Identification of Spiroplasma insolitum symbionts in Anopheles gambiae

Background: Insect symbionts have the potential to block the transmission of vector-borne diseases by their hosts. The advancement of a symbiont-based transmission blocking strategy for malaria requires the identification and study of Anopheles symbionts. Methods: High throughput 16S amplicon sequencing was used to profile the bacteria associated with Anopheles gambiae sensu lato and identify potential symbionts. The polymerase chain reaction (PCR) with specific primers were subsequently used to monitor symbiont prevalence in field populations, as well as symbiont transmission patterns. Results: We report the discovery of the bacterial symbiont, Spiroplasma, in Anopheles gambiae in Kenya. We determine that geographically dispersed Anopheles gambiae populations in Kenya are infected with Spiroplasma at low prevalence levels. Molecular phylogenetics indicates that this Anopheles gambiae associated Spiroplasma is a member of the insolitum clade. We demonstrate that this symbiont is stably maternally transmitted across at least two generations and does not significantly affect the fecundity or egg to adult survival of its host. Conclusions: In diverse insect species, Spiroplasma has been found to render their host resistant to infection by pathogens. The identification of a maternally transmitted strain of Spiroplasma in Anopheles gambiae may therefore open new lines of investigation for the development of symbiont-based strategies for blocking malaria transmission.


Abstract
: Insect symbionts have the potential to block the transmission of Background vector-borne diseases by their hosts. The advancement of a symbiont-based transmission blocking strategy for malaria requires the identification and study of symbionts.

Anopheles
: High throughput 16S amplicon sequencing was used to profile the Methods bacteria associated with and identify potential Anopheles gambiae sensu lato symbionts. The polymerase chain reaction (PCR) with specific primers were subsequently used to monitor symbiont prevalence in field populations, as well as symbiont transmission patterns.
: We report the discovery of the bacterial symbiont, , in Results Spiroplasma in Kenya. We determine that geographically dispersed Anopheles gambiae populations in Kenya are infected with at low Anopheles gambiae Spiroplasma prevalence levels. Molecular phylogenetics indicates that this Anopheles associated is a member of the clade. We gambiae Spiroplasma insolitum demonstrate that this symbiont is stably maternally transmitted across at least two generations and does not significantly affect the fecundity or egg to adult survival of its host.
: In diverse insect species, has been found to render

Introduction
Malaria remains a major health problem in many developing countries, particularly in Sub-Saharan Africa (WHO, 2015). Malaria transmission dynamics are dependent on aspects of the physiology and ecology of their vectors, Anopheles mosquitoes. Historically, the most successful malaria interventions have been aimed at controlling the vector to break the disease transmission cycle (Raghavendra et al., 2011). The wide distribution of insecticide treated bednets (ITNs) has had a significant impact on reducing the number of malaria cases over the past 15 years, accounting for more than 50% of the malaria deaths averted in this period (Bhatt et al., 2015). However, vector resistance to insecticides used in ITNs is spreading rapidly and there are clear signs of behavioral resistance; mosquitoes that formerly bit indoors are now biting outdoors where nets offer no protection (Kabbale et al., 2013). This may reverse significant reductions in malaria disease burden and therefore new strategies are desperately needed to control mosquito populations or their capacity to transmit parasites. One of the most promising new tools for controlling vector borne diseases involves bacterial symbionts that decrease the vectorial capacity of their insect hosts (Iturbe-Ormaetxe et al., 2011). These symbionts are maintained in host populations through maternal transmission and can spread through insect populations. These features render them potentially a much more sustainable and cost-effective strategy for the control of vector-borne disease transmission than conventional methods (McGraw & O'Neill, 2013).
In the last decade, there have been many significant advances in the development of symbiont-based strategies for arboviral disease control, primarily centred on the bacterial symbiont, Wolbachia (Jeffries & Walker, 2016). Wolbachia can be transinfected into Ae. aegypti and Ae. albopictus where it blocks the transmission of arboviruses including Dengue, Chikungunya, Yellow Fever and Zika (Bian et al., 2010;Blagrove et al., 2013;Blagrove et al., 2012;Dutra et al., 2016;Ferguson et al., 2015;Moreira et al., 2009;van den hurk et al., 2012;Walker et al., 2011;Ye et al., 2015). In addition, Wolbachia-induced reproductive manipulation (cytoplasmic incompatibility) can drive the rapid spread of this endosymbiont through wild Ae. aegypti populations (Dutra et al., 2016;Hoffmann et al., 2011). While there is much interest in using a similar strategy to control malaria, there has been limited progress in identifying suitable, maternally transmitted symbionts in Anopheles mosquitoes. Numerous studies failed to identify Wolbachia from Anopheles species (Ricci et al., 2002), and although transinfection of Anopheles stephensi has been achieved (Bian et al., 2013), attempts to generate stable transinfected lines of An. gambiae have remained unsuccessful. Wolbachia has more recently been reported at low frequency, and very low apparent density, from certain field populations of Anopheles coluzzi and An. gambiae (Baldini et al., 2014). The natural Wolbachia -Anopheles gambiae system reported seems unlikely to have the characteristics required for development as a transmission blocking strategy since it has not been possible to select lines with high density and stable maternal transmission (Shaw et al., 2016).
To advance the prospect of a symbiont-based strategy for malaria control it will be important to continue to identify, generate and study a broad range of Anopheles -symbiont systems. Spiroplasmas are members of the Mollicutes, a bacterial group that split from a Gram-positive clostridial lineage of the eubacteria around 600-800 mya and has undergone degenerative evolution. Spiroplasmas are arthropod 'specialists' and all known species have some form of interaction with this clade (Gasparich et al., 2004). Members of this genus are functionally diverse, exhibiting a broad array of infection and transmission strategies: they can be pathogens, commensals or mutualists and rely on vertical or horizontal transmission (Regassa & Gasparich, 2006). In addition, Spiroplasma can confer a variety of insect hosts with resistance to a range of eukaryotic parasites, including nematodes, parasitoids and fungal pathogens (Jaenike et al., 2010;Łukasik et al., 2013;Xie et al., 2010), and they are therefore a good candidate for a symbiont that could be useful for control of Plasmodium.
Several Anopheles mosquito microbiome surveys have identified Spiroplasma from pathogenic clades (Lindh et al., 2005;Segata et al., 2016). In this study, we detected the presence of a novel strain of Spiroplasma in Anopheles gambiae mosquitoes. We sampled Anopheles gambiae sensu lato (s.l.) populations from geographically dispersed study sites in Kenya and found that the strain was present at low frequencies across both regions. We have also demonstrated that this Anopheles associated Spiroplasma is maternally transmitted. High-throughput 16S rRNA amplicon sequencing To maximize our chances of detecting potential symbionts we pooled 10 mosquitoes from each location (Central Kenya and Western Kenya). The pools were comprised of DNA from mosquito ovaries (5 mosquito samples) and whole mosquitoes (5 mosquito samples), since endosymbionts are generally at highest density in ovaries but can also be found in high densities in other tissues. The DNA samples were sent to the Research and Testing Laboratory (Lubbock, Texas, USA) for PCR amplification with 'universal' 16S rDNA primers (Lane, 1991;Lane et al., 1985), followed by MiSeq illumina sequencing. Samples were amplified in a two-step process that involved 25 µl reaction using Qiagen Hotstart Taq

Molecular phylogenetic analysis
Sequence alignments were performed using Clustal W in Geneious 8.1.9 software (www.geneious.com, Kearse et al., 2012). The trees were constructed by the maximum-likelihood method with a Tamura-Nei model in Geneious 8.1.9 software. Support for tree topology assessed by bootstrap resampling. To determine the phylogenetic position of Spiroplasmas identified in this study relative to previously identified Spiroplasmas, we compared sequence of 16S rRNA, rpoB and ftsZ genes. Nucleotide sequences of the other Spiroplasma species were derived from GenBank database (accession numbers shown in Figure 3). The length of the compared sequences was 301 bp for 16S rRNA, 210 bp for rpoB and 260 bp for ftsZ.

Establishment of iso-female lineages
Anopheles gambiae larvae collected from Central Kenya region were reared in the icipe mosquito insectary in Mbita, Kenya. Female mosquitoes that successfully reached adult stage were placed in standard 30cm x 30cm x 30cm rearing cages at a density of 30-100 mosquitoes per cage, ensuring a minimum of 30% males. Mosquitoes were then blood fed on Plasmodiumuninfected human blood, as previously described (Gouagna et al., 2003) and allowed to individually oviposit. After oviposition, eggs were counted and each female was screened for the presence of Spiroplasma. Adult progeny from infected mothers were counted for egg to adult survival rates and maintained to investigate transmission across multiple generations using the same experimental design. The eggs and adult progeny from some uninfected mothers were also counted to determine egg to adult survival rates. The effects of Spiroplasma on female fecundity were determined using female mosquitos collected as larvae from Central Kenya using the same strategy described above.
To reduce a potential bias from non-mated females, only broods consisting of more than 10 eggs were used to evaluate fecundity.

Mitochondrial DNA analysis
To determine the diversity of mosquito mtDNA, the 655bp ND5 gene was amplified using the primers described by Besansky (Besansky et al., 1997). Single PCR reactions were performed on the Veriti Thermal Cycler (Applied Biosystems, Carlsbad, CA, USA). PCR cycling conditions were initial denaturation at 95°C for 15 min, followed by 35 cycles of denaturation at 95°C for 30 s, annealing at 55°C for 30 s, elongation at 72°C for 30 s, then a hold temperature of 72°C for 7 min. PCR products were visualized on a 1% agarose gels. The PCR products were purified using ExoSap-IT purification protocol (USB Corporation Cleveland, OH). The DNA Sequences were cleaned and aligned using the MUSCLE algorithm (Edgar, 2004) in Geneious 8.1.9 software. Minimum spanning haplotype network (Bandelt et al., 1999) was constructed using Popart (http://popart.otago.ac.nz).

Results
Spiroplasma sequences isolated from An. gambiae Two pools of whole mosquitoes and ovaries were used to generate DNA for High-throughput sequencing of 16S rDNA, which resulted in 195,592 and 18,921 high-quality 16S rRNA sequences, from the Central Kenya pool (CK) and the Western Kenya pool (WK), respectively. Enterobacteriacea was most predominant group in CK with approximately 79% of the sequences, whereas both Propionibacteriaceae and Enterobacteriacea dominated in WK with 26% and 22%, respectively (see Figure 1). In the CK sample, a relatively small fraction of the 16S sequence reads (0.02%) were from Spiroplasmataceae. The Spiroplasma 16S sequence reads matched the 16S rDNA gene of Spiroplasma insolitum strain M55 with 100% identity. Spiroplasma insolitum M55 was originally isolated from a flower in Maryland, USA (Hackett et al., 1993).
Spiroplasma insolitum prevalence in field populations of An. gambiae s.l We developed a set of primers to target the rpoB gene of Spiroplasma insolitum. These primers were designed based on several Spiroplasma insolitum rpoB sequences from previous studies (Watanabe et al., 2013). The specificity of these primers for Spiroplasma insolitum was investigated on a panel of diverse Spiroplasma species. These primers were then used to determine the population-level prevalence of Anopheles gambiae associated Spiroplasma insolitum in mosquito samples obtained from Western Kenya (Kirindo, Nyawiya and Mageta Island) and Central Kenya (Mwea), see Figure 2. In all sites, mosquitoes were collected by mouth aspiration in houses across one rainy season (October-December 2016 or April-June 2017). In Mwea, we also collected Anopheles gambiae larvae, which were allowed to eclose before being screened for Spiroplasma as 21 day old adults. In Mwea, approximately 8% (n=490) of An. gambiae s.l harbored Spiroplasma. When collected as larvae, the rate of infection was higher, at 14% (n=163). In Western Kenya, the prevalence of Spiroplasma was generally lower and absent from one site. In Kirindo, the rate of Spiroplasma prevalence was 4% (n=173) and in Mageta the prevalence was 3% (n=66), whereas no infections were found in mosquitoes obtained from Nyawiya (n=222).

Molecular phylogenetic analyses of Spiroplasma
To determine the phylogenetic position of Anopheles associated Spiroplasma insolitum relative to other members of this clade, and to determine if multiple Spiroplasma strains are present in these populations, we developed primers to specifically target and amplify a region of the Spiroplasma insolitum ftsZ gene (Supplementary Table 1). In addition, we sequenced the region of rpoB amplified by our Insolitum-specific primers. The highthroughput sequencing that we carried out to investigate microbial diversity enabled us to obtain the sequence of a fragment of 16S rDNA. These sequences were used for the construction of phylogenetic trees, which indicate the strain of Spiroplasma from An. gambiae s.l can be classified into the citri-clade and confirm it clusters with Spiroplasma insolitum (see Figure 3). The 16S rDNA   fragment sequence was found to be identical to that of three previously described strains of Spiroplasma insolitum, M55, TU-14 and NBRC. The sequenced region of the rpoB gene from Anopheles associated Spiroplasma was also found to be identical to M55, TU-14 and NBRC and two strains of S. insolitum that are endosymbionts of flower bugs of the genus Orius, SpOriA/B (Watanabe et al., 2013). The region we sequenced of the ftsZ gene from Anopheles associated Spiroplasma was identical to M55, TU-14 and NBRC. Notably, ftsZ sequence data is not available for SpOriA/B. These results indicate that this Anopheles associated Spiroplasma strain is Spiroplasma insolitum, henceforth referred to as S. insolitum GAMB. Since all the S. insolitum GAMB genes (16S rDNA, rpoB and ftsZ) we sequenced were identical, we find no evidence for multiple strains of S. insolitum co-existing in the populations of Anopheles gambiae s.l. studied.

Spiroplasma insolitum GAMB is maternally transmitted in
Anopheles gambiae s.l but does not bias sex ratio or affect egg to adult survival To determine if Spiroplasma insolitum GAMB is maternally transmitted, we collected mosquitoes from the field and established iso-female mosquito lineages. We collected larvae from Mwea (where Spiroplasma insolitum GAMB prevalence was highest) and maintained them until they eclosed as G 0 adults, at which point they were blood fed then allowed to oviposit prior to screening for Spiroplasma infection. Three G 0 females carried Spiroplasma and from these, individual F1 female offspring were maintained to enable further screening for Spiroplasma. Most F1 did not produce viable offspring; in only one instance we obtained F2 females (see Figure 4). This is not altogether surprising, as field caught Anopheles gambiae s.l. are known to perform poorly prior to becoming 'acclimatized' to laboratory conditions (Diop et al., 1998). We found that Spiroplasma insolitum GAMB is maternally transmitted with very high efficiency, but that transmission efficiency did vary slightly between iso-female lineages. In two cases we observed perfect maternal transmission, whereas the rest had transmission efficiencies between 43% and 87%. The iso-female lineage that produced F2s showed 100% transmission from G0 to F1 and 83% transmission from F1 to F2.
Maternally transmitted symbionts are known to manipulate the sex ratio of their hosts to gain a transmission advantage (Hurst & Majerus, 1993). To determine if Spiroplasma insolitum GAMB affects the sex ratio of Anopheles hosts, we monitored the sex ratio of offspring in Spiroplasma-infected isofemale lineages. The sex ratio did not differ substantially from the expected 50% female/male in the two lineages producing greater than 10 progeny (see Figure 4), and therefore we conclude that Spiroplasma insolitum GAMB is not a male-killer.
We also monitored the fecundity and egg to adult survival rate for Spiroplasma infected and uninfected iso-female lineages (see Figure 5). We did not observe any significant difference between the fecundity and survival rate of Spiroplasma infected and uninfected individuals, indicating that Spiroplasma insolitum GAMB is not pathogenic.

Figure 4. Vertical transmission and sex ratio in Spiroplasma infected iso-female lineages.
Anopheles gambiae mosquitoes collected from Mwea as larvae were used to establish iso-females lineages. Offspring from Spiroplasma-infected iso-female lineages were screened for the presence of Spiroplasma at adult stage. The number of progeny screened are shown for each female and in the subsequent generation (G 0 , F 1 and F 2 ). At each generation the infection levels and sex ratio were monitored. The bars represent the % sex ratio of the total number of offspring from each iso-female lineage. The presence of between 40%-60% males in all but one family demonstrates that Spiroplasma is not a male killer. The observed maternal transmission efficiency (TE), ranged between 43% and 100% with an average of 82.6%.

Association between Spiroplasma insolitum GAMB and mtDNA haplotypes
We investigated possible associations between mtDNA haplotypes and Spiroplasma insolitum GAMB infection. Symbionts that have recently infected an insect population and are maintained by high efficiency maternal transmission can be expected to be associated with one or a few mitochondrial DNA haplotypes. In contrast, if a symbiont infection is associated with many or most mtDNA haplotypes, this suggests an older infection, paternal transmission or an appreciable level of horizontal transmission in addition to maternal transmission. We sequenced the ND5 mtDNA gene, which has been widely used for haplotyping Anopheles gambiae mosquitoes (Besansky et al., 1997). Of 21 Anopheles gambiae specimens collected in Central Kenya (Mwea), 11 were shown to be Spiroplasma insolitum GAMB positive based on PCR based screening. We identified a total of 6 haplotypes, four of these are identical to haplotypes reported previously (Aboud et al., 2014;Besansky et al., 1997), while two were novel (M1 and M2, see Figure 6). In the presence of Spioplasma in both species (Segata et al., 2016). In both studies, the Spiroplasma identified appears to be closely related to Spiroplasma ixodetus, which was initially discovered as a pathogen associated with ticks (Tully et al., 1995), and is thus quite different from the Spiroplasma identified in our study.
We demonstrate that Spiroplasma insolitum GAMB is found at relatively low frequencies in Anopheles gambiae s.l. mosquito populations. Frequencies tended to be higher in the Central Kenya Region (Mwea) than in the Western Kenya Region (Mageta, Kirindo and Nyawiya). These two regions have quite different mosquito habitats; in western Kenya most mosquitoes emerge from isolated puddles whereas Mwea (Central Kenya) is a rice growing region where Anopheles larvae are abundant in rice paddies and irrigation canals. While both sites experience an increase in mosquito abundance during the rainy season, the difference is less pronounced in Mwea due to year around irrigation providing more permanent larval habitats. We also noted that a higher infection rate was observed in Mwea when we collected larvae instead of adults. A possible explanation for this is that a greater number of these samples had Spiroplasma levels that were above the detection limit, or above levels required for the bacteria to be maintained through pupal morphological re-organization into the adult stage. This could be due the favorable laboratory larval and adult rearing conditions, which would likely result in more nutrients available to host and symbiont. Additionally, the mosquitoes collected as larvae (aged to 21 days) were likely to be older than field caught mosquitoes (unknown age). Spiroplasma densities in insects are known to significantly increase over the life of the host (Goto et al., 2006;Herren et al., 2014) and this could also have caused the observed increase in number of Spiroplasma positives.
Since insect associated Spiroplasma are known to exhibit a variety of different transmission patterns it was important to determine if Spiroplasma insolitum GAMB is maternally transmitted. We established iso-female lineages from infected field collected larvae and demonstrated that maternal transmission does occur with a high level of efficiency. We note that transmission efficiency does appear to vary slightly between the iso-female lineages tested. Two families exhibited very high maternal transmission to F1s, whereas transmission efficiency for the third was about 50%, however the third only generated 6 female offspring. While the spatial localization and mechanistic basis of Spiroplasma insolitum GAMB maternal transmission were not investigated, the closely related Spiroplasma poulsonii is known to achieve trans-ovarial maternal transmission by subversion of the yolk uptake pathway in Drosophila melanogaster (Herren et al., 2013). Other microbial symbionts that persist in the intestinal tract are more likely to achieve maternal transmission by the fecal-oral route. While transovarial maternal transmission tends to be higher efficiency, there are reports of very high efficiency transmission via the fecal-oral route (Hosokawa et al., 2013).
Many maternally transmitted insect symbionts have evolved strategies to bias sex ratio towards females to gain a transmission advantage (Werren & O'Neill, 1997). The most common manifestation of this is male-killing in which the endosymbiont confers male-specific embryonic lethality (Hurst & Majerus, 1993). Male 4 most common haplotypes, we found both Spiroplasma insolitum GAMB infected and uninfected individuals, suggesting that this symbiont is either an ancient infection or exhibits appreciable levels of horizontal transmission.

Discussion
We have identified a strain of Spiroplasma that is associated with Anopheles gambiae mosquitoes in Kenya. This Spiroplasma was initially identified from 16S rDNA high throughput sequencing reads from a pool of Anopheles gambiae mosquitoes from Central Kenya (Mwea). 16S rDNA sequence revealed that this Anopheles associated Spiroplasma strain was a member of the citri clade and grouped with Spiroplasma insolitum. Most of the known insect endosymbiotic Spiroplasmas are found in this clade, for example, S. poulsonii, S. citri and S. insolitum are all species which have been studied as insect endosymbionts (Haselkorn, 2010;Watanabe et al., 2014). It is notable that Spiroplasma insolitum GAMB is very closely related to SpOriA/B, a strain of Spiroplasma insolitum that is an endosymbiont of flower bugs in the genus Orius (Watanabe et al., 2014). We are aware of two other studies that identified Spiroplasma sequence associated with Anopheles mosquitoes. An investigation on midgut bacteria in Anopheles gambiae and Anopheles funestus from Western Kenya detected 16S sequence corresponding to Spiroplasma in Anopheles funestus (Lindh et al., 2005). Another study investigated the microbiome of the reproductive tracts of Anopheles gambiae and Anopheles coluzzi in Burkina Faso and found evidence for the killing has been observed in numerous strains of endosymbiotic Spiroplasma (Anbutsu & Fukatsu, 2011). We monitored the sex ratio of the Spiroplasma-carrying lineages and found very close to 50% male offspring in the two lineages where more than ten offspring could be examined, suggesting that Spiroplasma insolitum GAMB is not a male killer.
A number of Spiroplasma are known to be pathogenic to their arthropod hosts (Clark et al., 1985;Nunan et al., 2005). By monitoring the fecundity and egg to adult survival rate we determined that Spiroplasma insolitum GAMB was not pathogenic. This finding when coupled with its phylogenetic position suggests that Spiroplasma insolitum GAMB is likely to either be a commensal or mutualist, although adult fitness assays are also needed. From the standpoint of developing Spiroplasma insolitum GAMB as part of a future microbe based transmission blocking strategy this is advantageous, as a pathogenic phenotype could limit the capacity for Spiroplasma to spread through the host population.
We did not observe a clear correlation between mtDNA haplotype and Spiroplasma infection. This could be due to two major possibilities. The first, that Spiroplasma infection has been maintained in this species for a very long period of time enabling diversification of mtDNA within the infected lineage (as is the case for obligate symbionts, Moran, 2006). A second possibility is that there is significant horizontal transmission of Spiroplasma between the An. gambiae mosquitoes and/or that paternal as well as maternal transmission occurs, resulting in the wide and almost even distribution of Spiroplasma infection between mitochondrial haplotyes. Given that Spiroplasma insolitum GAMB is not an obligate symbiont (as it is not found in all Anopheles gambiae s.l. individuals), it seems most likely that Spiroplasma insolitum GAMB is both horizontally and vertically transmitted. The phylogeny of Spiroplasma also suggests a high frequency of horizontal as well as vertical transmission (Haselkorn et al., 2009) and many Spiroplasma likely utilize both forms of transmission. From the standpoint of using Spiroplasma as a tool for blocking VBD transmission, the prospect of strains being both vertically and horizontally transmitted is of considerable interest and could render them easier to spread through host populations.
We have reported the identification of novel strain of Spiroplasma in Anopheles gambiae s.l. mosquitoes. Questions that now need to be addressed, once stable infected colonies have been successfully created, include the effects of these Spiroplasma on Plasmodium transmission, effects on adult lifespan, where it localizes within the mosquito, and mechanisms of vertical and horizontal transmission. Spiroplasmas are known to protect a variety of insect hosts from diverse parasites (Jaenike et al., 2010;Łukasik et al., 2013;Xie et al., 2010) and therefore the discovery of Spiroplasma insolitum GAMB could provide a step towards the development of novel malaria control strategies. distinct tissues used to make the libraries.
For all PCR assays, it would be nice to have more information regarding negative controls. To do so, the authors could have used mosquitoes that they know are negative. I say this because BLAST Spiroplasma analysis of the and primers used in this study revealed that they exhibit very high identity with Rpob FtsZ these genes in other bacteria. Can the authors conclusively rule out non-specific amplification from bacterial members of the microbiota that are not ? Spiroplasma Is the work clearly and accurately presented and does it cite the current literature? Yes

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? Yes
No competing interests were disclosed.

Competing Interests:
I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. There is great interest in using host-associated microbes to control insect disease vectors. Maternally transmitted microbial symbionts are especially promising because many have evolved sophisticated strategies to spread through host populations and/or suppress interfere with insect parasites and pathogens. In this study, the authors report the discovery and initial characterization of a strain of Spiroplasma insolitum that infects Anopheles gambiae mosquitoes. It occurs at low frequency in western and central Kenya. The authors establish infected isofemale lines in the lab, rear them for two generations, and show that offspring are infected as well (although transmission efficiency is often low). Mitochondrial sequence analysis suggests that horizontal transmission is pervasive. This is an important contribution to the study of mosquito symbiosis and it will be interesting to characterize this symbiont in greater detail.
It would be useful to have more information about the establishment and rearing of the iso-female lines (this is not always clear), because it is a bit difficult to interpret the results on transmission to offspring. It would be useful to have more information about the establishment and rearing of the iso-female lines (this is not always clear), because it is a bit difficult to interpret the results on transmission to offspring. Were some or all offspring screened as eggs? Were some or all offspring reared individually? If offspring were reared in the same container as larvae, then it is possible that there was horizontal transmission from adult females, and then between larvae. It would be useful to rear infected and uninfected individuals together. If uninfected mosquitoes do not pick up the infection from infected ones, then vertical transmission may predominate.

Some other comments:
It would be useful to have a sentence explaining why Spiroplasma-infected mosquitoes were only kept in the lab for 2 generations.
The fact that the strain is identical at all genes sequenced in the study to S. insolitum from insects from other orders also suggests predominantly horizontal transmission. It would also be useful to mention more clearly that a number of cultivable/horizontally transmitted Spiroplasma have been isolated from (non-anopheline) mosquitoes.
In Figure 4, it would be useful to add arrows to indicate which F1 lines are the mothers of the F2.
It would be useful to have more information about Figure 5. Was transmission to offspring measured in these Spiroplasma-infected females? Which mothers are these (connect to Figure 4? F1?)?
Why were mosquitoes collected as larvae older than field-caught mosquitoes (second last page)?

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound? Yes

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? Partly
Collaborate with former supervisor of two of the coauthors.

Competing Interests:
Referee Expertise: Insect parasites and symbionts I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.