help button home button Genetics AJP: Lung
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by White, E. J.
Right arrow Articles by Kaback, D. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by White, E. J.
Right arrow Articles by Kaback, D. B.
Genetics, Vol. 167, 51-63, May 2004, Copyright © 2004

In Vivo Analysis of Synaptonemal Complex Formation During Yeast Meiosis

Eric J. White1,a, Carrie Cowan2,b, W. Zacheus Candeb, and David B. Kabacka
a Department of Microbiology and Molecular Genetics, UMDNJ-Graduate School of Biomedical Sciences, UMDNJ-New Jersey Medical School, International Center for Public Health, Newark, New Jersey 07101-1709
b Department of Molecular and Cell Biology, University of California, Berkeley, California 94720

Corresponding author: David B. Kaback, UMDNJ-New Jersey Medical School, International Center for Public Health, 225 Warren St., P.O. Box 1709, Newark, NJ 07101-1709., kaback{at}umdnj.edu (E-mail)

Communicating editor: M. JOHNSTON


*  ABSTRACT
*TOP
 *ABSTRACT
*MATERIALS AND METHODS
 *RESULTS
 *DISCUSSION
 *LITERATURE CITED

During meiotic prophase a synaptonemal complex (SC) forms between each pair of homologous chromosomes and is believed to be involved in regulating recombination. Studies on SCs usually destroy nuclear architecture, making it impossible to examine the relationship of these structures to the rest of the nucleus. In Saccharomyces cerevisiae the meiosis-specific Zip1 protein is found throughout the entire length of each SC. To analyze the formation and structure of SCs in living cells, a functional ZIP1::GFP fusion was constructed and introduced into yeast. The ZIP1::GFP fusion produced fluorescent SCs and rescued the spore lethality phenotype of zip1 mutants. Optical sectioning and fluorescence deconvolution light microscopy revealed that, at zygotene, SC assembly was initiated at foci that appeared uniformly distributed throughout the nuclear volume. At early pachytene, the full-length SCs were more likely to be localized to the nuclear periphery while at later stages the SCs appeared to redistribute throughout the nuclear volume. These results suggest that SCs undergo dramatic rearrangements during meiotic prophase and that pachytene can be divided into two morphologically distinct substages: pachytene A, when SCs are perinuclear, and pachytene B, when SCs are uniformly distributed throughout the nucleus. ZIP1::GFP also facilitated the enrichment of fluorescent SC and the identification of meiosis-specific proteins by MALDI-TOF mass spectroscopy.

DURING meiosis, replicated homologous chromosomes (homologs) pair, undergo reciprocal recombination (crossing over or chiasma formation) and synapsis, and then segregate from each other in two distinct division stages, meiosis I and meiosis II, to reduce the number of chromosomes by half. Reciprocal recombination between each pair of homologs takes place during meiotic prophase and is essential for their proper segregation at the first meiotic division (meiosis I). Meiotic prophase can be further divided into leptotene, zygotene, pachytene, diplotene, and diakinesis stages defined by the state of chromosome pairing and condensation. At leptotene, chromosomes begin to condense and each replicated homolog forms an axial element. Near the end of this stage, telomeres cluster near the nuclear periphery; the chromosomes form a bouquet, and double-strand breaks (DSBs) in DNA that initiate recombination begin to appear. At zygotene, the axial elements synapse at the sites of the DSBs and tripartite proteinaceous structures called synaptonemal complexes (SCs) begin to form between homologous chromosomes (BAKER et al. 1976 Down). At pachytene, the SCs have matured into ribbon-like structures that intimately connect each pair of homologs from end to end and appear embedded at each end in the nuclear periphery (ESPONDA and GIMENEZ-MARTIN 1972 Down; BYERS and GOETSCH 1975 Down). In the yeast Saccharomyces cerevisiae, strand invasion and Holliday junction recombination intermediates are observed throughout zygotene and pachytene (ALLERS and LICHTEN 2001 Down; HUNTER and KLECKNER 2001 Down). At diplotene, the SCs come apart but homologs are joined by mature chiasmata at or near sites where reciprocal recombination has taken place. At this stage it is likely that Holliday junctions have been resolved and reciprocal recombination has been completed. At diakinesis chromosomes undergo further condensation prior to alignment on the spindle at metaphase.

The SC has been proposed to function in the regulation of meiotic recombination including a phenomenon called crossover interference (EGEL 1978 Down; SYM and ROEDER 1994 Down). Crossover interference is an observed suppression of double crossovers that suggests a specific mechanism for distributing crossovers. This mechanism is also thought to help insure that each pair of chromosomes undergoes crossing over (KABACK et al. 1999 Down). Very little is known about the molecular mechanisms of this regulation and it is hoped that studies on SCs will enable them to be better understood (EGEL 1978 Down).

Many proteins known to be involved in recombination have been shown to be associated with SCs (BISHOP 1994 Down; ROSS-MACDONALD and ROEDER 1994 Down; EIJPE et al. 2000 Down). In S. cerevisiae most of the known SC-associated proteins display a discontinuous distribution along the longitudinal axis of this structure and appear to be associated with recombination intermediates. In addition, several proteins believed to have a structural role in the SC have also been found (HEYTING et al. 1985 Down; HOLLINGSWORTH et al. 1990 Down; SYM et al. 1993 Down; SMITH and ROEDER 1997 Down). One of these, Zip1, appears to be an integral component of the SC and is found throughout its entire length (SYM et al. 1993 Down). It is the only known S. cerevisiae SC component that behaves this way and is believed to be the ortholog of SCP1 from rat testis (MEUWISSEN et al. 1992 Down). Zip1 contains 875 amino acids that have been suggested to form two large central {alpha}-helical coiled-coil domains flanked by globular domains (SYM et al. 1993 Down). Sporulating yeast cells that are deleted for ZIP1 do not form SC (SYM and ROEDER 1994 Down) and show 20–50% of the wild-type (wt) level of reciprocal recombination and an absence of crossover interference. Most zip1 strains complete sporulation but show high levels of spore inviability (SYM et al. 1993 Down). However, there are also some zip1 strains that arrest in meiotic prophase and fail to complete sporulation (SYM et al. 1993 Down).

Many aspects of SC formation have yet to be elucidated since these structures cannot be easily observed in living cells. Cytological studies on the SC usually involve fixatives, nuclear swelling, and surface spreading techniques that destroy nuclear architecture. These techniques make it difficult to examine the distribution of the SCs within the nucleus or follow their assembly. Therefore, to analyze the formation and structure of SCs in living cells, a functional ZIP1::GFP fusion was constructed and introduced into yeast. This construct produced brilliant fluorescent SCs that allowed an investigation of SC formation and structure in relationship to the nucleus in both wt and meiotic mutant cells. The results indicated that SCs undergo dynamic rearrangements and suggest that pachytene may be subdivided into two distinct morphological substages. The ZIP1::GFP fusion was also used as a convenient and easy-to-follow marker for the partial purification of SCs from pachytene-arrested cells.


*  MATERIALS AND METHODS
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
*RESULTS
 *DISCUSSION
 *LITERATURE CITED

DNA manipulations:
Standard techniques were used for recombinant DNA plasmid construction and amplification in Escherichia coli (MANIATIS et al. 1982 Down).

C-terminal fusions of GFP to ZIP1 were nonfunctional (not shown). Therefore, GFP was inserted in the ZIP1 coiled-coil domain that appears to tolerate small deletions without significant loss of function (SYM and ROEDER 1995 Down). A 2.4-kb promoter-containing fragment containing the N-terminal portion of the ZIP1 gene was amplified by PCR (MULLIS and FALOONA 1987 Down) from pR1630 (a gift from G. S. Roeder) using the forward primer A, 5'-GCTAGGTACCTATACAACCGATCGACAAATTAT-3' and the reverse primer B, 5'-AAGCATGGCGCGCCAAGAGAGCTAATAATCTG-3'. A KpnI site was added to the forward primer (underlined) that facilitated introduction into the pRS series of plasmid vectors (SIKORSKI and HIETER 1989 Down). An AscI site introduced into the reverse primer (underlined) facilitated in-frame fusion to the coiled-coil region of ZIP1 (DONG and ROEDER 2000 Down). A 0.7-kb fragment containing GFP was amplified from pPS904 (KAHANA et al. 1995 Down) using the primers 5'-AAGCATGGCGCGCCTGGAGGTATGGCTAGCAAAGGAGAA-3' and 5'-AAGCATGGCCGGCCGCAGCCGGATCCTTTGTAT-3'. AscI and FseI sites (underlined) were introduced at the 5' and 3' ends of GFP, respectively, to facilitate its integration into ZIP1. Primer C, 5'-AGATGAGGCCGGCCAGCACAATATGAAGATTTGGTC-3' and primer D, 5'-ATGCTAGCGGCCGCGACCTCTTTTGTTTTTACTAGAG-3' were used to amplify a 1.2-kb fragment from pR1630 encoding the C terminus and 3' untranslated region (UTR) of ZIP1. An FseI site was introduced at the GFP fusion site and a NotI site was introduced in the 3'-UTR to facilitate ligation into pRS plasmids. PCR products were cut with restriction enzymes corresponding to their introduced restriction sites and ligated between the KpnI and NotI sites in CEN/ARS plasmid pRS316 (SIKORSKI and HIETER 1989 Down) to produce pEJW2. The resultant Zip1::GFP fusion protein was deleted for amino acids 526–528 (TKE) of the wt protein and contained 251 additional amino acids that included the green fluorescent protein flanked by peptide linkers. The amino acid sequence of the N-terminal fusion site of GFP (italics), including linker amino acids (underlined) to Zip1 (double underline) at amino acid 525, is IISSLGGAPGGMASKGE. The amino acid sequence of the C-terminal fusion site of GFP (italics), including linker amino acids (underlined) to Zip1 (double underline) at amino acid 529, is YKGSGCGRPAQYEDL. All PCR incubations were carried out at 94° for 10 min; followed by 34 cycles of 94° for 30 sec, 55° for 30 sec, and 72° for 1–2 min; and then followed by 72° for 7 min.

For genomic integration a 4.3-kb KpnI-NotI fragment containing ZIP1::GFP from pEJW2 was ligated into KpnI- and NotI-digested integration vector pRS306 (SIKORSKI and HIETER 1989 Down) yielding plasmid pEJW1.

Yeast strains and yeast cell growth:
Strains and genotypes are listed in Table 1. All strains were derived from SK1 (KANE and ROTH 1974 Down) and were maintained, grown, and sporulated on standard media as previously described (SHERMAN et al. 1986 Down). Nancy Kleckner generously provided all NKY strains. Asci were dissected and analyzed as previously described (SHERMAN et al. 1986 Down). All growth and sporulation in liquid medium were carried out with vigorous rotatory shaking at 30°.


 
View this table:
[in this window]
[in a new window]

  		Table 1. Yeast strains

Recombinant DNA molecules were introduced into yeast using 0.3 M lithium acetate as previously described (ITO et al. 1983 Down). ZIP1::GFP was introduced either on a centromere- containing plasmid or by two-step gene replacement (BURKE et al. 2000 Down) into haploid MATa and MAT{alpha} isogenic mating pairs. Mating produced diploids with noted genotypes. The wt and heterozygous control strains were made using NKY278a and NKY278b. ndj1::URA3 was introduced into EW201 and EW202 by one-step gene replacement as previously described (CONRAD et al. 1997 Down). ZIP1::GFP was introduced into ndt80::URA3 strains NKY2292 and NKY2293 by two-step gene replacement. Correct integration of all constructs was confirmed by DNA blot hybridization (SOUTHERN 1975 Down).

Synaptonemal complex spreads and conventional fluorescence microscopy:
Sporulating cells were spheroplasted and spreads were produced using the method of LOIDL et al. 1994 Down. Unstained spreads were visualized directly for fluorescent SCs. Fluorescence microscopy was performed at x1000 magnification using either an FITC filter for green fluorescent protein (GFP) or a 4',6-diamidino-2-phenylindole (DAPI; Hoechst) filter for stained DNA. Electron microscopy at x3000 magnification was carried out on surface-spread material that was stained with AgNO3 as previously described (LOIDL et al. 1994 Down).

Functional activity of ZIP1::GFP:
The percentage of viable ascospores from four-spored asci was determined by dissection and spore growth on YPD agar plates. Percentage of functional activity of ZIP1::GFP is the observed percentage of viable spores normalized between 0 and 100 where 0 was the percentage of viability obtained for the zip1::LYS2 mutant and 100 was the percentage of viability obtained for the wt control strain. Percentage of functional activity = 100 x (ZIP1::GFP spore viability – zip1 spore viability)/(wt spore viability – zip1 spore viability).

Synchronous sporulation for analyzing kinetics of SC formation and completion of meiotic divisions:
Cells were sporulated essentially as previously described (PADMORE et al. 1991 Down) using a single fresh colony from a YPD plate that was first streaked onto a YEP 2% (v/v) glycerol plate and grown for 3–4 days. A large colony that arose was then inoculated into 1.0 ml liquid YPD medium and grown to saturation overnight. A 1:100 dilution was made into 100 ml YEP 1% acetate and grown for 13.5 hr. Cells were harvested by centrifugation [1000 x g, 5 min at room temperature (RT)], washed with water, and resuspended in 500 ml 2% (w/v) potassium acetate sporulation medium, pH 7.0, at T = 0 hr. Cells were visualized using an Olympus BX60 fluorescence microscope. Meiotic progression was monitored at hourly intervals by counting the number of nuclei in ~200 DAPI-stained cells (LOIDL et al. 1994 Down). Kinetics of SC formation was monitored at hourly intervals by assigning ~200 sporulating nuclei into one of four categories, no fluorescence, diffuse fluorescence, punctate fluorescence, and full-length fluorescent strings, and plotting the percentage of each category as a function of time. Each kinetic analysis was carried out using two independently transformed strains. In all cases no differences were observed between the two transformants.

Fluorescence deconvolution light microscopy:
Microscopy was performed on a wide-field Olympus IX70 inverted microscope using a x100 1.35NA PlanApo lens, as part of the DeltaVision microscope system (Applied Precision, Issaquah, WA). Data sets were collected for GFP and Hoechst in three dimensions. Z-axis images were collected 0.2 µm apart. Three-dimensional data sets were deconvolved using the DeltaVision software package. To rotate three-dimensional nuclei, the VolumeViewer program in DeltaVision was used to create 36 projections around 180°.

Enrichment of fluorescent SCs:
Meiotic nuclei were prepared by an adaptation of the method of ROUT and KILMARTIN 1998 Down. A fresh YPD-grown overnight culture of strain EW104 was inoculated into 8 liters of YPA medium containing 2% (w/v) potassium acetate (10 ml inoculum/1 liter medium/2-liter flask) and grown to a density of 80–100 Klett units. Cells were harvested (1500 x g, 5 min at RT), washed with water, resuspended in 4 liters 2.0% (w/v) potassium acetate, pH 7.0 (500 ml/2-liter flask), and incubated for 10 hr with vigorous shaking at which time ~90% of cells were arrested at pachytene with full-length green fluorescent SC. The arrested cells were harvested (1500 x g, 5 min at RT); washed with water; resuspended in four 50-ml plastic tubes, each containing 40 ml of 100 mM Tris-HCl and 10 mM dithiothreitol (DTT), pH 9.4 (~10 g of cells wet weight/tube); and incubated for 10 min at RT. The cells were then pelleted (1500 x g, 5–10 min), washed with distilled water and then with 1.1 M sorbitol at RT, and resuspended in 8 ml 1.1 M sorbitol/tube. A total of 20 mg Zymolyase 20T (Seikagaku America, Ijamsville, MD), 4 mg lytic enzymes (Sigma, St. Louis), and 800 µl Glusulase (New England Nuclear, Boston) were added per tube and the suspension was incubated for ~2 hr at 30°. Spheroplasting was monitored by phase-contrast microscopy by adding ~1 µl 10% (w/v) sodium dodecyl sulfate to ~50 µl of sample and observing the extent of lysis. Spheroplasting was considered complete when the detergent lysed ~95% of the cells. Ice-cold 1.1 M sorbitol was then added to fill each centrifuge tube and the spheroplasted cells were pelleted (1500 x g for 5 min at 4°), gently resuspended, and washed in ice-cold 1.1 M sorbitol. The spheroplasts were gently resuspended in 20 ml ice-cold 1.1 M sorbitol/tube and each tube was overlaid onto 12 ml 7.5% Ficoll type 400 in 1.1 M sorbitol in a 40-ml centrifuge tube. Cells were centrifuged through the Ficoll layer at 10,500 x g for 15 min at 4° and the supernatant was carefully evacuated from the top to avoid any loss of the spheroplast-containing pellet. Each tube of spheroplasts was resuspended in 20 ml PVP lysis solution. PVP lysis solution is made up from PVP buffer plus the additions listed below. Each liter of PVP buffer contains 80 g polyvinyl pyrolidone (PVP-40), 1.57 g KH2PO4, 1.46 g K2HPO4, and 1.0 mmol MgCl2. To make PVP lysis solution, 400 µl 1.0 M DTT, 800 µl 2.0% (w/v in 100% ethanol) phenylmethylsulfonyl fluoride (PMSF; Sigma), 200 µl 10% (w/v) Triton X-100, 16 µl 1.0% (in DMSO) pepstatin (Sigma), 16 µl 1.0% (w/v in DMSO) chymostatin (Sigma), and 16 µl 1.0% (w/v) leupeptin (Sigma) were freshly added to 80 ml of PVP buffer just prior to use. While preparing lysis solution, it was also necessary to freshly prepare 50 ml of PVP-sucrose buffer. To prepare this solution 50 µl 1.0 M DTT, 100 µl 2.0% (w/v) PMSF, 10 µl 1.0% pepstatin, 10 µl 1.0% chymostatin, and 10 µl 1.0% leupeptin were added to 50 ml PVP buffer containing 0.3 M sucrose. Spheroplast lysis was performed by homogenization with a Polytron PT3000 (Brinkman/Kinematica AG, Lucerne, Switzerland) for 1 min at 12,000 rpm, 1 min at 15,000 rpm, and 1 min at 17,000 rpm with tubes placed on ice for 3 min in between each Polytron homogenization step. Immediately following homogenization, the extracts were underlaid with 12 ml of PVP-sucrose solution per tube and centrifuged at 16,500 x g for 20 min at 4° in an HB4 (Sorvall) rotor. The crude nuclear pellet that resulted was resuspended in 7.0 ml Bis-Tris buffer; 50 ml of Bis-Tris buffer contained 10 mM Bis-Tris, pH 6.75, 50 µl 1.0 M DTT, 100 µl 2.0% PMSF, 10 µl 1.0% pepstatin, 10 µl 1.0% chymostatin, 10 µl 1.0% leupeptin, and 10 µl 5 M MgCl2. Resuspension of the crude nuclear pellet was carried out using the Polytron at 12,000 rpm for two 30-sec homogenizations with tubes placed on ice between each step. Bis-Tris buffer (5 ml) containing 2.5 M sucrose was added to each tube and mixed by vigorous shaking to bring the final sucrose concentration to ~1.1 M. The homogenates were overlaid onto four step gradients containing 10 ml 2.5 M sucrose and 14 ml 1.2 M sucrose in Bis-Tris buffer made in 36-ml polypropylene centrifuge tubes. The gradients were centrifuged at 20,000 rpm for 1 hr at 4° in an SW28 rotor (55,000 x g; Beckman Instruments, Fullerton, CA) and fractionated by removing material from the top by aspiration. The top 24 ml that was just above the 1.2 M sucrose-2.5 M sucrose interface was removed in two 12-ml fractions and discarded. A majority of fluorescent nuclei were located just below at this interface. This layer (~8 ml/tube; fraction 3) was removed, pooled (total volume ~35 ml), and stored at –80°. The remaining material containing unlysed cells and spheroplasts was discarded. Fraction 3 was defrosted and ~15 ml Bis-Tris buffer was added to a total volume of ~50 ml that was divided among three 16-ml polypropylene ultracentrifuge tubes and centrifuged at 20,000 rpm for 1 hr at 4° in an SW28 rotor. The supernatant was aspirated and discarded. The pellet from each tube was resuspended in 1 ml cold Bis-Tris buffer containing 40% sucrose by vigorous pipetting. Twenty microliters of 2.0% (w/v) DNAseI (Sigma) and 100 µl 1.0% (w/v) RNAseA (Sigma) were added to each tube and the mixture was allowed to sit at RT for 15–20 min. Nine milliliters of Bis-Tris buffer containing 40% (w/v) sucrose plus 2.0% (v/v) polyoxyethylene 9 lauryl ether (polidocanol, Sigma) was added per tube and the tubes were placed on ice for 30 min. Only solid polidocanol was used as this detergent liquefies and loses its ability to efficiently lyse nuclei with age. Tubes were then incubated for 10 min at RT and then for an additional 10 min at 37°; 10.0% (w/v) N-lauroyl sarcosine sodium salt (Sigma) was added to a final concentration of 0.2% (w/v) and lysis was continued for an additional 30 min at 4°. Following lysis, the suspension from each tube was divided into 5-ml polypropylene tubes and centrifuged for 6 min at 6000 rpm (3300 x g) at 4° in an SW50 rotor to remove large nonresuspendible particles. The supernatant (5 ml/tube) was removed by aspiration and the volume was increased to 10 ml with a sucrose concentration of 1.5 M by the addition of 1.3 ml Bis-Tris buffer and 3.7 ml 2.5 M sucrose in Bis-Tris buffer. The lysed nuclear extract was divided and overlaid onto six step gradients consisting of 5.0 ml 2.3 M, 6.0 ml 2.0 M, and 15.0 ml 1.7 M sucrose in Bis-Tris buffer made up in 36-ml polypropylene tubes and centrifuged for 10 hr at 23,000 rpm (70,000 x g) at 4° in an SW28 rotor. The gradient was fractionated from the top by aspiration and the fractions comprising the 2.0 M sucrose layer containing the bulk of the fluorescent material were pooled and utilized for further study.

Quantitation of SC enrichment:
Equal amounts (200 µl) of each Zip1::GFP fraction and parallel wt Zip1 fractions were analyzed in nonfluorescent 96-well plates (Nalgenunc International, Rochester, NY) using a microplate fluorescence reader (FL500 version 1D1, Bio-tek, Winooski, VT) set at a sensitivity of 50 using an FITC filter set with excitation at 485 ± 20 nm and emission at 530 ± 25 nm. GFP-specific fluorescence was the difference between the Zip1::GFP and parallel wt measurements.

To determine the concentration of proteins in each sample the sucrose fractions were dialyzed overnight against 4 liters of sterile distilled H2O in the presence of 0.025% SDS at RT with continuous stirring using 6–8000 Mr cutoff Spectrapore dialysis tubing that had been boiled for 30 min and cooled (Spectrum Medical Industries, Los Angeles). To remove the polidocanol, 200 mM cholic acid (Sigma) and 5 mM N-ethylmalemide (Sigma) were added to each fraction. Following dialysis, fractions were lyophilized and resuspended in 500–1000 µl sterile H2O. Protein concentration from 100 µl of each sample was determined with the Bio-Rad Dc protein assay (Hercules, CA), which enables the accurate quantitation of protein in the presence of contaminating detergent.

PAGE and Western blotting:
Proteins were separated on either 8 or 10% discontinuous polyacrylamide gels containing SDS as described by LAEMMLI 1970 Down and visualized by staining with either Coomassie blue or silver nitrate (MORRISSEY 1981 Down; WILSON 1983 Down). Large-scale protein electrophoresis was performed with an SE 400 vertical slab gel (Amersham Biosciences, Piscataway, NJ).

For Western blotting, ~5 µg of protein was added per lane and separated by PAGE using a Bio-Rad Mini-PROTEAN III system and transferred to polyvinyl difluoride (PVDF) membrane (Amersham Biosciences) with the Bio-Rad Mini Trans-blot cell in the presence of Towbin buffer containing SDS as described by the supplier (TOWBIN et al. 1979 Down). GFP was detected using GFP Living Colors antibody from CLONTECH (Palo Alto, CA). Red1 antiserum was a gift from G. S. Roeder. The ECL+Plus system (Amersham Biosciences) was used for detection of antibody complexes.

Protein identification by MALDI-TOF mass spectroscopy:
Protein bands from Coomassie brilliant blue G-250-stained preparative gels were excised with a razor blade, minced, placed in siliconized 1.5-ml conical centrifuge tubes, and washed for at least 1 hr with 500 µl 0.1 M ammonium bicarbonate at RT. The gel pieces were rewashed twice for >1 hr in 500 µl of 0.1 M ammonium bicarbonate in 50% acetonitrile (ACN). ACN (100 µl) was added to shrink the gel pieces and removed. Gel pieces were dried and proteins were trypsinized as previously described (CHRISTOFFERS et al. 2003 Down). Salts and any remaining gel debris were removed from the peptides using a C18 ZipTip (Millipore, Bedford, MA) as described by the manufacturer. The peptides were then spotted onto plates and analyzed by MALDI-TOF mass spectroscopy (Voyager MALDI DE-PRO, Applied Biosystems, Foster City, CA) as described (CHRISTOFFERS et al. 2003 Down).


*  RESULTS
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
 *RESULTS
*DISCUSSION
 *LITERATURE CITED

Characterization of meiosis and sporulation in ZIP1::GFP cells:
To investigate the formation and structure of meiotic SCs in living S. cerevisiae cells, a ZIP1::GFP fusion was constructed as described in MATERIALS AND METHODS. To avoid interference with the nuclear localization signal (NLS) at the carboxyl terminal (BURNS et al. 1994 Down), GFP was inserted in the middle of the coiled-coil domain (Fig 1A). This region had been previously shown to be able to withstand small deletions while maintaining a significant level of function (TUNG and ROEDER 1998 Down). The ZIP1::GFP fusion was introduced on a stable centromere-containing plasmid (pEJW2) into a homozygous null zip1 mutant diploid (strain EW407). Examination of spreads from transformed sporulating cells (strain EW408) sampled at the appropriate time revealed the presence of nuclei with ~16 full-length brilliantly fluorescent SCs (Fig 1B). Most of these nuclei also contained an amorphous fluorescent body that probably was the polycomplex but may have also been a GFP aggregate. The polycomplex is an aggregate containing Zip1 protein and is frequently observed in wt meiotic cells (MOENS and RAPPORT 1971 Down; SYM and ROEDER 1995 Down). Electron microscopy of these fluorescent spreads revealed that the SCs appeared normal with a tripartite structure surrounded by chromatin (Fig 1C; SYM et al. 1993 Down). Therefore, Zip1::GFP is incorporated efficiently into SCs. Strains deleted for zip1 display substantially decreased spore viability compared to wt (SYM et al. 1993 Down). Strain EW408 (zip1::LYS2/zip1::LYS2) that contains plasmid pEJW2 (ZIP1::GFP) showed significantly enhanced spore viability (72 ± 7%, n = 216 spores dissected) compared to an isogenic zip1::LYS2/zip1::LYS2 control strain that spontaneously lost the plasmid (45 ± 6%, n = 240 spores dissected). However, the increased spore viability was an underestimate of the functional activity of ZIP1::GFP because plasmid pEJW2 instability lowered the value and could not be accurately quantitated. Therefore, ZIP1::GFP was stably integrated in the chromosome in place of the wt ZIP1 gene as described in MATERIALS AND METHODS and ZIP1::GFP/ZIP1::GFP homozygotes and ZIP1::GFP/ZIP1 heterozygotes were constructed. The homozygotes (strain EW102) produced brilliantly fluorescent SCs at the appropriate stage of meiosis while the ZIP1::GFP/ZIP1 heterozygotes (strain EW103) produced much less fluorescent SCs that were much more difficult to observe (not shown).



View larger version (41K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. ZIP1::GFP produces functional fluorescent synaptonemal complexes. (A) Structure of the Zip1::GFP fusion protein. Numbers denote amino acid residues of the wt protein. (B) Fluorescent SC spread obtained from 6-hr sporulating EW408 cells. Magnification, x1000. (C) Electron micrograph of surface-spread Zip1::GFP containing SCs from the same strain. Magnification, x3000; bar, 2.5 µm; inset, x2 enlargement of noted area showing tripartite SC structure. (D) Functional activity of ZIP1::GFP assayed by examining viability of dissected spores from 40–80 asci for each strain. Percentage of viable spores ±SE is shown. Percentage complementation shown is the percentage of viable spores normalized between 0 and 100% as described in MATERIALS AND METHODS.

To assay the functional activity of the fluorescent protein, its ability to increase spore viability was quantitated (Fig 1D). Spore viability of the ZIP1::GFP/ZIP1::GFP homozygote was only slightly less than that of the wt control while the ZIP1::GFP/ZIP1 heterozygote was indistinguishable from wt. Normalizing percentage of viability between 1.0 and 0 for wt and the homozygous zip1 null mutant, respectively, indicated that ZIP1::GFP in place of both copies of the wt gene restored viability to 85% of the wt value.

ZIP1 is required for normal levels of meiotic reciprocal recombination. To further assay the functional activity of the fluorescent protein, meiotic recombination was monitored in a 130-kb interval between an inserted URA3 gene on the left arm (CAO et al. 1990 Down) and the MAT locus on the right arm of chromosome III. The results [88PD:12NPD:90T = 43 cM for ZIP1::GFP (strain EW112) and 46P:14NPD:121T = 57 cM for wt (strain EW114)] indicated that the level of crossing over in the ZIP1::GFP homozygote was 75% that of wt, in close agreement with the restoration of viability.

Examination of meiotic nuclear division and sporulation at hourly intervals revealed that the ZIP1::GFP homozygote, ZIP1::GFP/ZIP1 heterozygote, and wt homozygote strains all sporulated to the same extent (Fig 2A). ZIP1::GFP/ZIP1 heterozygotes and wt homozygotes displayed no differences in the kinetics of meiosis and sporulation while the ZIP1::GFP homozygote exhibited an ~1-hr delay in the appearance of cells containing two nuclei (meiosis I). Thus, ZIP1::GFP produces easily observable fluorescent SCs and a near wt phenotype with respect to spore viability and meiotic reciprocal recombination. When heterozygous with a wt ZIP1 gene, there is no discernible phenotype, but SCs are less fluorescent and difficult to observe compared to the homozygote.



View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Kinetics of meiosis of ZIP1::GFP-containing strains. (A) Strains were sporulated and samples were taken at hourly intervals. Approximately 200 cells of each strain were scored for the presence of one, two, or four DAPI-stained nuclei by conventional fluorescence microscopy. Data for strains EW103 ZIP1/ZIP1::GFP ({blacksquare}, two nuclei; {square}, four nuclei), NKY278 ZIP1/ZIP1 ({blacktriangleup}, two nuclei; {triangleup}, four nuclei), and EW102 ZIP1::GFP/ZIP1::GFP (•, two nuclei; {circ}, four nuclei) are shown. Cells containing single nuclei are not shown. (B) Classes of nuclei from sporulating ZIP1::GFP cells. 1, nonfluorescent; 2, diffuse fluorescence; 3, punctate fluorescence; 4, fluorescent full-length SCs. (C) Time course of SC formation in ZIP1::GFP cells. Sporulating ZIP1::GFP (EW102) cells were sampled at hourly intervals and ~200 uniformly selected nuclei were examined. Each nucleus was classified as described in MATERIALS AND METHODS and the percentage of each class was plotted as a function of time. {blacksquare}, nonfluorescent; {square}, diffuse fluorescence (prezygotene); {blacktriangleup}, punctate fluorescence (zygotene); {circ}, full-length fluorescent SCs (pachytene).

SC formation was examined at hourly intervals by classifying nuclei into the four different morphological categories on the basis of the appearance and organization of the fluorescent protein (Fig 2B). Due to the difficulty of visually scoring the less fluorescent SCs in the heterozygote, a kinetic analysis of SC formation was performed solely on the ZIP1::GFP homozygote (Fig 2C). Zip1::GFP fluorescence was first observed at 2 hr and was diffusely distributed throughout the nuclei. By 3 hr many nuclei contained punctate fluorescent foci that we interpret to be due to the initiation of synapsis and the initial stages of SC formation at zygotene. At this point a few nuclei contained some full-length SCs. The number of nuclei containing full-length SCs increased at later times, reaching a maximum at 5 hr. From 7 to 8 hr there was another small increase in punctate nuclei that is likely to represent a very short-lived diplotene stage where SCs are disassembled. Concurrently, a striking increase in the number of non-fluorescent nuclei was observed, indicating that the fluorescent protein was degraded rapidly. By integrating the proportion of nuclei at each time point and correcting for sporulation efficiency, we calculated that zygotene lasted 0.4–0.6 hr, pachytene lasted 1.4–1.8 hr, and diplotene lasted 0.3–0.5 hr. These in vivo results are somewhat similar to previous studies using fixed wt ZIP1 cells (PADMORE et al. 1991 Down). However, zygotene was ~15 min longer and pachytene was ~30 min longer in the ZIP1::GFP homozygote. These results show that ZIP1::GFP strains produce the known stages in meiosis with a high degree of efficiency and synchrony.

Initiation of synapsis in zygotene nuclei:
Fluorescence deconvolution light microscopy was used to study zygotene synapsis and SC initiation. These studies were carried out only in the ZIP1::GFP/ZIP1::GFP homozygote because the observed fluorescence was too weak to score in the ZIP1::GFP/ZIP1 heterozygote. Sporulating cells were sampled at 2–3 hr when the percentage of presumed zygotene nuclei with punctate fluorescent dots was at a maximum. Optical sections of zygotene nuclei revealed that the GFP foci were uniformly distributed throughout the entire volume of each nucleus (Fig 3). Individual spots were not clearly resolved so the number of foci per nucleus could not be determined. These results suggest that zygotene pairing takes place all over the nucleus.



View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Uniform distribution of Zip1::GFP foci in zygotene nuclei. (A and B) Projections of 3-D reconstruction of optical sections of two zygotene nuclei from strain EW102 (ZIP1::GFP/ZIP1::GFP) showing punctate foci and short filaments indicative of zygotene synapsis. Zip1::GFP (green) is displayed above Hoechst (red)-stained DNA, which defines nuclear shape.

SC organization and distribution in pachytene nuclei:
At 5–7 hr, pachytene nuclei containing full-length SC were readily observable by fluorescence deconvolution microscopy in both the ZIP1::GFP/ZIP1::GFP homozygote and the ZIP1::GFP/ZIP1 heterozygote. In each nucleus all the SCs either were located at the nuclear periphery (Fig 4A and Fig B) or were more uniformly distributed throughout the entire nuclear volume (Fig 4C).



View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Perinuclear and uniform distribution of SCs. Orthographic projections of 3-D reconstructions of optical sections showing different pachytene SC distributions in EW102 (ZIP1::GFP/ZIP1::GFP) cells. (A) Symmetric distribution near the nuclear periphery. (B) Asymmetric or polarized distribution near the nuclear periphery. (C) Uniform distribution of SC within the entire nuclear volume. Images on the right are rotated 90° on the horizontal axis. (D) Hoechst-stained ZIP1::GFP/ZIP1::GFP nucleus showing diffuse perinuclear distribution of DNA. Inset shows perinuclear Zip1::GFP fluorescence from the same nucleus. (E) Hoechst-stained wt nucleus harvested at 3 hr showing diffuse perinuclear distribution of DNA. (F) Kinetics of SC distribution of wt, ndt80, and ndj1 cells containing ZIP1::GFP. Strains EW102 (wt), EW104 (ndt80), and EW105 (ndj1) were sporulated and cells were sampled at hourly time points. Three-dimensional reconstructions were made from ~50 pachytene nuclei per sample.

In most nuclei containing perinuclear SCs, the SCs were distributed in an asymmetric or polarized array (Fig 4B). In about one-fourth of the nuclei containing perinuclear SCs, the SCs were more symmetrically distributed around the entire periphery (Fig 4A). As there were no observable kinetic differences between the polarized and symmetrically distributed perinuclear arrays, the two arrays were treated as a single perinuclear class.

Quantitation of cells with perinuclear vs. uniformly distributed SCs between 4 and 7 hr in the heterozygote showed a decrease in the perinuclear population and an increase in the uniform population at the later time. However, the difference could not be considered significant. In contrast, when the homozygote was examined, the SCs were predominantly perinuclear at the earliest time point and predominantly uniformly distributed (90%) at the latest time points (Fig 4F). The studies on the homozygote suggest that pachytene consists of two distinct morphological stages, one where the SCs are predominantly perinuclear and a later stage where SCs are uniformly distributed throughout the nuclear volume. Alternatively, it is possible that pachytene is more dynamic with SCs oscillating between the perinuclear and uniform distributions, so that SCs spend more time at the periphery at early times than at later times.

Redistribution of SC in pachytene nuclei is dependent upon NDT80:
NDT80 mutants are blocked in pachytene and fail to properly resolve recombination intermediates into mature crossovers (ALLERS and LICHTEN 2001 Down). To examine whether this arrest corresponds to any of the proposed pachytene substages, ndt80 homozygotes containing ZIP1::GFP (strain EW104) were examined by fluorescence deconvolution microscopy. SCs did not appear to be exactly the same as wt but were narrower and seemed to be less condensed (Fig 5). Similar to the ZIP1::GFP homozygote, pachytene nuclei initially displayed predominantly perinuclear SCs. However, no time-dependent redistribution was observed. The percentage of cells with perinuclear SCs remained the same at all three times shown (Fig 4D) and did not appear to change after an additional 9 hr of incubation (data not shown). These results suggest that ndt80 mutants arrest with mostly perinuclear SCs. They are consistent with the idea that pachytene consists of two distinct morphological substages where the perinuclear association precedes the uniform distribution. They also are consistent with the idea that the release of SCs from the periphery is connected to the completion of crossing over.



View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. SC distribution and structure in ndt80 cells. Three-dimensional reconstructions of optical sections showing distribution of ZIP1::GFP-labeled SCs in strain EW104. (A) Symmetric distribution of SCs near the nuclear periphery. (B) Asymmetric or polarized distribution of SCs near the nuclear periphery.

Perinuclear chromosomes in meiotic nuclei lacking ZIP1::GFP:
Evidence for a perinuclear association could also be seen in appropriately staged wt cells where Hoechst-stained DNA gave a doughnut-like array similar to that observed in ZIP1::GFP-labeled cells with perinuclear SCs (Fig 4D and Fig E). Doughnut-like arrays were also readily observed in appropriately oriented DAPI-stained wt cells without the aid of optical sectioning and deconvolution. These arrays were indistinguishable from those shown in Fig 4D and Fig E. Quantitation by conventional fluorescence microscopy showed their appearance was induced by sporulation and was transient, appearing and disappearing at the appropriate time prior to the first meiotic division (Fig 6).



View larger version (18K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Meiosis-induced peripheralization of chromosomes in wt Zip1-containing nuclei. Cells incubated in sporulation medium were sampled at the times shown and the percentages of DAPI-stained nuclei containing either a doughnut-like pachytene DNA array or completed meiotic divisions (two nuclei/cell for meiosis I or four nuclei/cell for meiosis II) were determined using conventional fluorescence microscopy. Cells (~200) were scored for each time point and each cell was examined in several focal planes to maximize detection of doughnut-like arrays. Data for NKY278 wt ZIP1/ZIP1 cells ({blacksquare}, doughnut-like arrays; •, two nuclei/cell; {blacktriangleup}, four nuclei/cell), EW204 ndt80/ndt80 cells ({diamond}, doughnut-like arrays), and EW104 ndt80/ndt80 ZIP1::GFP/ZIP1::GFP cells ({diamondsuit}, doughnut-like arrays) are shown.

The ndt80 and ndt80 ZIP1::GFP homozygotes were also examined and exhibited sporulation-induced doughnut-like DNA arrays that were indistinguishable from those in wt cells. However, in these strains the doughnut-like arrays arose at the appropriate time and persisted and there was no nuclear division indicative of the pachytene arrest (Fig 6). These results provide further evidence for the perinuclear SC arrays and indicate that they are not an artifact of the ZIP1::GFP construct.

Perinuclear arrangement and redistribution of SC is not dependent upon NDJ1:
NDJ1 encodes a meiosis-specific telomere-associated protein required for the formation of leptotene-zygotene chromosomal bouquet. It has been suggested that NDJ1 plays a role in the interaction of the telomeres with the nuclear periphery. Therefore, it seemed reasonable to determine whether NDJ1 was required for the perinuclear SC array. Accordingly, SC formation and structure were examined in an ndj1 mutant containing ZIP1::GFP. Both perinuclear and uniformly distributed SCs were observed in the mutant (Fig 4F). In addition, there was a significant shift from perinuclear to uniformly arrayed SCs as pachytene progressed, similar to that observed in the ZIP1::GFP/ZIP1::GFP homozygote. The ndj1 mutants exhibited a longer pachytene as has been previously reported (CONRAD et al. 1997 Down). These results suggest that NDJ1 is not required for the establishment of the perinuclear SC array and any subsequent rearrangement. They are consistent with the idea that the arrangement of SC at the nuclear periphery does not directly involve telomeres.

Use of ZIP1::GFP for the enrichment of SCs:
ZIP1::GFP was next used as a fluorescent marker to assay the enrichment of SCs and attempt to identify meiosis-specific SC proteins. An ndt80 mutant homozygous for ZIP1::GFP (strain EW104) that was arrested in pachytene was utilized to increase the fraction of the cells containing SCs. In total, 90–95% of the nuclei showed bright fluorescent SC (Fig 7). Details for purification of nuclei and the enrichment of SCs are described in MATERIALS AND METHODS. Briefly, nuclei were isolated from spheroplasted cells by a modification of the method of ROUT and KILMARTIN 1998 Down. Purified nuclei showed a 6-fold increase in fluorescent units (FU)/mg protein (Table 2). Nuclei were lysed using 2.0% (w/v) polyoxyethylene 9 lauryl ether (polidocanol), which preserved fluorescence and enabled surface spreads of the SCs from enriched nuclei (Fig 8). Next, the lysed nuclei were treated with DNAseI and RNAseA to remove most of the chromosomal DNA and contaminating RNA and reduce the viscosity of the nuclear lysate. The extract was clarified by centrifugation, which removed insoluble material without significant loss of fluorescence. The supernatant was centrifuged onto a 1.5–2.3 M sucrose step gradient and the majority of fluorescent material was contained in the 2.0 M layer. This fraction was enriched 36-fold for Zip1::GFP and contained 60% of the total starting fluorescent material. Fluorescence microscopy revealed that this fraction contained fluorescent strings, punctate material, and some aggregates containing both strings and punctate material (Fig 9).



View larger version (89K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7. ZIP1::GFP-containing ndt80 cells arrested in pachytene. (A) Cells from strain EW104 were harvested after 8 hr in sporulation medium. Approximately 95% of cells were arrested in pachytene with fluorescent SCs. Distinct SCs cannot be seen in every cell because all nuclei were not in the same focal plane. (B) Phase-contrast microscopic image of the same cells. Magnification, x1000.



View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8. SC spreads from enriched nuclei. Strain EW104 (ZIP1::GFP/ZIP1::GFP, ndt80/ndt80) was sporulated and crude nuclei (nuclear fraction 3) were prepared as described in MATERIALS AND METHODS. Surface spreads were performed using 2.0% (w/v) polidocanol to lyse the nuclei. (A and B) Phase-contrast microscopy of silver-stained preparations. (C) Fluorescence light microscopy of unstained spread. Magnification, x1000.



View larger version (86K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 9. Fluorescence microscopy of enriched SC material. (A) Fluorescence micrograph of single SC-like structure from 2.0 M sucrose fraction prepared as described in MATERIALS AND METHODS. Magnification, x1000. (B) Low-magnification (x400) image of the same fraction.


 
View this table:
[in this window]
[in a new window]

  		Table 2. Enrichment of fluorescent material

The purification was examined on Western blots probed with antiserum to GFP (CLONTECH). The results revealed enrichment of an ~120-kD protein band, the expected size for Zip1::GFP (Fig 10), that was most intense in the fraction with the most fluorescent material (lane 6). This band was not present in the parallel sporulating wt ZIP1 nonfluorescent fraction (lane 7) or in vegetatively grown ZIP1::GFP cells (lane 9) and was much less intense in an unfractionated extract (lane 8). Little degradation of the protein was observed in any lane, suggesting that the Zip1::GFP from enriched SCs was intact. Interestingly, Zip1::GFP appears as a doublet, suggesting post-translational modification. The lower-molecular-weight form was enriched in the fractions containing the highest concentration of SCs. These Western blots were also probed with antibodies to Red1, another S. cerevisiae SC protein (ROCKMILL and ROEDER 1988 Down), which exhibited an almost identical enrichment in the fluorescent SC-containing fractions (not shown).



View larger version (31K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10. Western blot showing Zip1::GFP enrichment. In total, 5 µg of protein per lane was separated by PAGE, transferred to a PVDF membrane, and incubated with antibody to GFP; antigen antibody complexes were detected as described in MATERIALS AND METHODS. (Lanes 1–6) Enriched nuclei from sporulating strain EW104 (ZIP1::GFP/ZIP1::GFP, ndt80/ndt80) were lysed and fractionated on a sucrose step gradient as described in MATERIALS AND METHODS. Lanes 5 and 6 contained the majority of the fluorescent material and were collected as the enriched SC fractions. Little if any protein material and no fluorescent material were contained below these fractions. (Lane 7) Proteins from the equivalent fraction of lane 6 from sporulating wt Zip1 (nonfluorescent) cells (EW204; ndt80/ndt80) are included as a control to show that the antibody was specific for GFP. (Lane 8) Unfractionated whole-cell extract from sporulating strain EW104 shows 36-fold enrichment of Zip1::GFP. (Lane 9) Proteins from the equivalent fraction of lane 6 extracted from YPD-grown vegetative EW104 cells show that Zip1::GFP is sporulation specific.

Proteins extracted from the enriched SC preparations were analyzed by PAGE using proteins extracted from parallel fractions from identically treated vegetative nuclei from the same strain as a control (Fig 11). There was no clearly visible sporulation-specific 120-kD band indicative of Zip1::GFP. However, analysis of the proteins extracted from the 120-kD band using MALDI-TOF mass spectroscopy and the ProFound program (ZHANG and CHAIT 2000 Down) indicated the presence of Zip1-specific peptides. The most prominent meiosis-specific bands were extracted from the gel and analyzed by MALDI-TOF mass spectroscopy and the ProFound program (ZHANG and CHAIT 2000 Down). Four of five bands examined, Smc3, Spo22, Ams1, and Fox2, were conclusively identified and one was tentatively identified as Smc1. Smc3, a component of cohesin, was previously found to be on axial elements that form the SC. This protein was present in an ~140-kD doublet. The other member of this doublet could be only tentatively identified as Smc1 because this band contained contaminating peptides. Most interesting was the identification of the Spo22 protein. The gene that encodes this protein is required for efficient meiosis, produces a meiosis-specific transcript, and contains a phospholipase A signature (CHERRY et al. 2004 Down; PRIMIG et al. 2000 Down; G. TEVZADZE and R. E. ESPOSITO, personal communication). The other two proteins identified, Fox2 and Ams1, are peroxisomal (HILTUNEN et al. 1992 Down) and vacuolar enzymes, respectively (YOSHIHISA and ANRAKU 1990 Down). FOX2 produces a sporulation-induced transcript while AMS1 does not (CHU et al. 1998 Down). Two other meiosis-specific bands were analyzed but turned out to be previously characterized proteolytic cleavage products of Fox2 and Ams1 (YOSHIHISA and ANRAKU 1990 Down; HILTUNEN et al. 1992 Down). These results suggest that our SC preparations contain membrane contaminants, data consistent with the density of the fractions but also consistent with the perinuclear localization of SC that was most apparent in the ndt80 arrested nuclei. On the basis of these results it is likely that additional SC proteins will be identified by further studies on fractionating these preparations such as 2-D gel electrophoresis.



View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 11. Identification of sporulation-specific proteins from SC-enriched preparations. Coomassie blue-stained 10% SDS-polyacrylamide gel shows sporulation-specific proteins identified by MALDI-TOF mass spectrometry. M, molecular weight markers shown in kilodaltons. (Lane 1) SC-enriched fraction from sporulating strain EW104 (ZIP1::GFP/ZIP1::GFP, ndt80/ndt80) nuclei. (Lane 2) Parallel fraction from vegetatively grown strain EW104 nuclei.


*  DISCUSSION
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
 *RESULTS
 *DISCUSSION
*LITERATURE CITED

A ZIP1::GFP fusion that produces full-length fluorescent SCs enabled an analysis of these structures in living meiotic yeast cells and facilitated the partial purification of these supramolecular complexes. Zip1::GFP maintained its fluorescence in the presence of DTT, formaldehyde, and several mild detergents (data not shown) and produced SCs that also were capable of being visualized in surface spreads. On the basis of its ability to produce normal-looking SCs, rescue the spore inviability phenotype of zip1 deletions, and produce near wt levels of recombination, it appeared that the fluorescent fusion protein was mostly functional. However, complete function could not be assured because strains homozygous for the fusion exhibited a 1-hr delay in completing meiosis I. This delay is similar to that previously reported for several other mostly functional zip1 mutant alleles (TUNG and ROEDER 1998 Down). ZIP1::GFP/ZIP1 heterozygotes exhibited no delay but produced weakly fluorescent SCs. In spite of these minor limitations much was learned with this construct.

In S. cerevisiae, zygotene synapsis begins at DNA DSB sites that occur throughout the genome (CHUA and ROEDER 1998 Down). In our studies, we defined zygotene nuclei as those displaying fluorescent punctate dots and short filaments within the first 4 hr of sporulation. The zygotene foci that were observed were uniformly distributed throughout the nucleus, suggesting that the DSB initiated pairing interactions and early stages of SC formation occur all over the nucleus. This observation contrasts with those made on higher eukaryotes where synapsis is thought to begin near telomeres that are associated with the nuclear periphery (LOIDL 1990 Down). These contrasting observations nevertheless are consistent with the fact that recombination rates are highest near the ends of chromosomes in some of these higher organisms (VILLENEUVE 1994 Down; LANDER et al. 2001 Down; FROENICKE et al. 2002 Down) whereas recombination rates are more uniformly distributed throughout yeast chromosomes (MORTIMER and SCHILD 1980 Down; KABACK et al. 1989 Down).

In contrast to the uniform distribution of zygotene foci, mature full-length SCs were localized near the nuclear periphery in more than half of the early pachytene nuclei. The remaining nuclei exhibited a more uniform distribution of SCs that filled the entire nuclear volume. The perinuclear array appears to be genuine since it was present in both the ZIP1::GFP homozygote and the phenotypically wt ZIP1::GFP/ZIP1 heterozygote. In addition, a sporulation-induced, transient, but somewhat more diffuse perinuclear DNA array can be observed in Hoechst- or DAPI-stained wt nuclei that do not contain ZIP1::GFP. This array precedes meiosis I and occurs at a time that is consistent with pachytene. It furthermore persists in ndt80 homozygotes that are arrested in pachytene. While staging the perinuclear and uniform arrays in the ZIP1::GFP/ZIP1 heterozygote did not reveal a definitive temporal relationship, results with the ZIP1::GFP/ZIP1::GFP homozygote suggested that the perinuclear arrangement precedes the more uniform distribution. On the basis of these observations we propose that pachytene in S. cerevisiae consists of two distinct substages that we term pachytene A (perinuclear SC distribution) and pachytene B (uniform SC distribution). The inability to definitively stage these arrays in the heterozygote was mostly likely due to insufficient synchrony and the increased speed at which pachytene takes place compared to the homozygote. Furthermore, scoring was more difficult due to the weak fluorescence. It is also possible that pachytene is dynamic with SCs oscillating between the perinuclear and uniform distributions, so that SCs spend more time at the periphery at early times than at later times. Unfortunately, the ability to distinguish between these two alternatives was beyond our present technical capabilities.

Two perinuclear SC distributions could be observed, one where SCs were polarized to one side, the other where SCs were evenly distributed. During late zygotene or early pachytene in maize, the nucleolus moves from the center of the nucleus to the periphery, polarizing chromosomes to one side (DAWE et al. 1994 Down). While no kinetic differences were found for the two distributions in S. cerevisiae, perhaps the different perinuclear arrays are also somewhat dependent on nucleolar position, which has been shown to be perinuclear in thin-sectioned pachytene nuclei (BYERS and GOETSCH 1975 Down).

Nuclear structure was examined in ndt80 mutants that were arrested in pachytene. Approximately two-thirds were scored with perinuclear SC (pachytene A) while the remaining cells contained uniformly distributed SC (pachytene B). The fact that not all nuclei were perinuclear means that either a fraction of the nuclei may have been scored improperly due to limits in optical resolution or a fraction of the cells are capable of advancing to pachytene B without ndt80.

Initial observations of the ndt80 mutant suggest that SCs are not as fully condensed as in wt. In contrast, silver-stained surface spreads of ndt80 SC appear normal (XU et al. 1995 Down). At present we have no explanation for the difference other than to raise the possibility that ndt80 mutant SCs are not fully matured.

NDT80 has been described as being required for exit from pachytene. Since most ndt80 cells appeared to be arrested at the pachytene A stage, the NDT80 gene product may be more accurately described as being required for progression from pachytene A to pachytene B. The ndt80 mutants also suggest that the distinct SC distributions could correspond to the presence of specific recombination intermediates (ALLERS and LICHTEN 2001 Down; HUNTER and KLECKNER 2001 Down). Mutant ndt80 cells arrest with mostly perinuclear SCs and unresolved double Holliday junction intermediates. These results suggest that efficient processing of these intermediates may require the perinuclear localization. Perhaps the resolution of these intermediates is a prerequisite for transitioning from pachytene A to B. On the basis of the ndt80 results, the reorganization of SC from the periphery might be considered the first morphologically distinct phase following the pachytene arrest checkpoint. The mechanism for this reorganization is not known but could involve either the release of SC from the nuclear periphery or an overall change in nuclear membrane shape or nuclear volume.

We examined the role of NDJ1 and found that the Ndj1 protein, which interacts with the nuclear periphery-associated meiotic telomeres, is not involved in establishing the perinuclear organization of SCs. These results suggest that telomeres per se have little if any role in the perinuclear SC array. Nevertheless, it is possible that the uniform distribution of SCs at pachytene B requires the release of telomeres from the nuclear periphery.

The significance of the association of the SC with the nuclear periphery is not known. Perhaps the presumed movement of the nascent SC to the periphery facilitates pairing by physically removing the properly synapsed chromosomes from the unsynapsed ones. Another possibility is suggested by the observation that efficient repair of mitotic double-strand breaks requires tethering of chromosomes to the nuclear periphery (GALY et al. 2000 Down). Perhaps similar tethering of the SC to the nuclear periphery is required for the efficient repair of meiotic double-strand breaks as well. An association of SC near the nuclear pores might facilitate the localization of the meiotic recombination machinery following its import into the nucleus.

ZIP1::GFP was also found to be a useful marker to monitor enrichment of the SC from ndt80 pachytene-arrested cells. The crude SC preparations enabled the identification of several proteins that are known to be associated with the SC. In addition, a previously identified sporulation-induced protein Spo22 was also identified. These results suggest that this protein is a component of the SC. Nevertheless, since these preparations are considered crude, it is still possible that this protein is a contaminant.

Western blots showed that Zip1::GFP was highly enriched. It should not be considered surprising that no obvious 120-kD sporulation-specific band was visible in the stained gels, since many yeast proteins have a similar molecular weight. In addition, analysis of the 120-kD region by MALDI-TOF revealed Zip1 peptides, confirming the presence of this protein within this region. The Western blots also suggested the presence of multiple Zip1::GFP isoforms where the lower-molecular-weight form was more prominent in the purified fractions compared to the unfractionated nuclei. This observation suggests that the lower-molecular-weight band could be indicative of SC polymerization. Alternatively, the higher-molecular-weight band seen in unfractionated nuclei might be due to an association of ZIP1 with nucleic acid since this fraction was not treated with nucleases. As the overall yield of fluorescent protein exceeded 60% and SC-like structures were observable throughout the purification it is very likely that bona fide SCs were purified and unlikely that our preparation contained only polycomplex. Nevertheless, it is certainly possible that our preparations contained some polycomplex. It should be further pointed out that these results were all obtained with ndt80-arrested cells using a modified form of the Zip1 protein that exhibited near but not completely wt activity. Nevertheless, these observations on the purified SC material suggest that these methods will prove useful for the further analysis of the yeast SC.

To conclude, the ZIP1::GFP construct that is described here exhibited near wt behavior and produced fluorescent SCs that were easily visualized both in living cells and in chromosome spreads. Our studies showed that ZIP1::GFP is useful for observing the organization and behavior of SCs as well as serving as a convenient marker in the partial purification of these structures. This construct should be useful for studying other aspects of meiotic pairing and may also facilitate karyotyping since spreads do not require staining.


*  FOOTNOTES

1 Present address: Department of Molecular, Cellular and Developmental Biology, Yale University, New Haven, CT 06511. Back
2 Present address: Max Planck Institute of Molecular Cell Biology and Genetics, 01307 Dresden, Germany. Back


*  ACKNOWLEDGMENTS

We are most grateful to Shirleen Roeder, Beth Rockmill, Pam Silver, Nancy Kleckner, Scott Keeney, Breck Byers, Arnold Barton, and Rita Cha for strains, recombinant plasmids, antibodies, and/or valuable discussions. We especially thank Hong Li for invaluable help with the mass spectrometer and Sepp Loidl for valuable discussions, comments on the manuscript, generous help with microscopy, and opening his laboratory to us. These studies were supported by grants from the National Science Foundation (0136278 and 0100831).

Manuscript received October 20, 2003; Accepted for publication January 15, 2004.


*  LITERATURE CITED
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
 *RESULTS
 *DISCUSSION
 *LITERATURE CITED

ALLERS, T. and M. LICHTEN, 2001  Differential timing and control of noncrossover and crossover recombination during meiosis. Cell 106:47-57.[CrossRef][Medline]

BAKER, B. S., A. T. CARPENTER, M. S. ESPOSITO, R. E. ESPOSITO, and L. SANDLER, 1976  The genetic control of meiosis. Annu. Rev. Genet. 10:53-134.[CrossRef][Medline]

BISHOP, D. K., 1994  RecA homologs Dmc1 and Rad51 interact to form multiple nuclear complexes prior to meiotic chromosome synapsis. Cell 79:1081-1092.[CrossRef][Medline]

BURKE, D., D. DAWSON and T. STEARNS, 2000 Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Plainview, NY.

BURNS, N., B. GRIMWADE, P. B. ROSS-MACDONALD, E. Y. CHOI, and K. FINBERG et al., 1994  Large-scale analysis of gene expression, protein localization and gene disruption in Saccharomyces cerevisiae. Genes Dev. 8:1087-1105.[Abstract/Free Full Text]

BYERS, B. and L. GOETSCH, 1975  Electron microscopic observations on the meiotic karyotype of diploid and tetraploid Saccharomyces cerevisiae.. Proc. Natl. Acad. Sci. USA 72:5056-5060.[Abstract/Free Full Text]

CAO, L., E. ALANI, and N. KLECKNER, 1990  A pathway for generation and processing of double-strand breaks during meiotic recombination in S. cerevisiae.. Cell 61:1089-1101.[CrossRef][Medline]

CHERRY, M., C. BALL, K. DOLINSKI, S. DWIGHT, M. HARRIS et al., 2004 Saccharomyces Genome Database (http://genome-www.stanford.edu/Saccharomyces/).

CHRISTOFFERS, K., H. LI, S. KEENAN, and R. HOWELLS, 2003  Purification and mass spectrometric analysis of the µ opioid receptor. Mol. Brain Res. 118:119-131.[Medline]

CHU, S., J. DERISI, M. EISEN, J. MULHOLLAND, and D. BOTSTEIN et al., 1998  The transcriptional program of sporulation in budding yeast. Science 282:699-705.[Abstract/Free Full Text]

CHUA, P. R. and G. S. ROEDER, 1998  Zip2, a meiosis-specific protein required for the initiation of chromosome synapsis. Cell 93:349-359.[CrossRef][Medline]

CONRAD, M. N., A. M. DOMINGUEZ, and M. E. DRESSER, 1997  Ndj1p, a meiotic telomere protein required for normal chromosome synapsis and segregation in yeast. Science 276:1252-1255.[Abstract/Free Full Text]

DAWE, R. K., J. W. SEDAT, D. A. AGARD, and W. Z. CANDE, 1994  Meiotic chromosome pairing in maize is associated with a novel chromosome organization. Cell 76:901-912.[CrossRef][Medline]

DONG, H. and G. S. ROEDER, 2000  Organization of the yeast Zip1 protein within the central region of the synaptonemal complex. J. Cell Biol. 148:417-426.[Abstract/Free Full Text]

EGEL, R., 1978  Synaptonemal complex and crossing-over: Structural support or interference? Heredity 41:233-237.[Medline]

EIJPE, M., C. HEYTING, B. GROSS, and R. JESSBERGER, 2000  Association of mammalian SMC1 and SMC3 proteins with meiotic chromosomes and synaptonemal complexes. J. Cell Sci. 113:673-682.[Abstract]

ESPONDA, P. and G. GIMENEZ-MARTIN, 1972  The attachment of the synaptonemal complex to the nuclear envelope. Chromosoma 38:405-417.[CrossRef][Medline]