Fig. 1. Cic1 interacts with Pre6. (A) The Cic1 C-terminus is necessary and sufficient for interaction with Pre6 (amino acids 174–253) in the two-hybrid system. Yeast cells expressing the indicated GAL4 gene fusion were lysed in the presence of X-gal. Dark colony patches result from interaction of the plasmid-encoded proteins. Pre6 (amino acids 147–253) interacts with the C-terminus of Cic1 (amino acids 309–376) and with full-length Cic1 (amino acids 1–376) but not with C-terminally truncated Cic1 (amino acids 1–316). (B) Cic1 interacts with Pre6 in vitro. GST–Pre6 was purified from E.coli with gluthatione–Sepharose (Pharmacia). Immobilized GST–Pre6 and GST were incubated with in vitro translated [³⁵S]methionine-labeled Cic1. Cic1 binds to immobilized GST–Pre6 but not to GST. (C) Cic1 is essential for vegetative growth. Sporulation and tetrad dissection of a heterozygous CIC1/cic1Δ strain containing pSJ32 (CEN URA3 CIC1). All spores can grow on YPD and Sc-uracil plates. The spores containing the kanamycin deletion marker (growth on YPD G418) cannot grow on plates supplemented with 5-FOA, indicating that they cannot lose the URA3-marked plasmid containing CIC1. (D) Sequence similarities between Cic1, human Pbk1 (PIR accession No. T08693), yeast Ykr060w and a hypothetical protein (PIR accession No. T39179) from S.pombe. The sequences were aligned using the ClustAl program (HUSAR Version 3.0, DKFZ Heidelberg, Germany). Gaps (indicated by dots) were used to maximize alignments. Residues identical in at least two of the proteins are shaded in black.

Fig. 2. Cic1 is associated with the 26S proteasome. (A) Cic1 co-precipitates with proteasomal components. Extracts from cells expressing HA3::CIC1 (YS70) were subjected to immunoprecipitation with anti-HA antibodies chemically cross-linked to protein A–Sepharose. HA3-Cic1 was pulled down and the precipitate was analyzed by immunoblotting with anti-Pre6, anti-Scl1, anti-Pre4 and anti-Cim5 antibodies as indicated. The proteasomal subunits are present in the complex with precipitated HA3-Cic1 but not in a strain with untagged Cic1 (YS60). (B) Cic1 co-fractionates with the 19S subunit Cim5. An extract of the wild-type strain W303 was fractionated by gel filtration on Superose 6. The upper curve shows the proteasomal chymotrypsin-like activity in the relevant fractions (12–30). The void volume (v₀) and the marker proteins are indicated by dashed lines. The fractions were analyzed by SDS–PAGE, followed by immunodetection, with anti-Cic1, anti-Cim5, anti-Pre6 and anti-Pre4 antibodies. (C) A strain expressing HA3::CIC1 (YS70) shows the same distribution of HA3-Cic1 as Cic1 in the wild-type strain. (D) Analysis of Superose 6-fractionated proteasomal complexes from YS70 [HA3-Cic1; see (C)] by non-denaturating gel electrophoresis. Following western transfer, immunodetection was performed with anti-Pre6 and anti-Cim5 antibodies, respectively (left panel). Both blots were stripped and reprobed with anti-HA antibodies (right panel) to detect HA3-Cic1 that is only present in the 26S proteasome (RP₂CP).

Fig. 3. Cic1 localizes to the nucleus and is concentrated in a crescent-shaped part. Fixed and permeabilized YS70 cells were incubated simultaneously with 12CA5 (directed against the HA epitope of Cic1; upper row, left panel) and anti-Pre6 antibodies (lower row, left panel). 12CA5 was visualized with ALEXA₄₈₈-conjugated goat anti-mouse IgG, and anti-Pre6 antibodies with ALEXA₅₄₆-conjugated goat anti-rabbit IgG. DNA was stained with DAPI (upper row, right panel). The middle panel (upper row) shows a merge of antibody staining patterns of 12CA5 and DAPI. The middle panel (lower row) shows a merge of anti-Pre6 and 12CA5 staining. Untagged Cic1 wild-type cells show no staining with 12CA5 (data not shown).

Fig. 4. Cic1 is not necessary for general proteasomal degradation. (A) cic1-2 mutant cells (YS23) grow like wild-type (YS60) at 25°C, but show a strong growth phenotype at the restrictive temperature of 37°C. (B) Cic1 is not required for general proteasomal degradation. Pulse–chase analysis of Leu-β-gal and Ub-Pro-β-gal degradation in YS60 and YS23. Early logarithmically growing cultures were shifted to 37°C for 8 h prior to labeling with [³⁵S]methionine. Multiubiquitylated β-gal species are denoted with open-ended brackets. Asterisks indicate a stable 90 kDa cleavage product characteristic for short-lived β-gal-derivatives. The open square marks stable Pro-β-gal resulting from partial hydrolysis of Ub-Pro-β-gal. Quantification reveals no stabilization of the substrates in cic1-2 cells.

Fig. 5. Cic1 function is required for turnover of cell cycle regulatory proteins. (A) Cdc4 is stabilized in a cic1-2 mutant. Pulse–chase analysis of degradation of Cdc4 in wild-type (YS60.3) and cic1-2 mutant cells (YS23.3). Early logarithmically growing wild-type and cic1-2 mutant cells, expressing HA3-Cdc4 under control of the GAL1 promoter, were shifted to 37°C for 8 h prior to labeling with [³⁵S]methionine. Quantification reveals an ∼2.5-fold stabilization of Cdc4 in the cic1-2 mutant. (B) Grr1 is stabilized in a cic1-2 mutant. Pulse–chase analysis of degradation of Grr1 in wild-type (YS60.3) and cic1-2 mutant cells (YS23.3). Early logarithmically growing wild-type and cic1-2 mutant cells, expressing HA-Grr1^ΔN under control of the ADH1 promoter, were shifted to 37°C for 8 h prior to labeling with [³⁵S]methionine. Quantification shows that the half-life of Grr1 in the cic1-2 mutant is increased ∼2.5-fold. (C) Cic1 is not involved in ubiquitylation of Cdc4. Cultures of strain YS23.3 expressing only HA-ubiquitin, Flag-Cdc4 and untagged ubiquitin, or Flag-Cdc4 and HA-ubiquitin, and strain YS60.3 expressing Flag-Cdc4 and HA-ubiquitin were shifted to 37°C for 8 h prior to precipitation with anti-FlagM2 affinity gel. The precipitate was analyzed by immunoblotting with anti-HA antibodies. Both strains show the same ubiquitylation pattern of Cdc4. To confirm that the high molecular weight bands are due to ubiquitylation, we analyzed precipitates of YS60.3 expressing HA-ubiquitin, Flag-Cdc4 and untagged ubiquitin, or Flag-Cdc4 and HA-ubiquitin, with anti-Flag and anti-HA antibodies. With anti-Flag antibodies, the difference in molecular weight between ubiquitin residues and HA-tagged ubiquitin residues bound to Flag-Cdc4 can be detected. With anti-HA antibodies, only Flag-Cdc4 with HA-tagged ubiquitin residues can be detected.

Fig. 6. Cdc4 interacts with Cic1. (A) Cdc4 interacts with Cic1 in vitro. Flag-Cdc4 was precipitated with an anti-FlagM2 affinity gel from a pre1-1 pre2-2 strain expressing Flag-Cdc4. The precipitate was incubated with in vitro translated [³⁵S]methionine-labeled Cic1. The precipitate from cells carrying the expression vector alone was used as a control. Cic1 binds to immobilized Flag-Cdc4. (B) Cdc4 interacts with Cic1 in vivo. Extracts from cells expressing Flag-Cdc4 and expression vector, respectively, were subjected to immunoprecipitation with anti-FlagM2 affinity gel. The precipitated proteins were analyzed by immunoblotting with anti-Cic1 and anti-Pre6 antibodies.