Intrasperm vertical symbiont transmission

Significance

Diverse organisms are commonly associated with bacterial endosymbionts, which often affect hosts’ biology and phenotypes in a variety of ways. The majority of these symbionts are generally present in the host cell cytoplasm and maternally transmitted through host generations. Here, however, this conventional knowledge is countered by our discovery of intrasperm vertical transmission of nuclear-targeting bacterial symbiont (Rickettsia) in an insect (leafhopper Nephotettix cincticeps), which potentially erodes the nuclear-cytoplasmic conflict that governs the majority of endosymbiotic associations. The molecular and cellular mechanisms underlying the sperm head infection without disturbing sperm functioning are of not only basic but also applied interest, which may provide insights into the development of sperm-mediated genetic transformation and/or material delivery technologies.

Abstract

Symbiotic bacteria are commonly associated with cells and tissues of diverse animals and other organisms, which affect hosts’ biology in a variety of ways. Most of these symbionts are present in the cytoplasm of host cells and maternally transmitted through host generations. The paucity of paternal symbiont transmission is likely relevant to the extremely streamlined sperm structure: the head consisting of condensed nucleus and the tail made of microtubule bundles, without the symbiont-harboring cytoplasm that is discarded in the process of spermatogenesis. Here, we report a previously unknown mechanism of paternal symbiont transmission via an intrasperm passage. In the leafhopper Nephotettix cincticeps, a facultative Rickettsia symbiont was found not only in the cytoplasm but also in the nucleus of host cells. In male insects, strikingly, most sperm heads contained multiple intranuclear Rickettsia cells. The Rickettsia infection scarcely affected the host fitness including normal sperm functioning. Mating experiments revealed both maternal and paternal transmission of the Rickettsia symbiont through host generations. When cultured with mosquito and silkworm cell lines, the Rickettsia symbiont was preferentially localized within the insect cell nuclei, indicating that the Rickettsia symbiont itself must have a mechanism for targeting nucleus. The mechanisms underlying the sperm head infection without disturbing sperm functioning are, although currently unknown, of both basic and applied interest.

Endocellular bacterial symbionts are commonly found in diverse eukaryotes including animals, plants, fungi, and protists (1–7). In the majority of these cases, the symbionts are located in the cytoplasm of the host cells. Whereas the cytoplasmic symbionts are simply passed to daughter cells through host cell division in unicellular protists (6, 7), sex-related asymmetry in vertical symbiont transmission is generally observed in multicellular metazoans with sexual reproduction. Namely, the symbionts are transmitted vertically to the next host generation via infection to eggs in the maternal body, but not via infection to sperms (8, 9). Exceptional reports of paternal symbiont transmission are venereal transmission cases of several symbiotic bacteria (10, 11) and biparental transmission cases of some symbiotic viruses (12). Oocytes accumulate a large quantity of cytoplasm that provide a room for symbiont infection, whereas sperms discard their cytoplasm (together with inhabiting symbiotic bacteria) during spermatogenesis and transform into a streamlined shape with the small head consisting of condensed nucleus and the slender tail made of microtubule bundles for motility (13, 14). Therefore, if such symbiotic bacterial cells can be transmitted via sperm, a possible target may be the sperm head nucleus. Intranuclear bacterial symbionts, such as Holospora and Caedibacter, have been relatively well-documented from unicellular ciliates (6, 15), but reported only rarely from multicellular metazoans (16, 17). In insects and other arthropods, intracellular Rickettsia and Orientia pathogens/symbionts are sometimes observed to localize not only to the cytoplasm, but also to the nucleus of the host cells (18–21). In bathymodiolin mussels inhabiting hydrothermal vents and cold seeps, intranuclear bacterial parasites “Candidatus Endonucleobacter bathymodioli” have been described (16). Thus far, no case of paternal symbiont transmission via intrasperm passage has been reported. Considering the extremely streamlined sperm structure, bacterial infection to the sperm head nucleus is expected to impair genetic material and normal functioning of the sperm and, thus, intrasperm vertical symbiont transmission may seem unlikely to occur. In this study, however, we demonstrate that such a case exists in an insect.

The green rice leafhopper Nephotettix cincticeps (Uhler) (Hemiptera: Cicadellidae) (Fig. 1A), known as a notorious pest of rice in East Asia, is associated with two bacteriome-associated obligate symbionts, Sulcia and Nasuia, and a facultative symbiont of the genus Rickettsia (22). The Rickettsia symbiont of N. cincticeps represents a basal lineage of the genus Rickettsia (Fig. 1B) and exhibits high infection frequencies among N. cincticeps strains established from natural populations in Japan (Table 1). Previous histological studies described that Rickettsia-like bacterial cells are present not only in the cytoplasm but also in the nucleus of various cells and tissues of several leafhopper species including N. cincticeps (23, 24). Here we report that, in N. cincticeps, the Rickettsia symbiont efficiently targets and infects the host’s cell nuclei including sperm head nuclei, and vertically transmitted to the next host generation not only maternally via ovarial passage but also paternally via intrasperm passage with high fidelity.

Fig. 1.

The green rice leafhopper N. cincticeps and its Rickettsia symbiont. (A) An adult male of N. cincticeps. (B) Phylogenetic placement of the Rickettsia symbiont of N. cincticeps on the basis of 16S rRNA gene sequence. A Bayesian phylogeny inferred from 1,296 aligned nucleotide sites is shown. Posterior probabilities for the Bayesian phylogeny and bootstrap probabilities for the maximum likelihood phylogeny at 50% or higher are shown at the nodes, whereas asterisks indicate support values lower than 50%. Sequence accession numbers are shown in brackets. Major Rickettsia groups (40) are indicated on the right side.

Table 1.

Rickettsia infection frequencies in laboratory strains of N. cincticeps derived from different natural populations in Japan

Strain	Origin	Year	Infection rate,* %
Tsukuba-A	Yawara, Ibaraki	1988	100 (96/96)
Tsukuba-B	Tsukuba, Ibaraki	2006	100 (96/96)
Jyouetsu	Jyouetsu, Niigata	1993	100 (96/96)
Kagoshima	Kagoshima, Kagoshima	2001	100 (96/96)
Total	—	—	100 (384/384)

^*Diagnostic PCR was performed by using the primers NcRic_16S/f1 (5′-TGA CGG TAC CTG ACC AAG A-3′) and NcRic_16S/r1 (5′-AAG GGA TAC ATC TCT GCT T-3′) as described (22).

Results and Discussion

When we observed various cells and tissues of our N. cincticeps stocks by transmission electron microscopy, the Rickettsia symbiont was consistently found not only in the cytoplasm, but also in the nucleus of Malphigian tubule cells, midgut cells, and other types of host cells (Fig. 2 A and B). In an attempt to microbiologically characterize the Rickettsia symbiont, aseptically dissected ovaries of N. cincticeps were subjected to cultivation with insect cell lines, by which we established continuous Rickettsia cultures with the AeAl2 mosquito cell line and the aff3 silkworm cell line. In these heterospecific host cells, strikingly, the Rickettsia symbiont exhibited localizations not only to the cytoplasm but also to the nucleus (Fig. 2 C and D). The consistent nuclear localization in the different host species, which represent the distinct insect orders Hemiptera, Diptera, and Lepidoptera, strongly suggests that the Rickettsia symbiont itself must have some mechanism for targeting insect cell nuclei.

Fig. 2.

Transmission electron microscopy of nuclear localization of the Rickettsia symbiont in tissues of N. cincticeps and cell lines of other insects. (A) Malphigian tubule cell of N. cincticeps. (B) Midgut cell of N. cincticeps. (C) Cell line aff3 derived from the silkworm B. mori. (D) Cell line AeAl-2 derived from the mosquito A. albopictus. (E) Cross-section of sperm heads in testis of N. cincticeps. (F) Longitudinal section of sperm heads in testis of N. cincticeps. (G) Magnified image of the sperm heads. (H) Spermatheca of Rickettsia-uninfected female of N. cincticeps after mating with Rickettsia-infected male. mv, microvilli on the epithelium of spermatheca; n, nucleus; sh, sperm head; st, sperm tail. Asterisk in H highlights an intrasperm Rickettsia cell.

Our electron microscopic observations of the testis of N. cincticeps revealed that, surprisingly, almost all sperm head nuclei contained bacterial cells (Fig. 2 E–G). Of 296 sperm head nuclei we inspected on electron microscopic images representing a single cross-section of elongate sperm heads, 181 (61.1%) exhibited one or two bacterial cells on the sectioned plane (Fig. 2E). Longitudinal sections of the sperm heads revealed that the bacterial cells are arranged in a line along the long axis within each sperm head: More than 10 bacterial cells were often observed in a sperm head nucleus (Fig. 2 F and G). Fluorescence in situ hybridization of the sperm heads using specific oligonucleotide probes targeting 16S rRNA of the Rickettsia symbiont unequivocally demonstrated that the bacterial cells within the sperm head nuclei represent the Rickettsia symbiont (Fig. 3 A–C). Of 1,109 sperm heads inspected by in situ hybridization and fluorescence microscopy, 1,026 (92.5%) contained one or more Rickettsia cells, which were on average 5.33 ± 3.52 and ranging from 0 to 23 Rickettsia cells per sperm head (Fig. 4A).

Fig. 3.

In situ hybridization of the Rickettsia symbiont in mature sperm heads obtained from seminal vesicles of Rickettsia-infected males of N. cincticeps. (A) Red hybridization signals due to 16S rRNA of the Rickettsia symbiont. (B) Blue signals due to DNA staining of sperm heads. Note the unstained areas within the sperm heads, reflecting endonuclear localization of the Rickettsia symbiont cells. (C) Merged image. Note that a number of Rickettsia cells are arranged in a row within each sperm head.

Fig. 4.

Number of Rickettsia cells in mature sperm heads and vertical transmission rates of Rickettsia. (A) Number of Rickettsia per sperm head based on in situ hybridization images of 1,109 sperm heads obtained from seminal vesicles of four Rickettsia-infected adult males. (B) Vertical transmission rates of Rickettsia upon all mating combinations within and between the Rickettsia-infected and uninfected host strains. The numbers of parent pairs and those of total offspring are indicated above the columns.

Using a selective symbiont curing technique by rifampicin administration via rice seedlings (25), we established Rickettsia-infected (R⁺) and uninfected (R^–) strains of N. cincticeps under the same genetic background, with the obligate bacteriome symbionts Sulcia and Nasuia remaining intact. These R⁺ and R^– insect strains exhibited similar levels of fecundity, growth, and survival (Table 2), indicating no remarkable positive/negative effects of the symbiont infection on the host fitness at least under our rearing condition, although the possibility cannot be excluded that some effects at moderate levels would be detected with larger sample sizes. Eggs produced by mating with R⁺ males exhibited high rates of egg development almost comparable to those with R^– males (Table 3), indicating normal functioning of the Rickettsia-infected sperms. Meanwhile, statistical analysis showed that the cross between R^– females and R⁺ males produced significantly less offspring than the other crosses (χ² test; P < 0.001), suggesting slightly but significantly lower performance of the Rickettsia-infected sperms. These patterns may look like a low level of cytoplasmic incompatibility, but more data should be accumulated to test this hypothesis. Electron microscopy of dissected spermathecae confirmed transfer of intrasperm Rickettsia symbiont cells from R⁺ males to R^– females (Fig. 2H). When all possible mating combinations within and between the R⁺ and R^– insect strains were examined, the Rickettsia symbiont exhibited 100% maternal transmission and, notably, 61.8% (ranging from 40.0 to 80.0%) paternal transmission (Fig. 4B). This result indicates that only a few Rickettsia cells residing in the sperm head nucleus are sufficient for establishing the paternal symbiont transmission. The paternally transmitted Rickettsia symbiont exhibited 100% vertical transmission in subsequent host generations.

Table 2.

Comparison of fitness parameters between Rickettsia-infected and uninfected strains of N. cincticeps

	Fecundity*	Growth rate†	Survival,‡ %
Infected strain R+§	4.5 ± 2.8 (n = 8)	17.1 ± 0.7 (n = 42)	89.9% (89/99)
Uninfected strain R–§	6.6 ± 3.0 (n = 9)	17.0 ± 0.5 (n = 38)	91.5% (86/94)
P value¶	P = 0.16	P = 0.75	P = 0.70

^*Number of offspring per female in 2 d.

^†Nymphal duration in female (d).

^‡Adult emergence rate (%).

^§The uninfected strain R^– was generated from the infected strain R⁺ by an antibiotic treatment (Materials and Methods).

^¶P values were estimated by t test for fecundity and growth rate, and by χ²-test for survival.

Table 3.

Egg development rates in crosses within and between Rickettsia-infected and uninfected strains of N. cincticeps

	Females of infected strain R+, %	Females of uninfected strain R–, %
Males of infected strain R+	89.5* (196/219)†	75.9* (211/278)†
Males of uninfected strain R–	90.7* (107/118)†	94.4* (204/216)†

^*Significantly different from expected values (χ² test; P < 0.001).

^†Egg development rate (number of eggs with eyespots/total number of eggs inspected).

In conclusion, we discovered a previously unknown phenomenon: The intranuclear Rickettsia symbiont of N. cincticeps is transmitted efficiently through host generations via an intrasperm passage. In general, endosymbiotic bacteria are maternally transmitted and, thus, potentially incur an evolutionary conflict with their host organisms whose traits are biparentally inherited, which underlies such striking symbiont-mediated host phenotypes as cytoplasmic incompatibility, male-killing, parthenogenesis induction, and feminization (26). The not only maternal, but also efficient, paternal symbiont transmission in N. cincticeps potentially erodes the nuclear-cytoplasmic conflict that governs the majority of endosymbiotic associations (27, 28), thereby providing a unique empirical model that may shed light on the evolutionary aspects of symbiosis. For example, the biparental symbiont transmission may entail occasional mixing of different symbiont lineages, which would potentially lead to the evolution of some virulent phenotypes of the symbiont (29). Considering that some Rickettsia and allied pathogenic bacteria exhibit nuclear infections in somatic cells of their hosts (18–21), it seems plausible, although speculative, that the Rickettsia symbiont of N. cincticeps has coopted the pathogenic nuclear infection mechanism for establishing the sperm head infection and enabling the efficient paternal transmission. The molecular and cellular mechanisms underlying the sperm head infection without disturbing sperm functioning are of not only basic but also applied interest, which would potentially provide insights into the development of sperm-mediated genetic transformation and/or material delivery technologies that have long been anticipated but not yet realized (30, 31).

Materials and Methods

Insect Rearing.

N. cincticeps was reared on rice seedlings at 25–26 °C under a 16-h light and 8-h dark cycle either in plastic boxes (30 cm × 28 cm × 24 cm) for stock culturing, in glass bottles (180 mm high and 95 mm diameter) for mass-rearing experiments, or in glass test tubes (130 mm high and 16 mm diameter) for individual rearing. The insect strain mainly used in this study was originally collected from a rice field at Kagoshima, Japan, and harbor the bacteriome symbionts Sulcia and Nasuia and the systemic symbiont Rickettsia (22).

Antibiotic Treatment.

Rice seedlings were grown in glass test tubes with a small cotton block at the bottom, to which water containing 200 μg/mL rifampicin was added. Newborn nymphs were introduced into the test tubes and reared to adulthood. Offspring of the antibiotic-treated insects were transferred to new test tubes without the antibiotic, whereby isofemale lines were generated and maintained. After several generations, Rickettsia-negative isofemale lines were screened by diagnostic PCR (22), by which the Rickettsia-uninfected strain of N. cincticeps was established.

Rickettsia Cultivation.

The Rickettsia symbiont was cultivated with the AeAl-2 cell line derived from the mosquito Aedes albopictus (32) by inoculating a part of dissected ovary of Rickettsia-infected N. cincticeps into a plastic dish containing the cells, according to the cultivation procedure for Wolbachia (33). The cultivated Rickettsia symbiont was further transferred to the aff3 cell line from the silkworm Bombyx mori (34). The AeAl-2 and aff3 cells were grown in medium IPL-41 (GIBCO BRL 11505) supplemented with 5% (vol/vol) FBS (Sigma) at 26 °C.

Electron Microscopy.

The tissue and cell samples were prefixed with 0.8% glutaraldehyde and 1% paraformaldehyde in 0.06 M phosphate buffer for 1–2 h on ice, postfixed with 2% (wt/vol) osmium tetroxide for 1 h at room temperature, and dehydrated through an ethanol series. The dehydrated samples were embedded in Spurr resin, processed into ultrathin sections, stained with 2% (wt/vol) uranyl acetate and Sato’s lead solution, and observed under a transmission electron microscope (JEM-1010; JEOL).

Molecular Phylogenetic Analysis.

A multiple alignment of the nucleotide sequences was generated by the program MAFFT version 7.127b (35). The nucleotide substitution model, GTR + I + G, was selected by using the program jModelTest 2 (36, 37). The phylogenetic analyses were conducted by Bayesian and maximum-likelihood methods using the programs MrBayes v3.2.2 (38) and RAxML version 7.2.6 (39), respectively. Posterior probabilities were calculated for each node by statistical evaluation in Bayesian analysis, and bootstrap values were obtained with 1,000 replications in maximum-likelihood analysis.

In Situ Hybridization.

By dissecting male seminal vesicles in PBS (137 mM NaCl, 8.1 mM Na₂HPO₄, 2.7 mM KCl, and 1.5 mM KH₂PO₄), mature sperm suspension was prepared, smeared on MAS-coated glass slides (Matsunami Glass Ind., Ltd.), and air-dried. The sperms were fixed with 4% (wt/vol) paraformaldehyde in PBS for 60 min at room temperature. After rinsing twice with PBS, the samples were treated with 0.1 mg/mL pepsin in 0.01 M HCl for 15 min at 37 °C, and washed with 100% ethanol twice and air-dried. Approximately 150 μL of hybridization buffer [20 mM Tris⋅HCl at pH 8.0, 0.9 M NaCl, 0.01% SDS, and 30% (wt/vol) formamide] containing 100 nM each of three oligonucleotide probes specifically targeting 16S rRNA of Rickettsia spp., whose 5′ end was labeled with AlexaFluor647 dye, namely Apis-Ric16R1 (5′-TCC ACG TCA CCG TCT TGC-3′), Ric-R1071 (5′-CTT ATA GTT CCC GGC ATT AC-3′), and Ric-R1405 (5′-ACC CCA GTC GCT AAT TTT AC-3′), was applied onto the samples, covered with coverslip, and incubated in a humidified chamber at room temperature overnight. For removing nonspecific probe binding, the samples were washed in the hybridization buffer without the probes for 30 min at 42 °C. After thorough washing, the samples were mounted in Slowfade antifade solution (Invitrogen) supplemented with 0.25 μM SYTOX Green (Invitrogen), and observed under a laser confocal microscope (Pascal 5; Carl Zeiss).

Fitness Measurement.

Fecundity, nymphal growth, and survival were examined for the Rickettsia-infected (R⁺) and uninfected (R^–) strains of N. cincticeps. Fecundity was evaluated in terms of the number of offspring produced by a young female in 2 d. In each of glass bottles containing rice seedlings, six adult females and three adult males (3-d-old) were kept for 2 d, and the number of newborn nymphs that emerged in each of the bottles (8 bottles for the infected strain and 9 bottles for the uninfected strain) was counted 12 d later and divided by 6. Growth rate was evaluated in terms of days of nymphal duration to adulthood. Each newborn nymph was reared in a test tube containing rice seedlings, and duration to adult emergence was recorded individually. Survival was evaluated in terms of percentage of adult emergence. Each newborn nymph was reared in a test tube containing rice seedlings until it became an adult or died. Number of the nymphs individually reared in test tubes (#N) and number of the insects that reached adulthood (#A) were recorded, and #A ÷ #N × 100 was calculated.

Crossing Experiments.

For evaluating egg development rates, each of four mating combinations of 30 females and 15 males (R⁺ × R⁺, R⁺ × R^–, R^– × R⁺, and R^– × R^–) was reared in each of four plastic boxes containing a rice plant with four or five leaves. After 7 d, the plants were carefully dissected under a binocular microscope to isolate eggs, and the eggs were inspected for red eyespots as an indicator of embryonic development. For evaluating vertical transmission rates of the Rickettsia symbiont, four types of mating combinations between a female and a male (R⁺ × R⁺, R⁺ × R^–, R^– × R⁺, and R^– × R^–; 13, 9, 8, and 7 pairs for each type, respectively) were created in glass test tubes containing rice seedlings. From each of the test tubes in which next generation nymphs emerged, 5–28 nymphs were picked up, and 94–187 nymphs in total for each type were subjected to DNA extraction and diagnostic PCR detection of the Rickettsia symbiont.