Rad24- and Rad25-interacting domains of Byr2

It has been shown that Byr2 is negatively regulated by its interaction with 14-3-3 proteins encoded by rad24 and rad25 (Ozoe et al. 2002). To explore which regions of Byr2 interact with 14-3-3 proteins, we constructed plasmids expressing different regions of Byr2 tagged with HA. Byr2 was separated into seven regions (B2d1–B2d7) (Fig. 1) and each region was expressed in a SPSA (byr2∆) strain from the plasmid pSLF272 or pSLF273. Rad24 or Rad25 were also expressed as GFP fusions. For expression of B2d1–B2d4 and B2d5–B2d7, pSLF272 and pSLF273 vectors, respectively, were used (Fig. S1). Both vectors contain the same nmt41 promoter, but the HA tag is located at the C-terminus of B2d1–B2d4, while in B2d5–B2d7, it is located in the N-terminus. We then examined their interaction by immunoprecipitation. Rad24 or Rad25 was immunoprecipitated with anti-GFP. Immunoblotting of the immunoprecipitates with anti-HA antibody detected various forms of Byr2. Six Byr2-derived proteins, B2d1, B2d2, B2d3, B2d5, B2d6, and B2d7, were co-precipitated with Rad24-GFP and Rad25-GFP, but B2d4 was not (Fig. 2). Interaction of B2d1 and B2d7 with Rad25-GFP was weak, but a band was still detected. Because B2d4 contains the S/T rich domain, but not the Ras1-interacting domain or C-terminal domain, the S/T rich domain is unlikely to be the domain that interacts with Rad24 and Rad25. Thus, the immunoprecipitation experiments indicated that the Ras-binding domain and the C-terminal domain interacted independently with 14-3-3s.

Fig. 1

Domain structure of Byr2. Byr2 consists of four domains. The N-terminus region between residues 1–71 contains the Ste4 binding domain, the region between residues 71–180 contains the Ras-binding domain, the central region between residues 180–392 contains the S/T rich domain, and the C-terminus region between residues 392–659 contains the catalytic domain. Some regions of the byr2 gene were cloned into pSLF272 or pSLF273 and expressed under the control of the nmt41 promoter. B2m, B2d2m, and B2d7m regions containing Byr2 mutants in plasmid pSLF272 or pSLF273 are also shown

Fig. 2

Rad24 and Rad25 interaction domains of Byr2. A SPSA strain (h90byr2::ura4::ADE2) harboring plasmids pSLF272, pSLF272-B2, pSLF272-B2d1, pSLF272-B2d2, pSLF272-B2d3, pSLF272-B2d4, pSLF273-B2d5, pSLF273-B2d6, or pSLF273-B2d7 and pREP41-GFP, pREP41-rad24-GFP, or pREP41-rad25-GFP were grown on EMM2 medium to log phase. Anti-GFP antibody was used for immunoprecipitation of Rad24-GFP and Rad25-GFP and the immunoprecipitates were immunoblotted with anti-HA to detect Byr2-HA tagged proteins

Amino acid residues in the Ras-binding region and C-terminal region of Byr2 important for interaction with Rad24 and Rad25

The consensus sequence of 14-3-3 binding sites are reported to be RSXpSXP (Fu et al. 2000), but no sequence that completely matches this sequence is found in Byr2. Two S/TXP sequences, one situated at residue S136 in the Ras-binding domain and the other situated at residue S654 in the C-terminal domain, are found in Byr2 (Fig. S2). We also considered the 14-3-3 binding site in S. pombe Cdc25 (Zeng and Piwnica-Worms 1999) and selected seven amino acid residues to be mutated: three residues (S87, T94, and S136) in the Ras-binding region and four residues (S402, S566, S650, and S654) in the C-terminal region. Mutations were first introduced into the Ras-interacting domain of Byr2. S87A, T94A, and S136A mutations were introduced in the B2d construct expressing residues 71–180 of Byr2. We first verified that the expression level of various B2d mutants and Rad24-GFP or Rad25-GFP did not vary between the tested trains (Fig. 3A). Then, Rad24-GFP or Rad25-GFP was immunoprecipitated with anti-GFP antibody and the immunoprecipitates were blotted with anti-HA antibody to detect Byr2 containing residues 71–180 (Fig. 3B). Mutation of all three residues (S87A, T94A, and S136A) abolished the interaction with Rad24, while the S87A and T94A mutations compromised the interaction with Rad24. These mutations did not affect interaction with Rad25 (Fig. 3).

Fig. 3

The Ras-interacting domain of the Byr2 mutant and its binding to Rad24 and Rad25. A The Ras-interacting domain mutants of Byr2 (B2d2, B2d2-S87A-T94A, B2d2-S136A, B2d-S87A-T94A-S136A) and Rad24-GFP or Rad25-GFP were expressed in the SPSA strain. Expression of B2d2 and B2d2 mutants was detected by anti-HA antibody, and Rad24-GFP and Rad25-GFP were detected by anti-GFP antibody. B The same strains were used for immunoprecipitation of Rad24-GFP and Rad25-GFP with ant-GFP and immunoprecipitates were immunoblotted with anti-HA antibody to detect Byr2-HA tagged proteins

We next introduced alanine substitutions at residues S402, S566, S650, and S654 in the C-terminal domain (residues 393–659) of the B2d7 construct of Byr2. The mutants were expressed from plasmid pSLF273 together with concomitant expression of Rad24-GFP or Rad25-GFP in the SPSA strain (Fig. 4). We first verified that the expression levels of various B2d7 mutants and Rad24-GFP or Rad25-GFP did not vary between the tested trains (Fig. 4A). Rad24 or Rad25 was immunoprecipitated with anti-GFP antibody and the immunoprecipitates were blotted with anti-HA antibody to detect the Byr2 region including the 393–659 residues (Fig. 4B). The mutation of four amino acid residues (S402, S566, S650, and S654) to alanine abolished the interaction of the B2d7 construct with Rad24 and Rad25, and the mutation of three amino acid residues (S402, S650, and S654) to alanine compromised the interaction of the B2d7 construct of Byr2 with Rad24 and Rad25.

Fig. 4

The C-terminal Byr2 mutants and their binding to Rad24 and Rad25. A The C-terminal domain mutants of Byr2 and Rad24-GFP or Rad25-GFP were expressed in the SPSA strain. Expression of B2d7 and B2d7 mutants was detected by anti-HA antibody, and Rad24-GFP and Rad25-GFP were detected by anti-GFP antibody. B The same strains were used for immunoprecipitation of Rad24-GFP and Rad25-GFP with ant-GFP antibody and immunoprecipitates were immunoblotted with anti-HA antibody to detect the C-terminal Byr2-HA tagged mutant proteins

Mutation of full-length Byr2

To see how the mutations in the full-length Byr2 affected its interaction with 14-3-3s, Byr2 mutants were expressed from the plasmid pSLF272 together with concomitant expression of Rad24-GFP or Rad25-GFP. We first verified that the expression levels of various Byr2 mutants and Rad24-GFP or Rad25-GFP did not vary between the tested strains (Fig. 5A). Rad24-GFP or Rad25-GFP was immunoprecipitated with anti-GFP antibody and the immunoprecipitates were blotted with anti-HA to detect full-length Byr2 (Fig. 5B). Mutation of all seven residues (S87A, T94A, S136A, S402A S566A, S650A, and S654A) abolished the interaction of Byr2 with Rad24 and Rad25. Introduction of five mutations (S87A, T94A, S402A S650A, and S654A) into full-length Byr2 abolished the interaction of Byr2 with Rad24 and Rad25. Mutations of additional residues including these five mutations failed to interact with Rad24 or Rad25. The S402A, S566A, S650A, and S654A mutants failed to interact with Rad24 but weakly interacted with Rad25. The remaining constructs retained their interactions with Rad24 and Rad25. To disable the interaction of Byr2 with 14-3-3s, mutations of amino acids in the Ras-binding domain as well as in the C-terminal domain were necessary, supporting the observation that 14-3-3s interact with Byr2 via both regions.

Fig. 5

Interaction of Rad24 and Rad25 with point mutants of full-length byr2. A Mutants of full-length Byr2 and Rad24-GFP or Rad25-GFP were expressed in the SPSA strain. Expression of Byr2 and Byr2 mutants was detected by anti-HA antibody, and Rad24-GFP and Rad25-GFP were detected by anti-GFP antibody. B The same strains were used for immunoprecipitation of Rad24-GFP and Rad25-GFP with anti-GFP antibody and immunoprecipitates were immunoblotted with anti-HA antibody to detect full-length Byr2-HA tagged mutant proteins

Expression of the Byr2 domain and mutants affects mating

As 14-3-3 proteins interacted separately with the N- and C-terminal domains of Byr2 (Fig. 1), we next tested the effects of expression of different domains of Byr2 on mating. The mating ratios of wild-type strains harboring different plasmids were examined by incubating them on EMM2-N plates at 30 °C for 7 h (Table 2). Expression of B2d8 and B2d9 constructs bearing the N-terminal domain of Byr2 lowered the mating ratio to 10.54% and 14.26%, respectively, while the mating ratio of the wild-type strain (SP870A) harboring the empty vector was 29.6% (Table 2). Expression of the Ras-interacting domains B2d1 and B2d2 lowered the ratio to 17.05% and 15.39%, respectively. Expression of the C-terminal domain (B2d7) of Byr2 increased the mating ratio to 39.88% which is consistent with previous observations (Kjaerulff et al. 2005).

Table 2 Mating ratio in WT by expression of a variety of truncated Byr2

Expression of Rad24 or Rad25 in the wild-type strain lowers the mating ratio, as reported previously (Ozoe et al. 2002). We examined how the expression of N-terminal Byr2, C-terminal Byr2, or full-length Byr2 in cells co-expressing Rad24 or Rad25 affected mating ratios after incubation for 24–41.5 h on EMM2 plates (Fig. 6). In these experiments, nitrogen-containing medium was used. While the mating ratio in the strain harboring the vector control was 15–20% (Fig. 6A), expression of full-length Byr2 elevated the mating ratio to 25–30% and expression of Rad24 slowed the time to maximum mating (Fig. 6B). The mating ratio in the strain expressing the N-terminal domain of Byr2 was 5–10% (Fig. 6C) after the same period of incubation. Expression of Rad24 or Rad25 did not affect the mating ratio in the strain expressing the N-terminal domain of Byr2 (Fig. 6C), indicating a dominant negative effect of the N-terminal domain of Byr2 on mating is not interrupted by expression of 14-3-3 proteins. Expression of the C-terminal domain of Byr2 rapidly increased the mating ratio to 30–35% (Fig. 6D) and expression of Rad24 or Rad25 reduced the time needed to observe the increase in the mating ratio. These results clearly show a negative regulation of the C-terminal function of Byr2 by 14-3-3 proteins.

Fig. 6

Effect of expression of Byr2 mutants on mating efficiency in cells expressing Rad24 or Rad25. The SP870A strain harboring pSLF172 (A), pSLF172-B2 expressing full-length Byr2 (B), pSLF172-B2d1 expressing the N-terminal region of Byr2 (C), or pSLF172-B2d7 expressing the C-terminal region of Byr2 (D) together with pREP41 (▲), pREP41-rad24 (●) or pREP41-rad25 (■) was used to examine mating ratios. The SPSA (byr2∆) strain harboring plasmids pSLF172-B2 (E), pSLF172-B2-S402A-S650A-S654A (F), pSLF172-B2-S87A-T94A (G), pSLF172-B2-S402A-S566A-S650A (H), pSLF172-B2-S402A-S566A-S654A (I), or pSLF172-B2-S566A-S650A-654A (J) together withpREP41 (▲), pREP41-rad24 (●) or pREP41-rad25 (■) was used to examine mating ratios. Cells were grown on EMM2 plates for indicated times (h) and 1,000 cells were counted for mating and sporulation. Mean and standard deviation of duplicate samples are shown

Because we observed some mutations in Byr2 compromised their interaction with 14-3-3 proteins (Fig. 5), we next monitored the mating efficiency of the SPSA (byr2∆) strain harboring different plasmids expressing mutant types of Byr2. While introduction of some mutations into Byr2 completely abolished its function, S402A, S650A, S654A or S87A and T97A mutations did not affect Byr2 function (Fig. 6F, G). The inhibitory effect of Rad25 was clearly reduced in the strain expressing the Byr2–S402A–S650A–S654A mutant, but mating was still inhibited by Rad24 expression (Fig. 6F). Introduction of the mutations S402A, S566A, and S650A, the mutations S566A, S650A, and S654A, or the mutations S402A, S566A, and S654A into full-length Byr2 abolished its complementary function in the byr2∆ strain (Fig. 6H, I, J). Therefore, S566 is crucial for Byr2 function. It was difficult to test the effect of 14-3-3s in the constructs containing the S566A mutation. Introduction of S87A and S94A into full-length Byr2 rather elevated Byr2 activity in the byr2 deletion strain. In this stain, the inhibitory effect of Rad24 was moderate, unlike Rad25, which still inhibited mating (Fig. 6G), supporting the observation that the Byr2–S87A–S94A mutant was still able to interact with Rad25 (Fig. 3). Because Rad24 and Rad25 interact with other proteins such as Ste11 and Mei2 (Kitamura et al. 2001), the effect of 14-3-3s on mating needs to take into consideration of these proteins’ role in addition to Byr2.