Insects are known to form successful long-term symbioses with endosymbiotic bacteria (Harris et al., 2010; Gupta and Nair, 2020). Such bacteria may be referred to as primary symbionts, if they are obligatory and thus essential for the survival of the host (Sudakaran et al., 2017), or as secondary symbionts, if they are not involved in functions essential for host survival or reproduction. Facultative association with secondary symbionts may affect host biology and ecology by influencing host fitness through altered traits and capabilities (Sudakaran et al., 2017).

Rickettsia (Alphaproteobacteria: Rickettsiales) are gram-negative obligate intracellular bacteria found within eukaryotic cells; they have a variety of interactions with their arthropod hosts (Weinert et al., 2015; El Karkouri et al., 2022). Rickettsia can be found in many insect tissues, including Malpighian tubules, gut compartments, oocytes, and, rarely, sperm cells (Machtelinckx et al., 2012; Watanabe et al., 2014; Dally et al., 2020). While Rickettsia were initially recognized as human and animal pathogens transmitted by blood-feeding arthropods such as ticks, mites, lice and fleas, in recent decades most species and strains have been shown to be non-pathogenic endosymbionts of arthropods (Weinert et al., 2015; Pilgrim et al., 2021).

Most insect hosts of Rickettsia belong to the orders Hemiptera, Coleoptera, Diptera and Hymenoptera (Weinert et al., 2015), in all of which the bacteria have a positive effect on host survival and reproductive success. In the pea aphid Acyrthosiphon pisum (Hemiptera: Aphididae), for instance, Rickettsia improve host resistance to a pathogenic Pandora fungus (Lukasik et al., 2013). In the sweet potato whitefly Bemisia tabaci (Hemiptera: Aleyrodidae), Rickettsia provide protection against Pseudomonas syringae infection (Hendry et al., 2014), and enhance fertility, longevity and development (Himler et al., 2011). Additionally, these bacteria stimulate oogenesis in the booklouse Liposcelis bostrychophila (Psocoptera: Liposcelididae) (Perotti et al., 2006). Rickettsia may also have negative effects on their hosts, for example by slowing development (Semiatizki et al., 2020) or inducing male-killing in the host (Werren et al., 1994; Giorgini et al., 2010).

Within the suborder Heteroptera (Order Hemiptera), the family Miridae includes some 10,000 species with a wide range of feeding habits including herbivory, carnivory, and omnivory (Wheeler, 2002). Several members of this family, such as Nesidiocoris tenuis, Macrolophus pygmaeus and M. melanotoma, serve as biological control agents against key crop pests (Schaefer and Panizzi, 2000; Castañé et al., 2011), yet these mainly predaceous species may also feed on plant materials (Perdikis and Lykouressis, 2000). The potential of these predators to reduce agricultural yields when prey is scarce has limited their use in biological control programs (Castañé et al., 2011; Moerkens et al., 2016). It is therefore important to explore the presence and role of symbionts in the nutritional ecology of these omnivorous biological control agents.

Omnivores, like most insects, serve as hosts to symbiotic bacteria. In previous studies, M. pygmaeus was found to harbor Wolbachia and two species of Rickettsia, R. bellii and R. limoniae. In contrast, M. melanotoma was observed to house only Wolbachia and R. limoniae. Regarding symbiont distribution, Wolbachia are located in the ovaries of M. pygmaeus, where they induce cytoplasmic incompatibility (Machtelinckx et al., 2009, 2012). The two Rickettsia species, in contrast, were found in our previous study to be distributed in both the digestive and reproductive systems of M. pygmaeus, each displaying a unique cellular occupancy and a specific distribution pattern along the digestive system compartment (Dally et al., 2020). The significance of these differences in distribution is not yet clear, but a link to the omnivorous host diet has been proposed. The current study revealed the presence of both Rickettsia species, R. bellii and R. limoniae, in M. melanotoma, in contrast with the earlier findings of Machtelinckx et al. (2012). Accordingly, our objective is to further explore the role of Rickettsia in omnivore diet and nutrition. To this end, the occurrence of Rickettsia species in M. melanotoma was described, their localization patterns were compared to those found in M. pygmaeus, and Rickettsia abundance and phylogeny in the two hosts were determined.

In this study we detected, characterized and compared two symbiotic Rickettsia species, R. limoniae and R. bellii, inhabiting two Macrolophus species, M. pygmaeus and M. melanotoma. Phylogenetic analyses showed that R. limoniae from the two bugs studied most closely resemble each other, and share higher sequence similarity than to the symbiont sequenced from M. pygmaeus reported from the United Kingdom. Further, these R. limoniae sequences cluster closer to a bacterium sequenced from an Odonata species, than to those found in various other Hemiptera. A similar phylogenetic pattern was obtained for R. bellii, but based on the genes sequenced, it seems there is a clade that includes the symbiont found in Hemiptera, which all cluster together, with the exception of a strain from Bemisia tabaci. Altogether, the emerging pattern agrees with previous studies that suggest occasional horizontal transfer of Rickettsia (Nováková and Šmajs, 2019). Such a transfer may occur via feeding, either directly by the consumption of infected prey, or indirectly via feeding on host plants shared with other phytophagous insects. It is also possible that additional insect species feeding on the same host plants might take up microorganisms transferred by the bug to the plant. Rickettsia have been shown to be acquired from environmental sources by two species of Spalangia (Hymenoptera: Pteromalidae), S. endius and S. cameroni (Tzuri et al., 2021). Furthermore, Chrostek et al. (2017) reviewed the transmission of Rickettsia, Wolbachia, and Cardinium through plants by the leafhopper Euscelidius variegatus. These findings indicate that a co-evolutionary process between Macrolophus species and their symbiotic Rickettsia is unlikely.

In M. melanotoma, R. bellii is clustered with the bellii group based on the CoxA and GltA genes, but is more similar to the R. limoniae group according to the 16S rRNA gene phylogeny. These results suggest a possible recombination between the two Rickettsia species (Jiggins, 2006). Recombination events are not rare in Rickettsia genomes (Wu et al., 2009; Merhej and Raoult, 2011), and play an important role in the evolution of these bacteria, notably by enabling them to adapt to new hosts (Thomas, 2016).

The two Rickettsia species were documented in all individuals of M. pygmaeus (Dally et al., 2020) and had a high rate of occurrence (84%) in screened M. melanotoma, adults. This is contrary to the findings of a previous study, in which only R. limoniae, was detected in M. melanotoma (Machtelinckx et al., 2012). This difference may stem from the different collection sites (Israel in the current study while Machtelinckx et al., 2012 collected in Greece and Italy), which may have various environmental conditions, selection pressures or infection histories. Moreover, this may be an indication of horizontal transfer of Rickettsia between the mirid species. Finding the two Rickettsia species in the salivary glands of their mirid hosts lends support to the possibility that the bacteria may be transferred to, and acquired from, their host’s food source. It is increasingly noted that bacteria can colonize the salivary glands of insects; this may have significant implications for plant–insect interactions, particularly for disease transmission by herbivorous insects (Kaiser et al., 2010; Body et al., 2013). Two well-documented examples are Citrus greening, the most destructive citrus disease in the world, caused by Candidatus Liberibacter asiaticus which resides in the salivary glands of the vector, the Asian citrus psyllid Diaphorina citri (Ammar et al., 2011), and Flavescence dorèe, a severe grapevine disease caused by Candidatus Phytoplasma vitis found in salivary glands of the leafhopper Scaphoideus titanus (Cicadellidae) (Marzorati et al., 2006). The endosymbiotic bacterium Cardinium was found to be injected into the plant by this leafhopper vector without any notable influence (Gonella et al., 2015). It can thus be seen that bacteria transmitted by herbivores to their host plants are not necessarily pathogenic. In some cases, bacterial symbionts have been shown to alter plant metabolic reconfiguration in a way that better meets insect nutritional needs. The endosymbiont Wolbachia has been shown to be transmitted by larvae of the leaf-mining moth Phyllonorycter blancardella to the leaves of apple seedlings (Malus domestica), where the bacterium alters the phytohormonal profile of the leaves, creating an optimal microenvironment for its host (Kaiser et al., 2010; Body et al., 2013). Similar transmission was reported from another hemipteran, the sweet potato whitefly Bemisia tabaci, which transmits and acquires Rickettsia through the phloem of cotton plants (Caspi-Fluger et al., 2012).

The distribution of the two Rickettsia species in M. melanotoma ovarioles resembles their localization in M. pygmaeus (Machtelinckx et al., 2012; Dally et al., 2020). In both cases, the symbionts were found primarily in the germarium, strongly suggesting that the bacteria are transmitted vertically from the mother to her offspring via the egg (transovarial transmission), a common transmission pathway in many symbiont-host systems (Monti, 2017).

The two Rickettsia species displayed a unique distribution pattern in the two studied Macrolophus species. In M. melanotoma, R. bellii was distributed throughout the entire digestive tract, while R. limoniae appeared mainly in the foregut and the midgut. In M. pygmaeus, on the other hand, the distribution of R. bellii was more restricted than that of R. limoniae. Many phenotypic, genetic and physiological factors may lead to such variation in Rickettsia distribution in the gut. In addition, environmental factors such as seasonal conditions and host plants may also be involved. In the present study, for example, M. pygmaeus was reared in the laboratory on frozen Ceratitis capitata eggs and tomato seedlings under optimal temperature conditions, whereas M. melanotoma was collected in the spring from Dittrichia viscosa plants. Because species in the Miridae, especially those of the omnivorous Dicyphini tribe, are the best-known group of arthropods specialized for foraging, feeding, and oviposition on sticky plants (Wheeler and Krimmel, 2015), the difference in the food plant of the two studied hosts may influence Rickettsia involvement in the bugs’ nutritional ecology, and their distribution pattern might be influenced by the food source.

Mutualistic microbes often inhabit the caeca connected to the most anterior portion of the midgut (Hosokawa et al., 2016; Nardi et al., 2019), yet to the best of our knowledge such caeca structures have not been reported from any mirid species studied so far. Instead, our earlier work on M. pygmaeus (Dally et al., 2020), as well as the present study, revealed the presence of a paired organ at the posterior end of the midgut. Although this caecum-like organ appears to differ morphologically between the two studied Macrolophus species, as well as between males and females, it harbors both Rickettsia species in separate cells or, more rarely, together in the same cell. The function of these caeca-like organs is as yet unknown.

Obligate blood-feeders such as ticks, bed bugs, and tsetse flies have evolved to rely on microbial endosymbionts to supplement several B vitamins – such as biotin (B7), folate (B9), and riboflavin (B2) – that are deficient in blood. For example, the endosymbiont Rickettsia buchneri provides its tick hosts with biotin and folate, essential components for cell growth in all eukaryotic and prokaryotic organisms (Rio et al., 2016; Narasimhan et al., 2021).

In many hemipterans, symbiotic bacteria found on the midgut epithelia supply their hosts with essential nutrients such as amino acids and vitamins (Kikuchi et al., 2009), and recycle metabolic wastes of the host (Ohbayashi et al., 2019). The presence of Rickettsia in the digestive tract of both M. pygmaeus and M. melanotoma, as well as other omnivorous mirids, such as the midgut epithelial cells of Stenotus binotatus (Chang and Musgrave, 1970) and the lumen of the digestive tract of Nesidiocoris tenuis (Caspi-Fluger et al., 2014), may be indicative of the nutritional role filled by Rickettsia symbionts in this group. However, the omnivorous habit of these bugs ensures a balanced diet, so other phenotypes could not be ruled out. Examples could be found in Liu and Guo (2019), which reviewed the roles Rickettsia and Wolbachia play in protecting hosts from stresses caused by natural enemies, heat, and toxins. Moreover, M. pygmaeus became more sensitive to freezing conditions when all three symbionts, Wolbachia, R. bellii and R. limoniae were removed, but it is unclear the absence of which symbiont causes the negative effect (Maes et al., 2012).

Coinfections are well known for Wolbachia (Ant and Sinkins, 2018) but have been less commonly recorded for other symbionts. The presence of both R. bellii and R. limoniae in the two studied bug species is, however, in agreement with the widespread occurrence of co-infections in the Torix group of Rickettsia (Pilgrim et al., 2021). Co-infection with Rickettsia from the Torix group was also observed in the two damselflies Coenagrion puella and Coenagrion pulchellum (Thongprem et al., 2021).

In conclusion, the presence of two Rickettsia species with specific distribution patterns that differed between two Macrolophus host species was described. These symbiont species are phylogenetically similar and were found in the vast majority of the field-collected host bugs. As sequencing the genome of the two Rickettsia species could elucidate the role of such symbionts in the feeding habits of their hosts, it would be warranted to expand this research by combining lab experiments on feeding behavior with investigations into molecular genetics and bioinformatics.

The data presented in the study are deposited in the GenBank repository, accession numbers OQ374915-OQ374918, OQ374921-OQ374924, OQ398710, OQ398711, OQ271380-OQ271381 and OQ410975.

MD: wrote the manuscript, performed, and designed the work. YZ: performed the phylogenetic analysis and wrote sections of the manuscript. EB: captured FISH images, and wrote the FISH method section. NM-D: contributed to the phylogenetic analysis section. MC and EZ-F: contributed to conception and design of the study. All authors contributed to manuscript revision, read, and approved the submitted version.

This research was supported by the Israel Science Foundation (grant no. 397/21 to EZ-F).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.