Fig 1.
A sensitive genetic system for assaying RIP in regions of interspersed homology.
(a) Tester constructs comprise pairs of closely-positioned DNA repeats: the base-pair sequence of the “left” repeat (“Reference”) is held constant and corresponds to an arbitrarily chosen segment of the Neurospora genome, the base-pair sequence of the “right” repeat (“Test”) can be varied as desired. (b) Patterns of interspersed homology (relative to the reference sequence) are defined by the three distances: X—the length of the homologous unit, Y—the separation between homologous units, and Z—the sequence position of the first homologous unit relative to the reference sequence. The sum of X and Y defines the periodicity of interspersed homology. (c) Alignment of a reference sequence and three different test sequences. A minimally effective pattern of interspersed homology is "3H-8N" [14]. One such pattern, with the sequence position parameter set to 1 (Z = 1), is shown as 3H-8N_1. Two homology patterns examined in this study are 4H-7N_1 and 4H-7N_7.
Fig 2.
Short regions of perfect homology promote RIP mutation in combination with the 500-bp interspersed homology 4H-7N_1.
(a) Partially-homologous repeats containing 500 base-pairs of the interspersed homology 4H-7N_1 (as defined in Fig 1A) are created by replacing the cyclosporin-resistant-1 gene (csr-1) with synthetic DNA. (b) Derivatives of the basic repeat construct include 0–100 base-pairs of perfect homology created by integrating additional DNA between restriction sites NotI and NdeI. The repeat construct with 0 base-pairs of perfect homology was reported previously [14]. (c) The mean number of RIP mutations detected across the entire sequenced region for repeat constructs described in (Panels b and d). X-axis: Arabic numerals refer to the length of perfect homology inserted between NotI and NdeI sites, as shown in (Panel b); Roman numerals refer to additional repeat constructs tested in (Panel d). Error bars represent the standard error of the mean. The total number of sequenced spores and replica crosses are provided in S1 Table. (d) RIP mutation profiles for selected repeat constructs. Roman numerals correspond to the following RepeatIDs (provided in S1 Table): i—“XIQ”, ii—“XIO”, iii—“XIN”, iv—“XJG”, v—“XIM”, vi—“XKE”, vii—“XJH”, viii—“XJY”. The number of mutations per site is expressed as percent of all spores sequenced for a given repeat construct.
Fig 3.
Short regions of perfect homology promote stronger mutation in combination with the 500-bp interspersed homology 5H-6N_1.
(a) The mean number of RIP mutations (measured as in Fig 2C) for repeat constructs with 0–100 base-pairs of perfect homology adjacent to the interspersed homology 5H-6N_1 or 4H-7N_1. Data for the interspersed homology 4H-7N_1 are replotted from Fig 2C. (b) The mean number of RIP mutations detected across the 500-bp reference region for repeat constructs with 0–100 base-pairs of perfect homology adjacent to the interspersed homology 5H-6N_1 or 4H-7N_1. (c) RIP mutation profiles for selected repeat constructs containing 0, 15, 50 and 100 base-pairs of perfect homology adjacent to the interspersed homology 5H-6N_1.
Fig 4.
Interplay between the 500-bp interspersed homology 4H-7N_1 and different patterns of interspersed homology in the adjacent 100-bp region.
(a) Reference/test combinations are created for a region derived from the 100-bp insert (Fig 2D: construct v). All interspersed homologies (as described in Fig 1) have a homologous unit length of 6 base-pairs and are varied with respect to periodicity (from 8 to 15) and sequence position (from 1 to 12, for the selected periodicity of 10). We note that the last homologous unit in 6H-4N_2, 6H-4N_3 and 6H-4N_4 was inadvertently extended by 3, 2 and 1 base-pairs, respectively. (b) The mean number of RIP mutations (measured as in Fig 2C) for repeat constructs described in (Panel a). (c) RIP mutation profiles for selected repeat constructs with periodicities of 8, 9, 10 and 11, corresponding to homology patterns 6H-2N_1, 6H-3N_1, 6H-4N_1 and 6H-5N_1, respectively, as described in (Panel a). (d) RIP mutation profiles for selected repeat constructs with sequence positions of 3, 5, 9, and 11, and corresponding to homology patterns 6H-4N_3, 6H-4N_5, 6H-4N_9 and 6H-4N_11, respectively, as described in (Panel a).
Fig 5.
RIP is modulated by the underlying DNA sequence.
(a) The 500-bp interspersed homologies 4H-7N_1 and 4H-7N_7 are tested in the absence of any adjacent homology (“none”, same as “0 bps”) or in combination with the 100-bp interspersed homologies 6H-4N_1 and 6H-4N_7. The last homologous unit in 4H-7N_1 was inadvertently extended by one base-pair. Roman numerals correspond to the following RepeatIDs (provided in S1 Table): i—“XIR” ii—“XKO”, iii—“XJJ”, iv—“XKM”, v—“XKC”, vi—“XKN”. (b) The mean number of RIP mutations (measured as in Fig 2C) for repeat constructs described in (Panel a). (c) RIP mutation profiles for repeat constructs described in (Panel a).
Fig 6.
Identification of a specific triplet sequence that modulates RIP.
(a) The number of triplets included in the homologous units of 4H-7N_1 and 4H-7N_7. Forward and reverse-complemented instances of each triplet are counted together. (b) Variants of the interspersed homology 4H-7N_7 that lack two GAC triplets (Δ2GAC), or all six GAC triplets (Δ6GAC), or seven TGA triplets (Δ7TGA). (c) The mean number of RIP mutations (measured as in Fig 2C) for repeat constructs described in (Panel b). (d) RIP mutation profiles for repeat constructs described in (Panel b). (e) The role of GAC triplets in the interspersed homology 6H-4N_11 was examined using the repeat construct described in Fig 4A. “ΔGAC”: both GAC triplets in the original sequence are deleted. "Mock": analogous mutations are made to delete two non-GAC triplets, as described in the text. (f) The mean number of RIP mutations (measured as in Fig 2C) for repeat constructs in (Panel e). (g) RIP mutation profiles for repeat constructs described in (Panel e).
Fig 7.
Homology recognition and ensuing mutation are not coupled with respect to the involved base-pairs.
(a) RIP mutation profiles of the two diagrammed repeat constructs are compared along their entire lengths. A 200-bp portion of the reference sequence that appears to have similar levels of RIP in both cases is outlined. We note that a distal portion of the same reference sequence (marked with “*”) escapes RIP in the case of 4H-7N_1 but not in the case of 4H-7N_7. (b) RIP mutation profiles along the 200-bp portion of the reference sequence at the base-pair resolution. “●●●●” indicates the positions, in the reference sequence, of the 4-bp homologous units defined by the corresponding test sequences 4H-7N_1 (upper profile) and 4H-7N_7 (lower profile). (c) Correlation of mutations observed for individual sites within the 200-bp segment (Panel b). r is the Pearson correlation coefficient.