Vulva

RNAi was performed as described elsewhere (Fraser et al. 2000 (link); Kamath et al. 2000 ) with minor adaptations when the rrf-3 strain was used: after transferring L3- to L4-staged hermaphrodites onto the first plate, we left them for 48 h at 15°C instead of 72 h and then plated single adults onto other plates seeded with the same bacteria. Furthermore, we did not remove the mothers from the second plates. The phenotypes assayed are these: Emb (embryonic lethal), Ste (sterile), Stp (sterile progeny), Brd (low broodsize), Gro (slow postembryonic growth), Lva (larval arrest), Lvl (larval lethality), Adl (adult lethal), Bli (blistering of cuticle), Bmd (body morphological defects), Clr (clear), Dpy (dumpy), Egl (egg-laying defective), Lon (long), Mlt (molt defects), Muv (multivulva), Prz (paralyzed), Pvl (protruding vulva), Rol (roller), Rup (ruptured), Sck (sick), Unc (uncoordinated) Thin and Pale. Emb was defined as greater than 10% dead embryos for N2 and greater than 30% dead embryos for rrf-3. Ste required a brood size of fewer than ten among fed N2 worms and fewer than five among rrf-3. Each postembryonic phenotype was required to be present among at least 10% of the analysed worms.

Total RNA was extracted from normal controls and cancer patients from cervix, endometrium and vulvar tissues and was hybridized on Affymetrix HG133 A 2.0 microarray chips corresponding to more than 14,500 uniquely represented genes (NetAffx 32). A total of 35 samples were used to identify potential biomarkers and signatures in each type of cancer. The data were analyzed with the R language (version 3.0.2) and bioconductor package (version 2.13) [55 (link)]. The RMA algorithm [56 (link)] and log2 transformation were used for background correction and normalization of the data. For those genes that were represented in more than one probe, only probes with the highest average value across all arrays were kept. In order to identify differentially expressed genes, we performed a Student's t-test in unlogged data between normal and cancer tissues, and those genes with a value of p < 0.05 and a fold change (±) greater than 1.5, were considered significant. The microarray data (GSE63678) were submitted to GEO [57 ].

The first crossing experiments were performed as previously reported^{31 (link)}. In short, three J4 juvenile females of dioecious species and six young-adult males of P. pacificus were crossed on Nematode Growth Medium (NGM) plates with a 50 µl OP50 lawn (N = 4 in each species). The following strains were used: P. pacificus, PS312; P. exspectatus, RS5522; P. occultus, RS5811; P. sikae, RS5901; P. arcanus, RS5527; P. kurosawai, RS5914; P. taiwanensis, RS5797; P. maxplancki, RS5594. Males were removed after two days and females were removed after four days to avoid backcross. Hybrids were allowed to cross on the same plate and cultured. To prevent starvation, at least 50 worms were transferred to a new plate to continue the culture. If newborn larvae continued to be produced for one month (~6 generations), we considered the hybrids to propagate. To confirm the reproducibility of the original cross between P. pacificus and P. exspectatus, we made eight additional replicates (Supplementary Table 6). To prevent starvation, 25% of the individuals on the plate were transferred before starvation (after day 5). Thus, the number of animals counted on day ten is reduced relative to day five because 75% of the animals were not transferred. Subsequently, >95% of worms were transferred in the second or later transfers. We counted the number of individuals all 5 days for 30 days, indicating the numbers of J2/J3 and J4 juveniles, as well as adults for both, hermaphrodites and males. Note that eggs were not counted. We found continuous production of juveniles in four of the eight replicates (50%). Note that the number of animals declines over time because some hybrid progeny die without producing a large number of progeny. Importantly, however, new juveniles were constantly observed throughout the duration of the experiment, and no trend of change of sex ratio was observed (Supplementary Table 6).
For quantitative reproduction tests, one virgin female and one young-adult male were mated on the NGM plate with a 50 µl OP50 lawn with egg laying for six days. Parents were transferred to new plates every second day. Progeny were grown for three to four days on these plates. The number of males, females (or hermaphrodites) and immature progeny were counted on the basis of their morphology. Because the hermaphrodites have the same morphology as females, we do not distinguish these two sexes. When the two-arm gonad and the vulva were observed, the worm was categorized as a female or hermaphrodite. When the one-arm gonad and connection of the gonad to the spicule were observed, the worm was categorized as a male. When these reproductive traits were not observed, the worm was categorized as an immature animal. The type strain of P. pacificus, PS312, and an inbred line of the type strain of P. exspectatus, RS5522B, were used in this experiment. For hybrid crosses, old females or hermaphrodites (four days after J4 stage) were used for mating to let hermaphrodites run out of self-sperm. We also tested the number of progeny of old P. pacificus hermaphrodites without mating at the same time (N = 18). Because only one progeny was found from all hermaphrodites (0.056 progeny per replication on average), the self-progeny is negligible in the analysis. For backcrossing, we first prepared F1 animals produced by P. exspectatus dam and P. pacificus sire or F1 animals produced by P. pacificus dam and P. exspectatus sire using the experimental set-up described in the preceding and backcrossed them with animals of the pure species. We used young J4-stage females or hermaphrodites for backcrossing. We did not test the backcross with P. pacificus hermaphrodites that produce ~200 self-progeny because that makes the interpretation difficult. For the test of hermaphroditic reproduction of F1 animals, F1 female or hermaphrodite was placed on the NGM plate with a 50 µl OP50 lawn without males. We tested the wild-type cross of P. exspectatus in each backcross as control. For intercrosses of F1 animals, F1 hybrids were crossed to each other to avoid the effect of inbreeding. The sample number of each experiment is listed in Supplementary Table 1. Asymptotic Wilcoxon–Mann–Whitney test was performed using wilcox_test function of an R package, coin.
The reproductive capacity of F1 males was compared between crosses of P. pacificus and different dioecious species using the same experimental scheme. The F1 males were produced by crosses between females of dioecious species and P. pacificus male and backcrossed with parental dioecious species. We used wild isolates of six dioecious species. Only presence or absence of progeny (BC1) was analysed in these experiments (N = 25 each).

Organotypic invasion assays were performed as previously described (Gifford and Itoh, 2019 (link)). Briefly, collagen I (BD Biosciences cat. No. 354249) and Matrigel (BD Biosciences cat. No. 354234) were mixed to yield a final collagen concentration of 4 mg/ml and a final Matrigel concentration of 2 mg/ml. After the gel had been left to set at 37°C for 1 hr, mixture of 5 × 10⁵ A431 cells and 5 × 10⁵ vulval CAFs (VCAFs) were plated on the top in complete medium. Twenty-four hours later, the gel was then mounted on a metal bridge and fed from underneath with complete medium (changed daily). After 6 days, the cultures were fixed with 4% PFA plus 0.25% glutaraldehyde in PBS and imaged using Zeiss LSM780 at a magnification of ×10 and ×20. Z-stack images spanning 100–150  μm were collected, and image stacks were processed by ZEN software to yield maximum-intensity projections.
For organotypic killing assay, the gels containing 5 × 10⁵ VCAFs were set without cancer cells and incubated for 5 days in complete media. Then the gels were incubated with the media with puromycin (5 µg ml^–1) for 48 hr to kill the fibroblasts and then washed three times with complete media (30 min per wash). 5 × 10⁵ cancer cells were then plated on top, and the assays proceeded as usual.
For quantification of the images, strand length and width were measured using Fiji software. Strand tapering was calculated by the following formula: strand width at 20% from the root/strand width at 80% from the root.

Human vulval CAFs are described in Gaggioli et al., 2007 (link). CAFs were cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% Fetal Bovine Serum (FBS) and 1% insulin–transferrin–selenium (Invitrogen, no. 41400–045) and 100  U/ml penicillin, and 100  μg/ml streptomycin. Human vulval SCC cell line A431 cells were grown in DMEM supplemented with 10% FBS, 100  U/ml penicillin, and 100  μg/ml streptomycin. For ROCK inhibitor treatment cells were treated with 10 µM Y27632.