Discovery and characterization of a novel TA

A) The gray line represents the frequency of XZ1516 alleles across the genome after four generations of intercrossing with QX1211. Each panel corresponds to a C. elegans chromosome and each x-axis tick indicates 5 Mb. The dotted blue line represents the expected allele frequency for each chromosome with no selection. The region highlighted in red on the right side of chromosome V shows the greatest allele frequency deviation from expectation. B) Crosses between XZ1516 (purple) and QX1211 (yellow) establish the inheritance pattern of the TA element. Bar plots show the fraction of dead L1s observed in each cross. Error bars indicate 95% binomial confidence intervals calculated using the normal approximation method. Crosses from left to right: selfing of XZ1516/QX1211 heterozygous hermaphrodites; XZ1516/QX1211 heterozygous hermaphrodites crossed to QX1211 males; XZ1516/QX1211 heterozygous males crossed to QX1211 hermaphrodites. The observed fraction of dead L1s was not significantly different from the expected fractions for a maternally inherited TA element, exact binomial test. C) Model of the TA inheritance. Punnett square shows the lethality pattern expected in progeny from selfing of XZ1516/QX1211 heterozygous hermaphrodites. A maternally deposited toxin (black square) is present in all progeny and causes L1 lethality unless a zygotically expressed antidote (white circle) is also present.

Identification of the TA components

A) Localization of the TA element genes in XZ1516. Top panel: Strain genotypes of near-isogenic lines are displayed as colored rectangles (XZ1516 in purple; QX1211 in yellow; Cas9-induced deletion in red) for chromosome V. The fraction of L1 lethality after selfing of the NIL/QX1211 hermaphrodites is shown to the right of each NIL. The bottom panel depicts a summary of QX1211 sequencing reads aligned to the XZ1516 genome corresponding to the mapped TA element. Gray bars denote short-read sequencing depth in 200 bp windows and red dots denote the number of variants detected between QX1211 and XZ1516 in each window. The XZ1516 and QX1211 genome are so diverged that short reads derived from QX1211 don’t align to the XZ1516 genome in the 200 bp windows with no corresponding read depth, as indicated by a lack of a gray bar. The toxin and antidote genes are highlighted in green and light blue, respectively. B) Knockout and transgenic rescue experiments define the TA components. Bar plots denote the fraction of dead L1s derived from selfing F1 heterozygous individuals. Error bars indicate 95% binomial confidence intervals calculated using the normal approximation method. Blue and green boxes with “A” and “T” indicate intact antidote and toxin genes, respectively; white boxes indicate deletions of these genes. XZ1516 genotypes are depicted in purple and QX1211 genotypes are depicted in yellow. Panels from top to bottom: XZ1516/QX1211 control cross, the observed lethality is not significantly different from the expected 25%; toxin knockout cross to QX1211 the observed lethality is significantly different from the expected 25% p = 1.38e-31; antidote transgenic rescue cross the observed lethality is significantly different from the expected 25% p = 1.22091e-53; toxin and antidote double knockout cross to XZ1516 the observed lethality is not significantly different from the expected 25%. An exact binomial test was used to determine significance.

Demographics of the tmrl-1/amrl-1 TA

A) A dendrogram showing the relatedness of 550 wild C. elegans strains at the TA locus. Branches are colored to represent the three distinct clades, where purple denotes the XZ1516-like clade, yellow denotes the N2-like clade, and pink denotes the NIC195-like clade. B) Isolation location of strains collected in Hawaii. Pie charts show the number of isolates from each clade when multiple strains were collected at one location, with colors as in A. C) Bar plots show the fraction of dead L1s in crosses between XZ1516 and NIC195 (left) and between XZ1516 and NIC195 with its antidote allele knocked out (right), indicating that this antidote is active against the XZ1516 toxin. Error bars indicate 95% binomial confidence intervals calculated using the normal approximation method. The observed lethality in the NIC195 x XZ1516 cross is significantly different from the expected 25% (p = 6.14e-19, exact binomial test), while the antidote knockout difference is not significantly different. D) Synteny plot of the TA region between the XZ1516 (top) and N2 (bottom) genomes. The TA components tmrl-1 and amrl-1 are colored green and blue, respectively. E) Percent amino acid identity of ∼5500 one-to-one orthologs identified between the XZ1516 and N2 genomes. Amino acid identity for tmrl-1 is indicated with a red line.

The N2 tmrl-1 allele contributes to larval arrest in the absence of MUT-16

A) Density plots showing the distribution of animal lengths on the x axis for the Δtmrl-1, Δmut-16, and the Δtmrl-18; Δmut-16 double knockout lines. The distribution of animal lengths are significantly different for all comparisons (Kruskal-Wallis test; p = 1.56e-133 for the Δmut-16 to double knockout comparision, p = 7.51e-67 for Δtmrl-1 to double knockout comparison, and p ≈ 0). B) Animal length data from A) were binned to approximate larval stages as described in the methods. Stacked bar charts of the fraction of animals for each developmental stage for the Δtmrl-1, Δmut-16, and the Δtmrl-1; Δmut-16 double knockout lines are shown. The fraction of the population is shown on the y axis for each developmental stage – yellow: L1, green: L2/L3, and blue: L4. The fraction of adults is omitted for clarity, but corresponds to the fraction that brings the total to 1 for each genotype.