Temperature-sensitive mutations made easy: generating conditional mutations by using temperature-sensitive inteins that function within different temperature ranges

Abstract

Reversible and easy to use, temperature-sensitive (TS) mutations are powerful tools for studying gene function. However, TS alleles are rare and difficult to generate and identify, and this has limited their use in most multicellular organisms. We have generated and characterized 41 intein switches, temperature-sensitive Sce VMA mutations that splice only at the permissive temperatures to generate intact host proteins. At nonpermissive temperatures, they fail to splice, resulting in a loss of function of the proteins in which they reside. By inserting an intein switch into a protein of interest, one can turn on and off the activities of the engineered protein with a simple temperature shift. The 41 TS inteins function in five different temperature ranges, with permissive temperatures ranging from 18 degrees to 30 degrees . This collection makes it possible to choose a TS-intein switch according to the optimal growth temperature of an organism or to suit a special experimental design.

Figure 1.—

Generation and characterization of TS inteins. (A) Principle of TS-intein function. (B) Flowchart of steps used to generate the second-generation TS inteins and to characterize the TS-intein mutations from both the first and second generation. See text for details. Numbers on the right represent the number of TS alleles retained at each step. An asterisk indicates the number of first-generation TS-intein alleles retained at each step.

Figure 2.—

Identification of TS inteins that splice only in a limited temperature range. (A) Screening for the second-generation TS inteins. About 3000 clones were screened (156 are shown). Colonies that grew well at 18° (top) but not at 30° (bottom) are circled. (The small circle indicated by an arrow in the bottom panel is a bubble.) (B) Eliminating leaky TS-intein alleles. Images show the growth profiles of yeast rescued by GAL4 containing selected intein alleles expressed from the low-copy-number plasmid pS5 (top) and the high-copy-number plasmid pU (bottom). Each row represents an eightfold dilution of the previous concentration. S16 and S22 behaved like other TS-intein alleles at 18°, but supported much better yeast growth at 30° than other mutations did. Note that Gal4-intein^WT (second column) allowed yeast to grow equally well as Gal4p without the intein inserted (first column), while intein^Dd (third column) made Gal4p nonfunctional, regardless of the temperature.

Figure 3.—

Growth profiles of yeast FY761 transformed with different pU-Gal4 intein alleles. From top to bottom, intein alleles are listed in order of increasing permissive temperatures, except for controls at the very bottom. Note that the yeast at the bottom grew better than those at the top. Although the transition from one allele to the next is gradual, for the convenience of description, we grouped the TS-intein alleles into five groups. Each row in a panel represents a 1:8 dilution series for the specified intein allele. SD-Ura: synthetic media lacking uracil, dextrose as sugar source; SG-Ura: synthetic media lacking uracil, galactose as sugar source; WT: GAL4 with wild-type intein inserted; Dd: GAL4 with dead intein inserted; F1 to F19 are first-generation TS-inteins; S1 to S88 are second-generation TS-inteins; three-digit numbers in all figures: ID numbers for the TS inteins.

Figure 4.—

Protocol for protein expression and assessing splicing. Yeast was first grown in the presence of methionine to the desired log-phase density and then placed in methionine-free media to induce transcription. Half an hour later, methionine was added to stop transcription. The yeast cultures were then divided and incubated at various temperatures for protein splicing. The results were analyzed by Western blotting.

Figure 5.—

Western blot of Gal4-intein splicing. (A) Group I. (B) Group II. (C) Group III. (D) Group IV. (E) Group V. (F) Controls. Temperatures (20° and 25°) common for testing all TS-intein alleles are in boldface type. The top band in each panel is the unspliced precursor protein, while the bottom band is the spliced product Gal4p (labeled only for alleles 307 and 308). The alleles marked with an asterisk splice efficiently in both Gal4p and eGFP (Figure 6 and Figure S3) and are recommended for future use. The numbering and marking system used in this figure remains the same in the figures that follow, as well as in the figures in the supporting information. Note that the TS-intein alleles that enabled the host yeast to grow at higher temperatures also spliced at higher temperatures.

Figure 6.—

Comparison of the splicing of selected TS-intein alleles in different host proteins. (A) Group I. (B) Group II. (C) Group III. (D) Group IV. (E) Group V. (F) Controls. WT: eGFP with wild-type intein inserted; Dd: eGFP with dead intein inserted. Temperature (20°) common for testing all TS-intein alleles is in boldface type. Note that TS-intein alleles spliced similarly in both host proteins. As in Gal4p, intein^WT splicing in eGFP was temperature independent, and intein^Dd did not splice from its host protein eGFP.

Figure 7.—

Time course of intein splicing in BasGFP. Western blot analysis with anti-GFP antibody of proteins from yeast carrying a BasGFP intein incubated for the indicated periods after stopping transcription induction at 18° or 30°. (A) Time courses of the splicing of intein^WT and intein⁵⁰³. At 30°, intein^WT (top left) spliced rapidly, while intein⁵⁰³ (top right) barely spliced. At 18°, intein⁵⁰³ (bottom right) spliced at a reasonable speed but was slower to splice than intein^WT (bottom left). (B) Time courses of the splicing of intein^WT and intein^Dd. Dead intein did not splice from its host protein. Note that the time courses of the splicing of a BasGFP-intein^WT at 30° done on 2 different days (top left in A and right in B) were very similar, indicating that the splicing reaction is reproducible. (C) Quantification of BasGFP-intein splicing. The y-axis is the ratio of the amount of unspliced BasGFP intein to the total amount of BasGFP and BasGFP intein. The x-axis is the time (in minutes) when yeast cells were harvested. The moment at which transcription induction was stopped was time zero. The calculated, pseudo-first order, splicing halftime is presented following the temperature of splicing and the name of each intein allele.