Management of corn leafhopper (Homoptera: Cicadellidae) and corn stunt disease in sweet corn using reflective mulch

Plastic reflective mulches significantly reduced populations of corn leafhopper, Dalbulus maidis (DeLong & Wolcott), adults and the incidence of corn stunt disease caused by Spiroplasma kunkelii (CSS) in late planted sweet corn (Zea mays L.). The reflective mulches were more effective than were either foliar or soil applied insecticides in managing both the leafhopper and the pathogen it transmits. Yields of marketable ears were 1.5 to 2 times greater in reflective mulch plots than from fallow plots. This was due to larger ears (individual ear weight and length) rather than an increase in the number of ears. The use of reflective mulches provides an alternative strategy to insecticides in the management of both D. maidis and corn stunt disease. Such a strategy may prove useful to growers in Latin America and to limited resource growers and organic growers in the United States who wish to grow corn without the use of insecticides.     

Habitat of the corn leafhopper (Hemiptera: Cicadellidae) during the dry (winter) season in Mexico

Although the corn leafhopper Dalbulus maidis (DeLong and Wolcott) is the most important vector of maize pathogens in Latin America, little is known about how and where it overwinters (passes the dry season), particularly in Mexico. The objectives of this study were (1) to monitor the abundance of D. maidis adults throughout the dry season in maize and maize-free habitats and (2) to determine where and how D. maidis adults, exposed or nonexposed to the maize pathogen Spiroplasma kunkelii Whitcomb, overwinter in a maize-free habitat. Work for the first objective was done during the two consecutive dry seasons of 1999-2000 and 2000-2001; the second objective was done during the dry seasons of 2003-2004 and 2005-2006. During the dry winter seasons, D. maidis was prevalent as long as maize was present in irrigated areas. The leafhopper was found in 52 of the 58 irrigated maize fields sampled in Mexico at the end of the dry seasons of 1999-2000 and 2000-2001. However, leafhopper adults were not found in nonirrigated maize-free habitats at high elevation during the dry winter season (February, March, and April), although leafhopper adults were prevalent on perennial wild grasses in January after maize harvest. Additional experiments revealed, however, that corn leafhopper adults, although few in number, survived the entire dry season in these nonirrigated maize-free fields. Also, no detectable difference in survival existed between leafhoppers exposed and those not exposed to S. kunkelli during the two dry seasons in the maize-free habitat.     

Field overwintering biology of Spiroplasma kunkelii (Mycoplasmatales: Spiroplasmataceae) and its vector Dalbulus maidis (Hemiptera: Cicadellidae)

We studied the corn stunt spiroplasma (CSS), Spiroplasma kunkelii (Mycoplasmatales: Spiroplasmataceae) and its vector the corn leafhopper Dalbulus maidis (Hemiptera: Cicadellidae) under field conditions in Mexico. We surveyed for the presence of CSS in D. maidis by using PCR on samples of adults collected during the 2000–01 and 2003–04 winter (dry) seasons from irrigated low‐elevation sites and un‐irrigated high‐elevation sites. Also, we determined the body size and number of mature eggs of D. maidis females collected during the dry season in 2004 and in females collected on maize seedlings in the first months (June and July) of the wet (summer) season in 2005. Our PCR results showed that CSS was present in leafhopper adults collected during the 2000–01 and 2003–04 dry seasons in irrigated low‐elevation sites. However, in un‐irrigated high‐elevation sites, CSS was present in corn leafhopper adults caught before, but not during, the dry seasons. In these un‐irrigated high‐elevation sites, adult leafhoppers without CSS were recovered during the first 2 months (November and December) of the dry season from the foliage of wild perennial grasses. Females collected on wild perennial grasses in December 2004 showed similar head width and wing length to females caught on maize seedlings in June 2005. However, females collected on maize seedlings in July 2005 had the widest heads, longest wings and highest number of mature eggs. The largest body size of these females that arrived in July 2005 indicates that they are immigrants and support the Roff’s hypothesis that insect migrants should be larger than nonmigrants.     

Transmission of sugarcane white leaf phytoplasma by Yamatotettix flavovittatus, a new leafhopper vector

Sugarcane white leaf disease is caused by plant pathogenic phytoplasmas that are transmitted to the plant by the leafhopper Matsumuratettix hiroglyphicus (Matsumura). To determine whether there are other insect vectors that transmit this disease pathogen, leafhopper species in sugarcane, Saccharum officinarum L., fields in northeastern Thailand were monitored by using light traps. Sixty-nine leafhopper species from family Cicadellidae were found. Using nested polymerase chain reaction (PCR) with specific primers, a 210-bp amplified DNA fragment corresponding to phytoplasma associated with sugarcane white leaf disease was detected from 12 species of leafhoppers [Balclutha rubrostriata (Melichar), Balclutha sp., Bhatia olivacea (Melichar), Exitianus indicus Distant, Macrosteles striifrons Anufriew, Matsumuratettix hiroglyphicus (Matsumura), Recilia distincta (Motschulsky), Recilia dorsalis (Motschulsky), Recilia sp., Thaia oryzivora Ghauri, Yamatotettix flavovittatus Matsumura, and Xestocephalus sp.]. The percentage of individual infection with phytoplasma varied from 5% in B. olivacea to 35% in Xestocephalus sp. The most abundant leafhopper species, i.e., E. indicus, Y. flavovittatus, and M. hiroglyphicus were used in transmission tests to determine their vector status for the sugarcane white leaf phytoplasma transmission. Infected insects were reared on healthy plants and specific PCR followed by sequencing of the amplicons was used to determine whether the phytoplasma was transmitted to the plants. The results showed that both Y. flavovittatus and M. hiroglyphicus, but not E. indicus, can transmit sugarcane white leaf phytoplasma to healthy sugarcane plants. The transmission efficiency of M. hiroglyphicus (55%) was higher than that of Y. flavovittatus (45%). We conclude that Y. flavovittatus is a newly discovered vector for sugarcane white leaf disease, in addition to M. hiroglyphicus. These two species peak at different times of the year and therefore complement each other in the transmission of the phytoplasma. Because there are no known alternative host plants for the sugarcane white leaf, management of the disease will necessarily require the control of both Y. flavovittatus and M. hiroglyphicus.     

Disruption of a gene predicted to encode a solute binding protein of an ABC transporter reduces transmission of Spiroplasma citri by the leafhopper Circulifer haematoceps

Spiroplasma citri is transmitted from plant to plant by phloem-feeding leafhoppers. In an attempt to identify mechanisms involved in transmission, mutants of S. citri affected in their transmission must be available. For this purpose, transposon (Tn4001) mutagenesis was used to produce mutants which have been screened for their ability to be transmitted by the leafhopper vector Circulifer haematoceps to periwinkle plants. With one mutant (G76) which multiplied in leafhoppers as efficiently as S. citri wild-type (wt) strain GII-3, the plants showed symptoms 4 to 5 weeks later than those infected with wt GII-3. Thirty to fifty percent of plants exposed to leafhoppers injected with G76 remained symptomless, whereas for wt GII-3, all plants exposed to the transmission showed severe symptoms. This suggests that the mutant G76 was injected into plants by the leafhoppers less efficiently than wt GII-3. To check this possibility, the number of spiroplasma cells injected by a leafhopper through a Parafilm membrane into SP4 medium was determined. Thirty times less mutant G76 than wt GII-3 was transmitted through the membrane. These results suggest that mutant G76 was affected either in its capacity to penetrate the salivary glands and/or to multiply within them. In mutant G76, transposon Tn4001 was shown to be inserted into a gene encoding a putative lipoprotein (Sc76) In the ABCdb database Sc76 protein was noted as a solute binding protein of an ABC transporter of the family S1_b. Functional complementation of the G76 mutant with the Sc76 gene restored the wild phenotype, showing that Sc76 protein is involved in S. citri transmission by the leafhopper vector C. haematoceps.     

Physical and genetic map of the Spiroplasma kunkelii CR2-3x chromosome

Spiroplasma kunkelii (class Mollicutes) is the characteristically helical, wall-less bacterium that causes corn stunt disease. A combination of restriction enzyme analysis, pulsed-field gel electrophoresis (PFGE), and Southern hybridization analysis was used to construct a physical and genetic map of the S. kunkelii CR2-3x chromosome. The order of restriction fragments on the map was determined by analyses of reciprocal endonuclease double digests employing I-CeuI, AscI, ApaI, EagI, SmaI, BssHII, BglI, and SalI; adjacent fragments were identified on two-dimensional pulsed-field electrophoresis gels. The size of the chromosome was estimated at 1550 kb. Oligonucleotide pairs were designed to prime the amplification of 26 S. kunkelii gene sequences in the polymerase chain reaction (PCR). Using PCR amplicons as probes, the locations of 27 S. kunkelii putative single-copy genes were positioned on the map by Southern hybridization analyses of chromosomal fragments separated in PFGE. The nucleotide sequence of the single ribosomal RNA operon was determined and its location mapped to a chromosomal segment bearing recognition sites for SalI, SmaI, EagI, and I-CeuI.     

Inoculation and acquisition of maize bushy stunt mycoplasma by its leafhopper vector Dalbulus maidis

Maize bushy stunt mycoplasma (MBSM), a mycoplasma-like organism, is transmitted in a persistent manner by the corn leafhopper, Dalbulus maidis, to maize (Zea mays). The influence of the duration of acquisition access and inoculation access periods on the transmission of MBSM by D. maidis was investigated. The proportion of plants infected by D. maidis increased significantly from 0 to 0.51 as the inoculation access time to a plant increased from 10 min to 72 h (X²= 101.5, P < 0.001). Likewise, the proportion of insects acquiring MBSM from infected plants increased from 0 to 0.19 as the acquisition access time to the source plant increased from 10 min to 72 h (X²= 53.2, P < 0.001). The data were fitted to a loglinear regression model. No significant association was found between the sex of the insects and vector ability.     

Virus-vector relationships of chickpea chlorotic dwarf geminivirus and the leafhopper Orosius orientalis (Hemiptera: Cicadellidae)

Chickpea chlorotic dwarf geminivirus (CCDV) is one of the viruses associated with chickpea stunt disease. It is transmitted by the leafhopper Orosius orientalis. The minimum acquisition access period (AAPmin) and inoculation access period (IAPmin) were found to be less than 2 min, while the minimum latency period (LPmin) was less than 2 h. The median AAP, IAP and LP were 8.0 h, 2.3 h and 27.7 h, respectively. No difference in transmission rates (proportion of leafhoppers able to transmit) was observed between male and female leafhop-pers. In serial transmission experiments, transmission was shown to be persistent, and after a 2-day AAP about 80% of the leafhoppers transmitted the virus for most of their life. The virus could be detected in individual leafhoppers by DAS-ELISA. It did not multiply in the leafhopper, but, instead, decreased in concentration during leafhopper feeding on a non-host of the virus.     

Corn Stunt Spiroplasma in Dicotyledonous Plants

Corn stunt spiroplasma (CSS) was transmitted by the leafhopper vector Euscelidius variegatus (Kirschbaum) and produced symptoms on four dicotyledonous plant species, Sinapis alba L. (mustard), Pisum sativitm L. (pea), Raphanus sativus L. (radish) and Spinacia oleracea L. (spinach). The vectors became infective by microinjection with a broth culture of CSS. This insect also acquired CSS from infected mustard plants and transmitted it to healthy ones.     

Disruption of a Gene Predicted To Encode a Solute Binding Protein of an ABC Transporter Reduces Transmission of Spiroplasma citri by the Leafhopper Circulifer haematoceps

Spiroplasma citri is transmitted from plant to plant by phloem-feeding leafhoppers. In an attempt to identify mechanisms involved in transmission, mutants of S. citri affected in their transmission must be available. For this purpose, transposon (Tn4001) mutagenesis was used to produce mutants which have been screened for their ability to be transmitted by the leafhopper vector Circulifer haematoceps to periwinkle plants. With one mutant (G76) which multiplied in leafhoppers as efficiently as S. citri wild-type (wt) strain GII-3, the plants showed symptoms 4 to 5 weeks later than those infected with wt GII-3. Thirty to fifty percent of plants exposed to leafhoppers injected with G76 remained symptomless, whereas for wt GII-3, all plants exposed to the transmission showed severe symptoms. This suggests that the mutant G76 was injected into plants by the leafhoppers less efficiently than wt GII-3. To check this possibility, the number of spiroplasma cells injected by a leafhopper through a Parafilm membrane into SP4 medium was determined. Thirty times less mutant G76 than wt GII-3 was transmitted through the membrane. These results suggest that mutant G76 was affected either in its capacity to penetrate the salivary glands and/or to multiply within them. In mutant G76, transposon Tn4001 was shown to be inserted into a gene encoding a putative lipoprotein (Sc76) In the ABCdb database Sc76 protein was noted as a solute binding protein of an ABC transporter of the family S1_b. Functional complementation of the G76 mutant with the Sc76 gene restored the wild phenotype, showing that Sc76 protein is involved in S. citri transmission by the leafhopper vector C. haematoceps.     

The parasitoid Gonatopus bartletti reduces presence of plant-pathogenic Spiroplasma kunkelii within the leafhopper vector Dalbulus maidis

Homopteran vectors (e.g., leafhoppers) of plant pathogens are vessels for reproduction of cell wall‐free bacteria. These vectors also serve as hosts for larval parasitoid dipterans, hymenopterans, and strepsipterans. However, no study has explored the relationship among these wall‐free bacteria and parasitoid larvae within the insect host. We studied the corn stunt spiroplasma (CSS), Spiroplasma kunkelii Whitcomb (Mycoplasmatales: Spiroplasmataceae), a bacterium that originated from secondary symbionts that cause corn stunt disease in maize, Zea mays L., and its reproduction in the haemolymph of the corn leafhopper, Dalbulus maidis (Delong and Wolcott) (Homoptera: Cicadellidae). We also studied the dryinid parasitoid Gonatopus bartletti Olmi (Hymenoptera: Dryinidae), the larva of which feeds in the corn leafhopper haemolymph. Our results showed that when CSS and the wasp coexisted in D. maidis, the development of the parasitoid was not affected by S. kunkelii. Parasitoid development was successfully completed when leafhoppers acquired S. kunkelii before or after parasitism and when CSS had median (10 days) and long (20 days) incubation periods in the leafhopper before parasitization. The presence of S. kunkelii did not affect parasitoid development to the adult stage. However, polymerase chain reaction showed that the presence (survival) of S. kunkelii in the leafhopper was negatively affected by the parasitoid larva. Fewer leafhoppers had CSS before and after parasitization compared with leafhoppers that only acquired the CSS. This negative effect helps to explain the high parasitism rate by G. bartletti in D. maidis and the low presence of S. kunkelii in the corn leafhopper when CSS and the wasp parasitoid overlap throughout their geographic distribution. The parasitoid larva may negatively affect S. kunkelii by (1) producing antibacterial peptides that are toxic to CSS; (2) producing teratocytes that take nutrients from the host for larval development, but these nutrients are required by CSS; (3) affecting, indirectly, CSS through other symbiotic microorganisms; and (4) producing proteins with antibacterial activity that are present in the venom of the wasp parasitoid.     

Effect of maize stunting mollicutes on survival and fecundity of Dalbulus leafhopper vectors

The effect of corn stunt spiroplasma (CSS) on survival and fecundity of three Dalbulus leafhopper species was determined. CSS significantly reduced the survival, as measured by the time to 50% (t₅₀) and 25% (t₂₅) survival, and by the scale parameter (b) of the Weibull model, for D. elimatus and D. gelbus. Fecundity of these two species, as measured by the net and gross reproductive rates, was also reduced by CSS. CSS did not significantly affect the corn leafhopper, D. maidis. In a separate experiment, maize bushy stunt mycoplasma (MBSM) reduced the survival and fecundity of D. maidis at temperatures from 20 to 29 °C. The effect of MBSM on D. maidis survival was less severe than CSS on D. elimatus and D. gelbus; t₂₅, but not t₅₀, was reduced by MBSM. Survival times and the cohort generation time generally declined with increasing temperature. Fecundity, however, generally increased with increasing temperature.     

Differential acquisition of chrysanthemum yellows phytoplasma by three leafhopper species

Chrysanthemum yellows (CY) phytoplasma has been transmitted with three leafhopper species: Euscelidius variegatus (Kirschbaum), Macrosteles quadripunctulatus (Kirschbaum) and Euscelis incisus (Kirschbaum): the first two species are reported as CY phytoplasma vectors for the first time. Leafhoppers were allowed to acquire the pathogen from the following source plants: Apium graveolens L., Catharanthus roseus L., Chrysanthemum carinatum Schousboe L. and C. frutescens L. DNA extracted from healthy or inoculative leafhoppers-exposed plants were analyzed by dot-blot and Southern hybridizations with a molecular probe constructed onto a fragment of European aster yellows phytoplasma DNA. The three leafhopper species were able to transmit CY phytoplasma after acquisition on chrysanthemum, but only M. quadripunctulatus and E. variegatus transmitted after feeding on periwinkle, and none acquired it from celery. All plant species tested were susceptible to CY, but while chrysanthemum and periwinkle were suitable for both inoculation and acquisition, celery did not seem to be a good source of phytoplasma for further inoculations. It is concluded that host plants influence leafhoppers' vectoring ability, possibly due to the different feeding behaviour of the insects on diverse plant species. Since CY, like several other phytoplasmas, can be transmitted by different insect species, it is likely that a close transmission specificity probably does not exist between phytoplasmas and their leafhopper vectors.     

Cell division gene cluster in Spiroplasma kunkelii: functional characterization of ftsZ and the first report of ftsA in mollicutes

Spiroplasma kunkelii is a helical, wall-less bacterium that causes corn stunt disease. In adaptation to its phloem-inhabiting parasitic lifestyle, the bacterium has undergone a reductive evolutionary process and, as a result, possesses a compact genome with a gene set approaching the minimal complement necessary for multiplication and pathogenesis. We cloned a much-reduced cell division gene cluster from S. kunkelii and functionally characterized the key division gene, ftsZ(sk). The 1236-bp open reading frame of ftsZ(sk) is capable of encoding a protein with a calculated molecular mass of 44.1 kDa. Protein sequence alignment revealed that FtsZ(sk) is remarkably similar to FtsZ proteins from other eubacteria, and possesses the conserved GTP-binding and hydrolyzing motifs. We demonstrated that overexpression of ftsZ(sk) in Escherichia coli causes transgression of the host cell division, resulting in a filamentous phenotype. We also report, for the first time, the presence of a ftsA gene in the cell division cluster of a mollicute species.     

Infection and replication sites of Spiroplasma kunkelii (Class: Mollicutes) in midgut and Malpighian tubules of the leafhopper Dalbulus maidis

Spiroplasma kunkelii distribution and infection mechanisms in the intestines and Malpighian tubules of Dalbulus maidis were investigated by transmission electron microscopy. Spiroplasmas were found between microvilli and in endocytic vesicles of the midgut epithelium. At the basal part, cytoplasmic vesicles contained multiple spiroplasmas with tube-like extensions and spiroplasmas accumulated between the laminae rara and densa of the basal lamina. Tip structures of flask-shaped spiroplasmas pierced the lamina densa that was discontinuous in close proximity to spiroplasmas. Spiroplasmas were found in hemolymph, crossed the basal lamina of Malpighian tubule epithelium and accumulated at high numbers in muscle cells that had cytopathogenic changes. S. kunkelii had perithrochous approximately 8nm diameter structures determined to be fimbriae protruding from the cell surface, and similar structures were adhering to the basal lamina of midgut epithelium and to external lamina of muscle cells. Further, spiroplasmas had pili-like appendages at one or both cell poles and appeared to conjugate. This is the first time that fimbriae and pili have been observed in a mollicutes.     

THE MANNER OF TRANSMISSION OF SOME BARLEY YELLOW-DWARF VIRUSES BY DIFFERENT APHID SPECIES

Some barley yellow-dwarf (BYD) viruses isolated from cereal crops in Great Britain were transmitted by Rhopalosiphum padi , L. and others were not. Sitobion fragariae (Walker), S. avenae (Fabricius), and Metopolophium dirhodum (Walker) all transmitted viruses of both types, but they usually transmitted those of which Rhopalosiphum was a vector less readily than did R. padi. The transmissibility of a virus by a given aphid species was not affected by transmission with another, less efficient, vector species. Neomyzus circumflexus (Buckt.) and Rhopalosiphum maidis (Fitch) transmitted the few viruses with which they were tested.
A few R. padi acquired virus from infected leaves during 30 min. feeding and inoculated healthy seedlings during 15 min. feeding, but the minimum total time taken to acquire and transmit was 10 hr. and 32 hr. were needed for about half the aphids that were able to acquire and transmit virus to do so. This may indicate the existence of a short latent period of the virus in the vector, although the evidence is not conclusive. The times spent on infected plants influenced the results more than those spent on healthy ones; many transmissions occurred with short feeding times on healthy plants so long as the time spent on infected leaves was long, but the reverse was not true. Nymphs of R. padi that moulted after they left infected plants on which they fed long enough to become infective, infected slightly fewer plants than adults fed for the same times.     

Variation in vector competency depends on chrysanthemum yellows phytoplasma distribution within Euscelidius variegatus

Phytoplasmas are plant‐pathogenic Mollicutes transmitted by leafhoppers, planthoppers, and psyllids in a persistent propagative manner. Chrysanthemum yellows phytoplasma (CY) is a member of ‘Candidatus Phytoplasma asteris’, 16Sr‐IB, and is transmitted by at least three leafhopper species, Macrosteles quadripunctulatus Kirschbaum, Euscelidius variegatus Kirschbaum, and Euscelis incisus Kirschbaum (all Homoptera: Cicadellidae: Deltocephalinae). Although M. quadripunctulatus transmits CY with very high efficiency (near 100%), 25% of E. variegatus repeatedly fail to transmit CY. The aims of this work were to correlate vector ability with different pathogen distribution in the insect body and to investigate the role of midgut and salivary glands as barriers to CY transmission. Euscelidius variegatus individuals acquired CY by feeding on infected plants or by abdominal microinjection of a phytoplasma‐enriched suspension. Insects were individually tested for transmission on daisy seedlings [Chrysanthemum carinatum Schousboe (Asteraceae)], and thereafter analysed by real‐time polymerase chain reaction (PCR) for CY concentration on whole insects or separately on heads and the rest of the body. Hoppers were classified as early and late transmitters or non‐transmitters, according to the time inoculated plants required for expression of CY symptoms. Similar transmission efficiencies were achieved following feeding or abdominal microinjection, suggesting that salivary glands may be a major barrier to transmission. Following acquisition from infected plants, all transmitters tested positive by PCR, and 60% of non‐transmitters also tested positive although with a significantly lower CY concentration. This indicates that a minimum number of phytoplasma cells may be required for successful transmission. The midgut may have prevented phytoplasma entry into the haemocoel of PCR‐negative non‐transmitters. Results suggest that both midgut and salivary glands may act as barriers. To assess the effect on CY transmission of a specific parasitic bacterium of E. variegatus, tentatively named BEV (Bacterium Euscelidius variegatus), we established a BEV‐infected population by abdominal microinjection of BEV bacteria. The presence of BEV did not significantly alter the efficiency of CY transmission.     

Horizontal transmission of the symbiotic bacterium <Emphasis Type="Italic">Asaia</Emphasis> sp. in the leafhopper <Emphasis Type="Italic">Scaphoideus titanus</Emphasis> Ball (Hemiptera: Cicadellidae)

BACKGROUND: Bacteria of the genus Asaia have been recently recognized as secondary symbionts of different sugar-feeding insects, including the leafhopper Scaphoideus titanus, vector of Flavescence dorée phytoplasmas. Asaia has been shown to be localized in S. titanus gut, salivary glands and gonoducts and to be maternally transmitted to the progeny by an egg smearing mechanism. It is currently not known whether Asaia in S. titanus is transmitted by additional routes. We performed a study to evaluate if Asaia infection is capable of horizontal transmission via co-feeding and venereal routes. RESULTS: A Gfp-tagged strain of Asaia was provided to S. titanus individuals to trace the transmission pathways of the symbiotic bacterium. Co-feeding trials showed a regular transfer of bacterial cells from donors to recipients, with a peak of frequency after 72 hours of exposure, and with concentrations of the administrated strain growing over time. Venereal transmission experiments were first carried out using infected males paired with uninfected females. In this case, female individuals acquired Gfp-labelled Asaia, with highest infection rates 72-96 hours after mating and with increasing abundance of the tagged symbiont over time. When crosses between infected females and uninfected males were conducted, the occurrence of "female to male" transmission was observed, even though the transfer occurred unevenly. CONCLUSIONS: The data presented demonstrate that the acetic acid bacterial symbiont Asaia is horizontally transmitted among S. titanus individuals both by co-feeding and venereal transmission, providing one of the few direct demonstrations of such a symbiotic transfer in Hemiptera. This study contributes to the understanding of the bacterial ecology in the insect host, and indicates that Asaia evolved multiple pathways for the colonization of S. titanus body.