Mosquitoes (A. aegypti) and Infection Phenotype Determination.
4. Kittayapong P, Baisley KJ, Baimai V, O’Neill SL (2000) Distribution and J Vis Exp: 220.
diversity of Wolbachia infections in Southeast Asian mosquitoes (Diptera: 30. Molina-Cruz A, Gupta L, Richardson J, Bennett K, Black Wt, et al. (2005) Effect
Culicidae). Journal of medical entomology 37: 340–345. of mosquito midgut trypsin activity on dengue-2 virus infection and
5. Serbus LR, Casper-Lindley C, Landmann F, Sullivan W (2008) The genetics dissemination in Aedes aegypti. Am J Trop Med Hyg 72: 631–637.
and cell biology of Wolbachia-host interactions. Annu Rev Genet 42: 683–707. 31. Tortosa P, Courtiol A, Moutailler S, Failloux A, Weill M (2008) Chikungunya-
6. Turelli M, Hoffmann AA (1991) Rapid spread of an inherited incompatibility Wolbachia interplay in Aedes albopictus. Insect Molecular Biology 17: 677–684.
factor in California Drosophila. Nature 353: 440–442. 32. Beerntsen BT, James AA, Christensen BM (2000) Genetics of mosquito vector
7. Hoffmann AA, Montgomery BL, Popovici J, Iturbe-Ormaetxe I, Johnson PH, competence. Microbiol Mol Biol Rev 64: 115–137.
et al. (2011) Successful establishment of Wolbachia in Aedes populations to 33. Osborne SE, Leong YS, O’Neill SL, Johnson KN (2009) Variation in antiviral
suppress dengue transmission. Nature 476: 454–457. protection mediated by different Wolbachia strains in Drosophila simulans. PLoS
8. Xi Z, Khoo CC, Dobson SL (2005) Wolbachia establishment and invasion in an Pathog 5: e1000656.
Aedes aegypti laboratory population. Science 310: 326–328. 34. Cirimotich CM, Dong Y, Clayton AM, Sandiford SL, Souza-Neto JA, et al.
9. Walker T, Johnson PH, Moreira LA, Iturbe-Ormaetxe I, Frentiu FD, et al. (2011) Natural microbe-mediated refractoriness to Plasmodium infection in
(2011) The wMel Wolbachia strain blocks dengue and invades caged Aedes
Anopheles gambiae. Science 332: 855–858.
aegypti populations. Nature 476: 450–453.
35. Masui S, Sasaki T, Ishikawa H (2000) Genes for the type IV secretion system in
10. Hughes GL, Koga R, Xue P, Fukatsu T, Rasgon JL (2011) Wolbachia infections
an intracellular symbiont, Wolbachia, a causative agent of various sexual
are virulent and inhibit the human malaria parasite Plasmodium falciparum in
alterations in arthropods. Journal of bacteriology 182: 6529–6531.
anopheles gambiae. PLoS Pathog 7: e1002043.
11. Kambris Z, Blagborough AM, Pinto SB, Blagrove MS, Godfray HC, et al. 36. Wu M, Sun LV, Vamathevan J, Riegler M, Deboy R, et al. (2004)
(2010) Wolbachia stimulates immune gene expression and inhibits plasmodium Phylogenomics of the reproductive parasite Wolbachia pipientis wMel: a
development in Anopheles gambiae. PLoS Pathog 6. streamlined genome overrun by mobile genetic elements. PLoS biology 2: E69.
12. Bian G, Xu Y, Lu P, Xie Y, Xi Z (2010) The endosymbiotic bacterium Wolbachia 37. Stein M, Rappuoli R, Covacci A (2000) Tyrosine phosphorylation of the
induces resistance to dengue virus in Aedes aegypti. PLoS Pathog 6: e1000833. Helicobacter pylori CagA antigen after cag-driven host cell translocation. Proc
13. Moreira LA, Iturbe-Ormaetxe I, Jeffery JA, Lu G, Pyke AT, et al. (2009) A Natl Acad Sci USA 97: 1263–1268.
Wolbachia symbiont in Aedes aegypti limits infection with dengue, Chikungunya, 38. Sessions OM, Barrows NJ, Souza-Neto JA, Robinson TJ, Hershey CL, et al.
and Plasmodium. Cell 139: 1268–1278. (2009) Discovery of insect and human dengue virus host factors. Nature 458:
14. Kambris Z, Cook PE, Phuc HK, Sinkins SP (2009) Immune activation by life- 1047–1050.
shortening Wolbachia and reduced filarial competence in mosquitoes. Science 39. Guo X, Xu Y, Bian G, Pike AD, Xie Y, et al. (2010) Response of the mosquito
326: 134–136. protein interaction network to dengue infection. BMC Genomics 11: 380.
15. Hedges LM, Brownlie JC, O’Neill SL, Johnson KN (2008) Wolbachia and virus 40. Ahantarig A, Trinachartvanit W, Kittayapong P (2008) Relative Wolbachia
protection in insects. Science 322: 702. density of field-collected Aedes albopictus mosquitoes in Thailand. J Vector Ecol
16. Teixeira L, Ferreira A, Ashburner M (2008) The bacterial symbiont Wolbachia 33: 173–177.
induces resistance to RNA viral infections in Drosophila melanogaster. PLoS 41. Lambrechts L, Scott TW, Gubler DJ (2010) Consequences of the expanding
Biol 6: e2. global distribution of Aedes albopictus for dengue virus transmission. PLoS Negl
17. Pan X, Zhou G, Wu J, Bian G, Lu P, et al. (2012) Wolbachia induces reactive Trop Dis 4: e646.
oxygen species (ROS)-dependent activation of the Toll pathway to control 42. Rosen L, Roseboom LE, Gubler DJ, Lien JC, Chaniotis BN (1985) Comparative
dengue virus in the mosquito Aedes aegypti. Proc Natl Acad Sci U S A 109: susceptibility of mosquito species and strains to oral and parenteral infection with
E23–31. dengue and Japanese encephalitis viruses. Am J Trop Med Hyg 34: 603–615.
18. Bourtzis K, Pettigrew MM, O’Neill SL (2000) Wolbachia neither induces nor 43. Chen WJ, Wei HL, Hsu EL, Chen ER (1993) Vector competence of Aedes
suppresses transcripts encoding antimicrobial peptides. Insect molecular biology albopictus and Ae. aegypti (Diptera: Culicidae) to dengue 1 virus on Taiwan:
9: 635–639. development of the virus in orally and parenterally infected mosquitoes. J Med
19. Rances E, Ye YH, Woolfit M, McGraw EA, O’Neill SL (2012) The relative Entomol 30: 524–530.
importance of innate immune priming in Wolbachia-mediated dengue interfer- 44. Jumali Sunarto, Gubler DJ, Nalim S, Eram S, et al. (1979) Epidemic dengue
ence. PLoS Pathog 8: e1002548. hemorrhagic fever in rural Indonesia. III. Entomological studies. Am J Trop
20. Frentiu FD, Robinson J, Young PR, McGraw EA, O’Neill SL (2010) Wolbachia- Med Hyg 28: 717–724.
mediated resistance to dengue virus infection and death at the cellular level. 45. Vazeille M, Rosen L, Mousson L, Failloux AB (2003) Low oral receptivity for
PLoS One 5: e13398. dengue type 2 viruses of Aedes albopictus from Southeast Asia compared with
21. Dobson SL, Bourtzis K, Braig HR, Jones BF, Zhou W, et al. (1999) Wolbachia that of Aedes aegypti. Am J Trop Med Hyg 68: 203–208.
infections are distributed throughout insect somatic and germ line tissues. Insect 46. Hotta S (1998) Dengue vector mosquitoes in Japan: the role of Aedes albopictus
Biochem Mol Biol 29: 153–160.
and Aedes aegypti in the 1942–44 dengue epidemics of Japanese Main Islands.
22. Gratz NG (2004) Critical review of the vector status of Aedes albopictus. Med
Medical Entomology and Zoology 49: 276–274.
Vet Entomol 18: 215–227.
47. Kobayashi M, Nihei N, Kurihara T (2002) Analysis of northern distribution of
23. Blagrove MS, Arias-Goeta C, Failloux AB, Sinkins SP (2012) Wolbachia strain
Aedes albopictus (Diptera: Culicidae) in Japan by geographical information system.
wMel induces cytoplasmic incompatibility and blocks dengue transmission in
J Med Entomol 39: 4–11.
Aedes albopictus. Proc Natl Acad Sci U S A 109: 255–260.
24. O’Neill SL, Pettigrew MM, Sinkins SP, Braig HR, Andreadis TG, et al. (1997) In 48. Metselaar D, Grainger CR, Oei KG, Reynolds DG, Pudney M, et al. (1980) An
vitro cultivation of Wolbachia pipientis in an Aedes albopictus cell line. Insect outbreak of type 2 dengue fever in the Seychelles, probably transmitted by Aedes
molecular biology 6: 33–39. albopictus (Skuse). Bull World Health Organ 58: 937–943.
25. Dobson SL, Marsland EJ, Veneti Z, Bourtzis K, O’Neill SL (2002) 49. Effler PV, Pang L, Kitsutani P, Vorndam V, Nakata M, et al. (2005) Dengue
Characterization of Wolbachia host cell range via the in vitro establishment of fever, Hawaii, 2001–2002. Emerg Infect Dis 11: 742–749.
infections. Appl Environ Microbiol 68: 656–660. 50. Kan B (1997) Opinion on dengue endemics and its surveillance in China.
26. Sim S, Dimopoulos G (2010) Dengue virus inhibits immune responses in Aedes Chinese journal of vector biology and control 8: 152–156.
aegypti cells. PLoS One 5: e10678. 51. Yan Z, Hu Z, Jiang Y (2008) Dengue vector; Zhang F, Yang Z, editors. Beijing:
27. Gao Y, Hernandez VP, Fallon AM (1999) Immunity proteins from mosquito cell Science Press. 293 p.
lines include three defensin A isoforms from Aedes aegypti and a defensin D 52. Jin LQ, Li D (2008) A recent survey of mosquito fauna in Guangdong
from Aedes albopictus. Insect Mol Biol 8: 311–318. Province,southern China, with a review of past records [corrected]. Med Vet
28. Troyer JM, Hanley KA, Whitehead SS, Strickman D, Karron RA, et al. (2001) Entomol 22: 359–363.
A live attenuated recombinant dengue-4 virus vaccine candidate with restricted 53. Cook PE, McMeniman CJ, O’ Neill SL (2007) Modifying Insect Population Age
capacity for dissemination in mosquitoes and lack of transmission from vaccinees Structure to Control Vector-Borne Disease; Aksoy S, editor. Austin: Landes
to mosquitoes. Am J Trop Med Hyg 65: 414–419. Bioscience 18060–18065.

www.plosntds.org 8 July 2012 | Volume 6 | Issue 7 | e1754