Slx5p‐Slx8p Promotes Accurate Chromosome Segregation by Mediating the Degradation of Synaptonemal Complex Components during Meiosis

Abstract Meiosis increases genetic diversity, yet the genome complement needs to be stable to ensure offspring viability. Both small ubiquitin‐like modifier (SUMO) and ubiquitin have been reported to participate in meiotic regulation, yet functions of the SUMO‐ubiquitination crosstalk in meiosis remain unclear. Here, it is reported that a SUMO‐targeted ubiquitin ligase, Slx8p, promotes accurate chromosome segregation during meiosis, since the deletion of SLX8 leads to increased aneuploidy due to a defect in synaptonemal complex (SC) component degradation. Both the RING domain and SUMO interacting motifs of Slx8p are essential for meiotic progression and maintaining spore viability, and the expression of tetraubiquitin fused with SUMO partially rescues meiotic defects in the SLX8‐deletion strain. Furthermore, Slx5p‐Slx8p can directly add ubiquitin to SUMOylated Zip1p and Ecm11p, and forced degradation of Ecm11p partially rescues the sporulation defects of the SLX8 deletion strain. These findings provide a mechanism for SC disassembly and reveal that the crosstalk between SUMOylation and ubiquitination facilitates accurate chromosome segregation by promoting SC component degradation during meiosis.


Introduction
Meiosis is a fundamental process of sexual reproduction in eukaryotes, which halves the genome by a single round of DNA replication followed by two consecutive cell divisions: meiosis the length of meiotic chromosome axes in two parallel tracks, a series of transverse filaments (TFs) that span the region between the two homologs and bridge the parallel homologous axes; and the central element (CE) proteins that are located along the center of the SC to stabilize the complex. [6,7] Several components of the SC have been identified: Red1p, Hop1p, and Mek1p are parts of LEs; Zip1p exists in TFs; Ecm11p, Gmc2p exist in CE in Saccharomyces cerevisiae. [3,7] Although the components of the SC share little conservation at the amino acid sequence level among different organisms, the ultrastructure and biological functions of the SC are similar to each other in most eukaryotes. [7][8][9] The SC dynamically changes during MI and goes through three stages that include assembly, a highly dynamic steady state, and disassembly. [3] In budding yeast, SC assembly begins with the recruitment of Zip3p and Zip1p at the sites of meiotic recombination events. [10,11] These events are followed by the interaction of SUMOylated Red1p with Zip1p and further recruitment of Zip2p, Zip4p, and Spo16p to form the synapsis initiation complex (SIC). [12,13] The maturation of the SC is regulated by a positive feedback system: Zip1p initiates the SUMOylation of Ecm11p; poly-SUMOylated Ecm11p promotes more Zip1p assembly; and then more Zip1p facilitates further loading of Ecm11p/Gmc2p and their SUMOylated forms. [14] At the pachytene stage, the SC is assembled at the interface of nearly all lengthwise-aligned homologous chromosome pairs. [15] Upon pachytene exit, the SC starts to disassemble in an asymmetric manner and SC proteins are retained at specific chromosome subdomains until late prophase I. [16,17] Compared with SC assembly, the mechanism underlying SC disassembly is not well understood.
During meiosis, both SC formation and disassembly are essential for proper segregation of homologs. [2,7,18] The SC formation can stabilize the chromosome structure and sense the presence of aberrantly entangled chromosomes. [19] Once the SC is disassembled, crossovers tether the homologs to each other and orient them for segregation. [2] In addition to mediating the close apposition of homologous chromosomes during mid-meiotic prophase, some SC components participate in centromere coupling and pairing. In the leptotene, Zip1 first appears as dispersed punctate foci at or near the centromeres, and mediates the homology-independent association of centromeres. [20,21] When the SC disassembles in late prophase, Zip1p remains at the paired centromeres, leaving the homologous partners joined by only chiasmata and centromere pairing. [22] Zip3p is also important for nonexchange chromosome segregation, and it was shown that ZIP3 deletion impairs both the tethering of nonexchange chromosomes and the localization of Zip1p to centromeres. [22] The disassembly of the SC at paired centromeres could allow for chromosome segregation at meiosis I. [7] Post-translational modifications (PTMs) of SC-related proteins, such as SUMOylation, phosphorylation, and N-terminal acetylation, have emerged as important regulators of SC dynamics. [3,7] Recent studies have reported that SUMO, ubiquitin, and proteasomes coordinately work to regulate the key events of meiotic prophase I. [23,24] RNF212 and HEI10 define an axis-associated, SUMO-ubiquitin-proteasome relay that may mediate turnover of a subset of recombination factors that act after DMC1-promoted homolog pairing. [23] SUMO E3 ligase Zip3p and SC protein Zip1p are required for proteasome particle recruitment to chromosomes during meiosis, which is essential for SC assembly and crossover formation. [24] However, the functional role and molecular mechanism underlying the SUMOubiquitin-proteasome system in prophase I remains poorly understood, and whether a mediator exists to conduct these PTM signals in prophase I still requires further investigation.
The SUMO-targeted ubiquitin ligase (STUbL) is defined as a specific class of ubiquitin ligases (E3) that recognize the SUMOylated substrates via SUMO-interaction motifs and ubiquitinate them via the RING domain, and this process is indispensable for the crosstalk between SUMOylation and ubiquitination. [25] To delineate the functional roles of SUMOubiquitin crosstalk during meiosis, we tested all the potential functions of the STUbLs during yeast meiosis. We found that only the disruption of Slx5p-Slx8p significantly influenced the meiotic progression in yeast, where it mainly participated in the degradation of SC components. The function of Slx8p during meiosis is dependent on its STUbL activity, and we identified Ecm11p as the key substrate. These findings reveal that the crosstalk between SUMOylation and ubiquitination promotes proper chromosomal segregation by mediating the degradation of SC components during meiosis.

Slx5p-Slx8p Is the Key STUbL that Regulates Meiosis
Three STUbLs have been reported to play roles in budding yeast, including Uls1p, Rad18p, and Slx5p-Slx8p. [25] To study the functional roles of the STUbLs during meiosis, we tested all STUbL enzymes by deleting ULS1, RAD18, SLX5, and SLX8 in SK1 background S. cerevisiae. After incubating in sporulation medium for 24 h, we found SLX5-and SLX8-deleted cells showed obvious impairment in the sporulation process ( Figure 1A). Furthermore, the double deletion of SLX5 and SLX8 also influenced the meiotic progression compared to the wild-type (WT) strain ( Figure 1A), suggesting that Slx5p-Slx8p was the key STUbL responsible for regulating yeast sporulation. To further confirm our finding, we evaluated the meiotic progression in WT, slx5Δ, slx8Δ, and slx5Δ slx8Δ strains, and found that sporulation in slx5Δ, slx8Δ, and slx5Δ slx8Δ strains showed a delay at the early stages of meiosis ( Figure 1B). Therefore, Slx5p-Slx8p is the key STUbL for the regulation of meiosis, and the disruption of Slx5p-Slx8p obviously impairs meiotic progression. As it was difficult to detect the spore formation in slx5Δ and slx5Δ slx8Δ cells after incubating in sporulation medium for 24 h ( Figure S1A, Supporting Information), we examined the spore viability in slx8Δ cells and found most of the tetrad-derived spores in the slx8Δ mutant cells were inviable ( Figure 1C,D). This result indicates Slx8p should be required for proper chromosome segregation during meiosis, and we selected Slx8p for further investigation.

The Disruption of SLX8 Impairs Meiotic Cell Division
To investigate the physiological function of Slx8p, we evaluated its expression during yeast meiosis. We found that the protein www.advancedscience.com Figure 1. Slx5p-Slx8p is the key STUbL to regulate meiosis. A) Functional screening of STUbL genes during yeast meiosis. WT and mutant strains were induced to sporulate and analyzed after 24 h. The divided nuclei were stained with DAPI, and more than 400 nuclei were scored for each experiment (n = 3 independent experiments). Data are presented as mean ± SEM. **P < 0.01. Two-sided t-tests were used for statistical analyses. B) The disruption of SLX5 and SLX8 impaired meiotic progression. WT, slx5Δ, slx8Δ, and slx5Δ slx8Δ cells were induced to undergo meiosis by transfer to SPM and analyzed at different time points, with more than 180 nuclei scored for each time point (n = 3 independent experiments). Data are presented as mean ± SEM. C) The spore viability of WT and slx8Δ cells. Tetrad analysis of WT and slx8Δ diploids was performed by dissection of tetrads after sporulation. D) Quantification of spore viability in WT and slx8Δ cells. More than 200 tetrads were scored for each time point (n = 3 independent experiments). Data are presented as mean ± SEM. Two-sided t-tests and G-tests were used for statistical analyses. The asterisks indicate statistically significantly improved viability frequency of slx8Δ cells compared to WT cells (P < 0.0001, G-test for homogeneity). P = 0.0014 proportion of four-viable spores (t-test, WT vs slx8Δ). E) Slx8p was predominantly expressed at the early stage of meiosis. The WT strain expressing Slx8p-9MYC in one allele was incubated in SPM, and samples were collected at different time points after sporulation induction. Expression of Slx8p-9MYC was analyzed over time by immunoblotting using an anti-MYC antibody. Pgk1p served as a loading control. F) Ratios of Slx8p/Pgk1p levels were normalized to the maximum Slx8p/Pgk1p ratio (n = 3 independent experiments). Data are presented as mean ± SEM. G) Premeiotic DNA replication was delayed in the slx8Δ strain during sporulation.

www.advancedscience.com
was predominantly expressed at the early stages of meiosis ( Figure 1E,F), indicating Slx8p might regulate early events of meiosis. To identify meiotic processes controlled by Slx8p, we first compared premeiotic DNA synthesis in the slx8Δ strain with that in WT cells using flow cytometry analysis. Although the premeiotic DNA replication showed a minor delay during the early stage of sporulation after SLX8 depletion, DNA replication was completed after 6 h in sporulation medium in both WT and slx8Δ cells ( Figure 1G; Figures S1B and S2, Supporting Information). This result suggests that the disruption of SLX8 has a modest effect on premeiotic DNA replication. Accompanying premeiotic DNA replication, the meiosis-specific cohesin subunit, Rec8p, is known to replace the mitotic cohesin subunit Mcd1p to form the predominant form of cohesin during meiosis. [26] We next observed the expression of the cohesin subunit during meiosis and found Mcd1p degradation and Rec8p expression were both somewhat delayed in slx8Δ cells ( Figure 1H-J). Furthermore, the degradation of Rec8p was obviously delayed in slx8Δ cells ( Figure 1J,K). Taken together, the SLX8 deletion delays premeiotic DNA replication and meiotic cohesin expression, and impairs cell divisions during meiosis.

SLX8 Deletion Leads to Extended Prophase I and Metaphase I
To further determine the precise phases of meiosis that were affected in slx8Δ cells, we examined the nuclear division and spindle formation in WT and slx8Δ cells by analyzing the states of the spindle and the spindle pole body (SPB) using mCherrytubulin and CNM67-GFP (Figure 2A). At 3 h postincubation, we found ≈80% of the WT and slx8Δ cells were in prophase I (Figure 2A,B). At 6 h, ≈50% of WT cells had exited prophase I, whereas more than 70% of slx8Δ cells were arrested in prophase I (Figure 2A,B). Even at 9 h, more than 40% of slx8Δ cells remained arrested in prophase I (Figure 2A,B). In addition, the proportion of cells in metaphase I was also significantly increased in slx8Δ cells at 9 h (Figure 2A,B). These results suggest that the disruption of Slx8p extends the period of meiotic prophase I and metaphase I, and Slx8p might play multiple functional roles during meiosis.

Slx8p Is Required for SC Component Degradation
Because meiotic prophase I can be further divided into leptotene, zygotene, pachytene, and diplotene stages as defined by the state of the SC, [27] we characterized the various stages using Zip1p staining at different time points during meiosis. At 3 h, the foci of Zip1p (leptotene) and dotty linear structures of Zip1p (zygotene) appeared in both WT and slx8Δ cells. This was followed by full linear Zip1p staining (pachytene) observed from 6 h ( Figure 2C,D). This result indicates the SC assembly occurred similarly in WT and slx8Δ cells. In the WT cells, the proportion of nuclei with entirely linear Zip1p decreased from 6 to 12 h, whereas the full linear SC was observed for an extended period in the absence of SLX8 ( Figure 2C,D), indicating that the pachytene period was extended in slx8Δ cells. To further validate our findings, we examined the expression of Ndt80p, which is an essential transcription factor for the exit from the pachytene stage. [28] In WT cells, Ndt80p was highly expressed at 6 h and decreased during sporulation ( Figure 2E,F), whereas the transcription factor was highly expressed from 8 to 12 h in slx8Δ cells ( Figure 2E,F), suggesting that SLX8 deletion stalls the pachytene exiting.
The extended persistence of the mature SC could be indicative of impaired meiotic recombination or delayed SC disassembly. [29][30][31] To distinguish between these two possibilities, we followed the degradation of SC components, including Red1p (LE), Zip1p (TF), and Ecm11p (CE), and the meiotic nuclear divisions. We found no obvious differences in the degradation of Red1p in slx8Δ cells compared with WT cells during meiosis (Figure 3A,B). However, some Zip1p protein was retained from 8 to 12 h, even in some proportion of cells that had finished at least one meiotic nuclear division at the same time of slx8Δ cells ( Figures 1B and 3C,D), indicating that the Zip1p degradation may be impaired in slx8Δ cells. Similar degradation defects could also be detected in Ecm11p ( Figure 3E,F). It has been reported that the Zip1p polycomplexes (PCs) are frequently juxtaposed with duplicated SPBs, [31] thus Zip1p PC accumulation suggests a SC component degradation defect in slx8Δ cells ( Figure S3, Supporting Information).

Slx8p Is Required for Accurate Chromosome Segregation during Meiosis
To exclude the effect on meiotic recombination in slx8Δ cells, we engineered spo11Δ and spo11Δ slx8Δ strains to block meiotic DSB and SC formation. [32] We found the spore viability in spo11Δ slx8Δ cells was similar to that in spo11Δ cells ( Figure  S4B,C, Supporting Information), yet deletion of SPO11 could not rescue the sporulation defect in slx8Δ cells ( Figure S4A, Supporting Information). This result suggests the meiosis delay conferred by the SLX8 deletion was not exclusively linked to meiotic recombination. It has been reported that the SC formation is severely defective in spo11Δ cells, [33] and Zip1p remains at the centromeres in spo11Δ cells to couple chromosome centromeres. [21] Absence of SC assembly due to the SPO11 deletion did not rescue the slx8Δ sporulation defect, indicating that the defect of SC disassembly alone on the chromosome arm region is not responsible for the slx8Δ sporulation defect ( Figure S4A, Supporting Information). At the same time, the dotty Zip1p signal was dramatically increased in spo11Δ slx8Δ cells compared to spo11Δ cells ( Figure S4D, Supporting Information). In slx8Δ cells, the proportion of dotty Zip1p signal and Zip1p PCs were significantly higher at 12 h than in the control groups ( Figure 2D; Figure S3, Supporting Information), and the proportion of Zip1p signal in the separated nuclei during meiotic division was increased after the deletion of SLX8 ( Figure S5, Supporting Information). These results indicate Slx8p may modulate SC component degradation along chromosome arms, centromeres, or even in regions not associated with chromatin and thereby facilitating meiotic divisions. As it has been proposed that defective SC disassembly might lead to homologous chromosome missegregation, [3,7] we next www.advancedscience.com Figure 2. The disruption of Slx8p causes extended prophase I and metaphase I. A) Characterization of meiotic cell cycle by SPB and spindle. CNM67-GFP (green)/mCherry-TUB1 (red) cells were sporulated in SPM and samples were collected at indicated time points, and nuclei were stained with DAPI (blue). B) The proportion of WT and slx8Δ cells at different meiotic stages. More than 180 cells were scored for each time point (n = 3 independent experiments). Values of P < 0.0001 prophase I at 6 h (WT vs slx8Δ), P < 0.0001 prophase I at 9 h (WT vs slx8Δ), and P < 0.0001 metaphase I at 9 h (WT vs slx8Δ) were observed. Data are presented as mean ± SEM. Chi-square tests were used for statistical analyses. C) Characterization of prophase I by morphology of SC (ZIP1-GFP). WT or slx8Δ cells transfected with pZIP1-GFP 700 were incubated in SPM, and harvested at 6 h, and meiotic chromosomes were spread for immunofluorescence microscopy, and nuclei were stained with DAPI (blue). D) The proportion of WT and slx8Δ cells at different prophase I stages. More than 200 nuclei were scored for each time point (n = 3 independent experiments). Values of P < 0.0001 dotty Zip1 at 12 h (WT vs slx8Δ), P < 0.0001 dot-liner Zip1 at 12 h (WT vs slx8Δ), and P < 0.0001 full SC at 12 h (WT vs slx8Δ) were observed. Data are presented as mean ± SEM. Chi-square tests were used for statistical analyses. E) Expression of Ndt80p was detected in WT and slx8Δ cells by immunoblotting during sporulation. WT or slx8Δ cells expressing the NDT80-9 × MYC in one allele were incubated in SPM, and samples were collected at different time points. Pgk1p served as a loading control. F) Ratios of Ndt80p/Pgk1p levels were normalized to the maximum Ndt80p/Pgk1p ratio (n = 3 independent experiments). Data are presented as mean ± SEM. See also Figure S3 in the Supporting Information.
www.advancedscience.com investigated whether Slx8p was required for proper chromosome segregation during meiosis using the TetO-TetR system. TetO arrays were integrated at HIS4 on chromosome III, and chromosome segregation was examined in the presence of TetR-GFP [34] (Figure 3G). We detected more than 10% missegregated chromosomes in SLX8-deleted cells during meiosis, but no more than 3% in WT cells ( Figure 3H,I). Our findings indicate that Slx8p is required for accurate chromosome segregation during meiosis.

Slx8p's Role in Meiosis Is Dependent on Its STUbL Activity
It has been reported that mutations in SLX8 result in slow growth and hypersensitivity to hydroxyurea due to the accumulation of SUMOylated proteins during mitosis. [35,36] Thus, we first characterized the SUMO and SUMOylated protein levels in slx8Δ cells during meiosis. Both in mitosis and meiosis, SLX8 deletion led to the accumulation of SUMOylated proteins and a reduction of SUMO monomers ( Figure 4A). Based on findings that showed growth defects caused by slx8Δ could be rescued by the deletion of SUMO ligase SIZ1 in mitosis, [36] we tested whether the deletion of SUMO ligase could rescue the sporulation defect in slx8Δ cells. By performing a small-scale screening of double or triple SUMO ligase mutants, we found that the sporulation efficiency of the mutants showed no significant differences compared with slx8Δ cells ( Figure S6, Supporting Information), suggesting the functional role of Slx8p during meiosis should be different from its functional role in mitosis.
Adv. Sci. 2020, 7,1900739   , and Slx8p simΔ . C,D) Slx8p participating in yeast meiosis is dependent on its E3 activity. The proportion of C) divided nuclei and D) spore viability in WT carrying pRS313, slx8Δ carrying pRS313, slx8Δ carrying pRS313-SLX8, and slx8Δ carrying pRS313-SLX8 rm cells. More than 150 nuclei and spores were scored for each experiment (n = 3 independent experiments). Data are presented as mean ± SEM. *P < 0.05. **P < 0.01. Two-sided t-tests were used for statistical analyses. E,F) Slx8p participating in yeast meiosis is dependent on its SUMO binding activity. The proportion of E) divided nuclei and F) spore viability in WT carrying pRS313, slx5Δ carrying pRS313, slx8Δ carrying pRS313, and slx5Δ slx8Δ carrying pRS313, slx8Δ carrying pRS313-SLX8 and slx8Δ carrying pRS313-SLX8 simΔ cells, slx5Δ slx8Δ carrying pRS313-SLX8, and slx5Δ slx8Δ carrying pRS313-SLX8 simΔ cells. More than 150 nuclei and spores were scored for each experiment (n = 3 independent experiments). Data are presented as mean ± SEM. *P < 0.05. **P < 0.01. Two-sided t-tests were used for statistical analyses. G) Slx8p loads onto meiotic chromosomes during prophase I. WT carrying pZIP1-GFP 700 and pRS313-SLX8-FLAG, and slx5Δ slx8Δ carrying pZIP1-GFP 700 and pRS313-SLX8 simΔ -FLAG cells were incubated in SPM and harvested at 6 h, and meiotic chromosomes were spread for immunofluorescence. Nuclei were stained with DAPI (blue). The pixel overlaps of Zip1p and Slx8p were quantified using the IMARIS software. H) The artificial Ub4-SUMO hybrid could be detected in yeast. Immunoblotting of FLAG was performed in WT carrying pYEP-CUP1-UB4-SMT3 cells by adding CuSO 4 (Cu 2+ ) or not (NC). I) Proportion of WT and slx8Δ cells carrying pYEP-CUP1 or pYEP-CUP1-UB4-SMT3 at the indicated stages of synapsis at 10 h in sporulation medium. More than 200 nuclei were scored for each time point (n = 3 independent experiments). P < 0.0001 for full SC at 10 h (WT vs slx8Δ). Data are presented as mean ± SEM. Chi-square tests were used for statistical analyses. J,K) Slx8p mediated SUMO-ubiquitination crosstalk is essential for yeast meiosis. The proportion of J) divided nuclei and K) spore viability in WT, slx8Δ, WT Ub4-Smt3, and slx8Δ Ub4-Smt3 cells. More than 200 nuclei and spores were scored for each experiment (n = 3 independent experiments). Data are presented as mean ± SEM. *P < 0.05. **P < 0.01. Two-sided t-tests were used for statistical analyses. See also Figure S6 in the Supporting Information.

www.advancedscience.com
Slx8p can cooperate with Slx5p to work as a SUMO-targeted E3 ubiquitin ligase responsible for the turnover of poly-SUMOylated proteins. [25] Since SUMOylated proteins accumulated in slx8Δ cells during meiosis, we speculated that Slx8p might regulate meiosis through its SUMO-targeted ubiquitin ligase activity, which includes E3 ligase activity and the recognition of SUMO. To test this possibility, we first generated two vectors: SLX8 and SLX8 with a RING domain mutant (SLX8 rm ) under the control of its own promoter ( Figure 4B). Each of the vectors was transformed into an SLX8 deletion strain. The expression of SLX8 in slx8Δ cells successfully rescued the sporulation defect ( Figure 4C), whereas the expression of SLX8 rm could not ( Figure 4C). Moreover, we found that the spore viability of SLX8 rm transformed cells was similar to that of the slx8Δ cells ( Figure 4D), indicating that Slx8p participation in yeast meiosis is dependent on its E3 activity. Next, we determined whether Slx8p modulated meiosis through its SUMOinteracting motif (SIM) (Figure 4B,E). As Slx5p also contributes to the SUMO binding activity of the Slx5p-Slx8p complex, [25] we transformed a SIM domain-deleted mutant of SLX8 into slx5Δ slx8Δ cells. We found that once the SUMO binding activity of Slx5p-Slx8p was abolished, sporulation efficiency and spore viability could not be rescued ( Figure 4E,F), which was a similar outcome with the RING domain mutant in Slx8p. Thus, both the E3 ligase activity and SUMO recognition of Slx8p are required for its functional roles during yeast meiosis, and Slx8p participation in meiosis is dependent on its SUMO-targeted ubiquitin ligase activity.

Slx8p Facilitates SC Components Degradation via SUMO-Ubiquitination Crosstalk
As many SC components could be modified by SUMO, [3,7,13,37,38] including Red1p, Zip1p, and Ecm11p, and our results indicated Slx8p was required for SC degradation, we speculated that Slx8p might recognize the SUMOylated SC components and promote their ubiquitination to facilitate SC degradation. To test this possibility, we first examined the localization of Slx8p in chromatin spreading during meiotic prophase I. At the leptotene stage, Slx8p rarely loaded onto the chromosomal axis ( Figure 4G). When homologous chromosomes were synapsed at the prophase stage, Slx8p was recruited to the SC and formed linear stretches or discontinuous signals that were co-localized with Zip1p ( Figure 4G). Next, we tried to discern whether this localization of Slx8p was dependent on its SIM that recognized SUMOylated proteins on the SC. As Slx5p and Slx8p may participate in meiosis by working as a heterodimeric complex, we transfected the SLX8 simΔ into slx5Δ slx8Δ cells to exclude the influence of the SIM domain on Slx5p and endogenic Slx8p. We found the Slx8p SIM mutant was not recruited to the SC ( Figure 4G). Thus, Slx8p might be recruited to the SC by recognizing the SUMOylated proteins via its SIM. The major function of Slx8p is to assemble polyubiquitin chains on SUMO or SUMOylated proteins, and Slx8p is also responsible for the turnover of SUMOylated proteins. [25,39] To evaluate this function, we directly fused a linear tetraubiquitin chain onto SUMO (Ub4-Smt3) and transformed it in slx8Δ cells to test whether Slx8p mediated SUMO-ubiquitination crosstalk is required for meiosis. We found the expression of this artificial Ub4-SUMO hybrid partially rescued the sporulation defect in slx8Δ cells ( Figure 4I-K), demonstrating that Slx8p mediated SUMO-ubiquitination crosstalk was required for yeast meiosis. Therefore, Slx8p might recognize SUMOylated SC components and promote their ubiquitination to facilitate chromosome segregation.

Ecm11p Is the Key Substrate of Slx8p during Meiosis
As Slx5p-Slx8p is responsible for the turnover of SUMOylated proteins by promoting their ubiquitination, [25,39] and the degradation of Zip1p and Ecm11p was delayed in slx8Δ cells ( Figure 3C-F), we speculated that Slx8p might promote the degradation of Zip1p and Ecm11p to promote meiotic progression. To test this possibility, we directly depleted Zip1p and Ecm11p proteins in slx8Δ cells via an auxin-induced degradation (AID) system. [41,42] The AID system relies on an auxin-induced interaction between the indole-3-acetic acid (IAA) family transcription factors and the F-box protein TIR1, which results in the ubiquitination of AID-fused proteins and their subsequent degradation by the proteasome. [42] An AID-tag was added (knockedin) to the C-terminus of ECM11 and ZIP1 in both alleles of the WT and slx8Δ strains. The AID-tag did not affect the meiotic www.advancedscience.com progression of WT strains (Figure 6E,F), suggesting that the tag did not influence the function of these proteins. Once IAA was added to the medium, both Ecm11p and Zip1p were degraded in slx8Δ cells ( Figure 6A-D). After establishing this system, we next monitored the effect of induced degradation of Ecm11p and Zip1p on sporulation, finding that only the degradation of Ecm11p could partially rescue the sporulation defect of slx8Δ cells ( Figure 6E,F). In addition, the chromosome missegregation defect and the spore viability of slx8Δ cells were all rescued to some extent ( Figure 6G,H). These results suggest that Adv. Sci. 2020, 7,1900739  Either D) 4 × SUMO-Zip1 or E) 4 × SUMO-Ecm11 were added to the Slx5p-Slx8p ubiquitination system in vitro, which contained E1, E2, Slx5p-Slx8p, ubiquitin, and ATP. The relative amounts of ubiquitinated 4 × SUMO-Zip1, 4 × SUMO-Ecm11 and free 4 × SUMO-Zip1, 4 × SUMO-Ecm11 were quantified using Odyssey software (n = 3 independent experiments). Data are presented as mean ± SEM. See also Figure S7 in the Supporting Information.

www.advancedscience.com
Ecm11p is the key substrate of Slx8p, and Slx8p might add ubiquitin to SUMOylated Ecm11p to promote its degradation, thus facilitating meiotic progression.

Discussion
STUbLs have been reported to control the turnover of SUMOylated proteins via ubiquitin-dependent protein degradation to ensure appropriate cellular levels of SUMOylated proteins. [43] During mitosis, SUMOylated proteins are accumulated due to STUbLs deletion, which results in slight growth defects and decreased genotoxin resistance. These defects could be suppressed by the deletion of a SUMO ligase SIZ1. [36] However, during meiosis, deletions of SIZ1, SIZ2, and ZIP3 could not rescue the sporulation defect in slx8Δ cells ( Figure S6, Supporting Information), indicating that Slx8p uses a different mechanism to regulate meiosis.
The SC is a meiosis-specific proteinaceous structure that stabilizes the interactions between homologous chromosomes during meiotic prophase I and allows for interhomolog crossover formation. [3,6,7] Although the mechanism of SC assembly has been well established and is regulated by a combination of nonstructural proteins and PTMs, such as SUMOylation, phosphorylation, and N-terminal acetylation, [3,7] the SC disassembly mechanism and the roles of PTMs remain poorly understood. The transition from prophase I to metaphase I, following SC disassembly, is a crucial event for equal distribution of genetic materials and genome stability. When cells exit from pachytene, the SCs could be disassembled on the chromosome arm region, [31] while some may remain at centromeres to produce a functional tether Adv. Sci. 2020, 7, 1900739   Figure 6. Ecm11p is the key substrate of Slx8p for accurate chromosome segregation. A-D) Conditional protein degradation of A) Zip1p and C) Ecm11p using the auxin-induced degradation (AID) system in slx8Δ cells. Expression of A) Zip1p-AID-9 MYC and C) Ecm11p-AID-9 MYC were analyzed by immunoblotting using an anti-MYC antibody. Pgk1p served as a loading control. Ratios of B) Zip1p/Pgk1p and D) Ecm11p/Pgk1p levels were normalized to the maximum Zip1/ Pgk1p and Ecm11p/Pgk1p ratios (n = 3 independent experiments). Data are presented as mean ± SEM. E,F) The proportion of divided nuclei after induced Zip1p or Ecm11p degradation. E) WT ZIP1-AID-9 × MYC, slx8Δ ZIP1-AID-9 × MYC cells and F) WT ECM11-AID-9 × MYC, slx8Δ ECM11-AID-9 × MYC cells were analyzed for sporulation and released from the pachytene arrest by the addition of 1 × 10 −6 m β-estradiol at 6 h. Auxin in DMSO or DMSO alone was added at the same time. More than 150 nuclei were scored for each experiment (n = 3 independent experiments). Data are presented as mean ± SEM. **P < 0.01. Two-sided t-tests were used for statistical analyses. G) The proportion of mis-separated chromosomes after induced Ecm11p degradation. More than 150 cells were scored for each experiment (n = 3 independent experiments). Data are presented as mean ± SEM. *P < 0.05. Two-sided t-tests were used for statistical analyses. H) The proportion of spore viability after induced Ecm11p degradation. More than 100 spores were scored for each experiment (n = 3 independent experiments). Data are presented as mean ± SEM. *P < 0.05. Two-sided t-tests were used for statistical analyses.

www.advancedscience.com
for homologous chromosomes. [22] It has been proposed that the degradation of SC components is essential for proper segregation of homologs during MI. [3,7] In the current study, we found that the disruption of SLX8 mainly leads to chromosome missegregation ( Figure 3). We also found that delayed SC component degradation is responsible for the meiotic defects in slx8Δ cells because forced degradation of the SC component partially rescued the chromosome segregation defect of the slx8Δ cells ( Figure 6). It has been reported that both ubiquitin and proteasomes could directly load onto the chromosome axes and co-localize with the SC during the late pachytene stage. [23,24] Consistent with these observation, we found that Slx8p-Slx5p, a heterodimeric STUbL, recognized the SUMOylated SC components via their SIM domains for recruitment onto the SC; Slx8p-Slx5p then directly added ubiquitin to SUMOylated Ecm11p and Zip1p to promote their degradation (Figures 4-6). The SUMO-ubiquitination crosstalk is an economical way of displacing SC components from the corresponding chromosome regions, as it is unnecessary to resolve the SUMOylation before their degradation. Although SC disassembly in chromosome regions could not fully address the sporulation defect of the SLX8 deletion, since disrupted SC formation in spo11Δ slx8Δ cells could not rescue its sporulation defect ( Figure S4A, Supporting Information), the dotty Zip1p signal in spo11Δ slx8Δ cells was dramatically increased compared to that in spo11Δ cells ( Figure S4D, Supporting Information). Moreover, the proportion of Zip1p PCs were significantly higher in slx8Δ cells than in the control groups ( Figure S3, Supporting Information). This result suggests Slx5p-Slx8p might also promote SC components degradation at the centromeres or even in some regions not associated with chromatin, thereby facilitating the homologous chromosome separation. Our findings demonstrate that persisting Zip1p is associated with chromosome missegregation, and Slx8p-dependent degradation of SC component Ecm11p partially rescues the chromosome segregation defect. Slx5p-Slx8p may coordinate ubiquitination and SUMOylation of SC components to promote their efficient degradation, thus facilitating proper chromosomal segregation (Figures 2,3, and 7; Figures S3-S5, Supporting Information).
Although far less is known about the mechanism of SC disassembly than SC assembly, some disassembly studies have provided enough details to advance our understanding of this process. In yeast, Cdc5p and Ipl1p have been shown to be essential for pachytene exit and SC destruction. [36,44] Ipl1p could Adv. Sci. 2020, 7,1900739  www.advancedscience.com coordinate synaptonemal complex disassembly with resolution of crossovers and cell cycle progression at the end of meiotic prophase. Cdc5p, the only Polo-like kinase in budding yeast, is regulated by Ndt80p and controls pachytene exit. [45] The expression of Cdc5p triggers an untimely disassembly of the SC and irregular resolution of recombination intermediates. [30] Cdc5p could collaborate with CDK1 to hyperphosphorylate Dbf4p, the regulatory subunit of Dbf4p-dependent Cdc7p kinase (DDK), and hyperphosphorylated DDK greatly accelerates SC destruction. [46] However, the substrates of Cdc5p, Dbf4p, and Ipl1p, as well as the precise mechanism underlying their triggering of SC disassembly, remain unknown. We noticed that the protein level of Ndt80p was increased in slx8Δ cells ( Figure 2E), suggesting that Cdc5p should be continuously induced by Ndt80p and promote SC disassembly in SLX8 deleted cells. Contrary to this expectation, degradation of SC components was delayed rather than accelerated in slx8Δ cells ( Figure 3C-F). Thus, Slx8p-dependent SC degradation appears to occur downstream of the Cdc5ppromoted process. Further studies are needed to explore the detailed relationships between Slx5p-Slx8p and these reported SC disassembly factors.
The artificial Ub4-SUMO hybrid and the induced degradation of Ecm11p could only partially rescue the sporulation defect in slx8Δ cells, indicating Slx8p may play multiple roles during meiosis. We found that premeiotic DNA replication showed a minor delay and the expression of some SC components and meiotic cohesin was impaired to some extent in slx8Δ cells (Figures 1G,J,K and 3A-F). In addition, Slx5p-Slx8p has been reported to modulate the crossover interference during meiosis. [47] Therefore, future efforts are still needed to address the functional roles of Slx5p-Slx8p in other aspects of meiosis.
Strains and Plasmids: All yeast strains and plasmids used in this study are described in Tables S1 and S2 in the Supporting Information.
Sporulation Conditions and Meiotic Nuclear Division Assays: Sporulation was induced using potassium acetate as previously described. [48] The strains were grown for 24 h in YPD medium (1% yeast extract, 2% peptone, and 2% glucose), diluted in liquid YPA medium (1% yeast extract, 2% peptone, and 2% potassium acetate) to OD600 = 0.3, and grown for 10 h at 30 °C. Cells were harvested, washed, and resuspended in a sporulation medium (SPM, 2% potassium acetate) to OD600 = 1.9, and sporulated at 30 °C for different durations. Meiotic nuclear divisions were visualized by staining chromosomal DNA with 1 µg mL −1 4′,6′-diamidino-2-phenylindole (DAPI): the samples were harvested at the indicated times and directly fixed in an equal volume of 100% ethanol for subsequent DAPI staining. Images were recorded and analyzed using a Nikon Eclipse Ti microscope (Nikon, Eclipse Ti-S, Tokyo, Japan). Spore viability was determined by dissection of tetrads from SPM that had been incubated for 3-4 days at 30 °C.
Immunoblotting: Cells were treated with mild alkaline regents and boiled in a standard electrophoresis loading buffer, as previously described. [49] The samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions and transferred to nitrocellulose membranes. The blots were incubated with a primary antibody and a fluorescent dye-labeled secondary antibody (926-32211, LI-COR, Lincoln, NE); the blots were then scanned using an Odyssey infrared 740 imager (9120, LI-COR Biosciences, Lincoln, NE).
Flow Cytometry: To analyze the DNA content, 1 × 10 7 cells were fixed with 1 mL cold 70% ethanol overnight and then resuspended in 1 mL of 50 × 10 −3 m sodium citrate. The samples were centrifuged at 2000 rpm for 5 min, and the supernatant was removed. The pellets were digested with RNase A for 2 h at 37 °C and then sonicated for 2 s at 20% power, stained with 1 × 10 −6 m Sytox Green (Molecular Probes, S-7020), and analyzed using a BD FACSvantage SE Flow Cytometry System (Franklin Lakes, NJ).
Meiotic Surface Spread Nuclei: Meiotic chromosome spread, staining, and imaging were carried out as previously described [47] with the following modifications: 80 µL 1× MES and 200 µL 4% paraformaldehyde were added to spheroplasted, washed cells, and then cells were lysed and spread on a glass microscope slide with 1% Lipsol (LIP Ltd., Shipley England) and fixed by 3% w/v paraformaldehyde with 3.4% w/v sucrose as described. [47] The slide was allowed to air dry until less than half of the liquid remained, and then washed in 0.4% Photo-flo as described. [50] Primary antibodies were added to the sections and incubated at 4 °C overnight, followed by incubation with the secondary antibodies. The nuclei were stained with DAPI. The images were taken immediately using an LSM 780 microscope (Zeiss, Oberkochen, Germany) or a TCS SP8 microscope (Leica, Wetzlar, Germany). To identify meiotic nuclei at prophase I, dotty, dot-linear, and full linear Zip1-GFP staining were used. For subsequent stages, mCherry-tubulin and CNM67-GFP were used to assess spindle and SPB behavior.
Homologous Chromosome Segregation Assay: A homologous chromosome segregation assay was performed as previously described. [34] The tetO array was inserted into the HIS4 locus consisting of 120 DNA fragments; each fragment contained the 19-bp tetO binding site and a unique 10-bp spacer. The pTetR-kGFP plasmid was constructed by PCR using the pTetR-GFP plasmid, which expresses TetR-GFP fusion protein under the control of the URA3 promoter. The plasmid was cut with AflII and integrated into the LEU2 locus. Microscopy was performed on an inverted microscope (Nikon, Tokyo, Japan) with a 100× oil immersion objective.