SMC Protein-Dependent Chromosome Condensation during Aerial Hyphal Development in Streptomyces

Members of the SMC (structural maintenance of chromosomes) proteinfamily play a central role in higher-order chromosome dynamicsfrom bacteria to humans. So far, studies of bacterial SMC proteinshave focused only on unicellular rod-shaped organisms that divideby binary fission. The conversion of multigenomic aerial hyphaeof the mycelial organism Streptomyces coelicolor into chainsof unigenomic spores requires the synchronous segregation ofmultiple chromosomes. Here we focus on the contribution of SMCproteins to sporulation-associated chromosome segregation inS. coelicolor. Deletion of the smc gene causes aberrant DNAcondensation and missegregation of chromosomes (7.5% anucleatespores). In vegetative mycelium, immunostained SMC proteinswere observed sporadically, while in aerial hyphae about toundergo sporulation they appeared as irregularly spaced fociwhich accompanied but did not colocalize with ParB complexes.Our data demonstrate that efficient chromosome segregation requiresthe joint action of SMC and ParB proteins. SMC proteins, similarlyto ParAB and FtsZ, presumably belong to a larger group of proteinswhose expression is highly induced in response to the requirementof aerial hyphal maturation.

INTRODUCTION

Top

INTRODUCTION

Higher-order chromosome structure is particularly critical inthe segregation of replicated chromosomes during cell divisionin eubacteria, eukaryotes, and archaea. Recent studies haveshown that members of the SMC (structural maintenance of chromosomes)protein family play a central role in higher-order chromosomedynamics, acting either as condensins or cohesins (15-17, 30,32, 36, 40). SMC are large proteins (110 to 170 kDa) that formdimers. Their N- and C-terminal parts form a single head domain(see references 13, 15-17, and 38 for reviews). SMC proteinsexhibit ATPase activity that is required for their dynamic interactionwith DNA (17). The results of recent studies demonstrated thatbacterial SMC protein functions in complexes with other proteins,i.e., ScpA and ScpB (Scp's are segregation and condensationproteins) (6, 29, 37).

In contrast to eukaryotes, in eubacteria the cell-cycle events,including replication, condensation, and segregation, take placesimultaneously (16, 26). Thus, during the bacterial cell cycle,chromosomes must undergo dynamic changes. So far, studies regardingchromosome dynamics have focused on Bacillus subtilis, Caulobactercrescentus, and Escherichia coli. In these organisms, SMC proteins(or MukB, a functional homologue in E. coli) appear to playsimilar roles; they are involved in chromosome organizationand segregation. The genes encoding these proteins are not essential,although their deletion results in temperature-sensitive growth(2, 9, 22, 31, 32, 38). In B. subtilis, smc deletion mutantsexhibit complex phenotypes under permissive conditions, includingpoorly compacted nucleoids and a relatively large amount ofanucleate cells (up to 15% of the cells were anucleate) (10).Interestingly, the inactivation of smc enhances the chromosomesegregation defect of spo0J- or soj-null mutants (25). The proteinsencoded by these genes, Soj (ParA, an ATPase) and Spo0J (ParB,a DNA-binding protein), form a nucleoprotein complex at theoriC region that ensures its proper segregation. Many attemptshave been made to localize SMC proteins in bacterial cells.Despite some differences in SMC protein localization, comprehensivemicroscopic observations in B. subtilis revealed that SMC proteinsform foci or a punctate pattern associated with the nucleoid(42). SMC protein colocalizes with its interacting partnersScpA and ScpB, and the specific localization of SMC proteindepends on both Scp proteins, showing that all three componentsof the SMC complex are required for proper localization (41).

Here we extend the studies of SMC proteins, which have previouslyfocused only on unicellular rod-shaped bacteria that divideby binary fission, to the more-complex bacterium Streptomycescoelicolor. Streptomycetes are gram-positive multicellular sporulatingbacteria that grow by tip extension and hyphal branching toform a dense mycelial network of vegetative hyphae in whichseptation occurs at some distance from the tips (7). Compartmentsof the vegetative hyphae contain several copies of the large(8 to 9 Mbp), linear uncondensed chromosome (4). Older Streptomycescolonies differentiate to form aerial hyphae, which subsequentlydevelop into chains of spores (3, 4). The rapid growth of theaerial hyphae is accompanied by intensive replication of thechromosomes, such that 50 or more uncondensed nonsegregatedchromosomes may be present in one long tip compartment (33).After an aerial hypha has stopped growing, a large number ofFtsZ rings form a regular ladder in the long tip compartment,giving rise to sporulation septa that delimit the prespore compartments(8). At this stage, dozens of chromosomes are condensed anduniformly aligned along the hyphal tip, ensuring that each presporecompartment receives a single copy (21).

Very little is known about the orchestration of major cell cycleprocesses, particularly the synchronization of condensationand the segregation of several dozen chromosomes during theformation of unigenomic spores. Here we focus on the contributionof SMC proteins to sporulation-associated chromosome segregationin S. coelicolor A3(2), the model organism for the Streptomycesspecies.

MATERIALS AND METHODS

Top

MATERIALS AND METHODS

DNA manipulation and bacterial strain growth conditions. DNA manipulations were carried out according to standard protocols(34). Enzymes were supplied by Fermentas, New England BioLabs,or Roche; fluorescent reagents were from Invitrogen (MolecularProbes); and oligonucleotides were from The Institute of Biochemistryand Biophysics (Poland). The S. coelicolor and E. coli strainsare listed in Table 1. Culture conditions, media, antibioticconcentrations, and transformation and conjugation methods followedthe general procedures for E. coli (34) and Streptomyces (23).S. coelicolor was cultivated in Trypticase soy broth-yeast extract-maltextract (1:1) complex liquid medium or on mannitol soy flouragar plates unless otherwise stated.

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

TABLE 1. Strains used in this study

Construction of strains carrying a deletion of smc or scpAB. A PCR-targeted gene replacement strategy (11, 12) was used forthe construction of deletions of the smc and scpAB genes. Deletionof smc was created by replacing the smc gene within cosmid 7A1with apra-oriT, an apramycin resistance (Apra^r) cassette flankedby FLP recombinase recognition target sites. The cassette wasamplified from pIJ773 (12) with the pair of primers p_{Delsmc_f}and p_{Delsmc_r} (Table 2). The resulting construct, 7A1 ${Delta}$ smc::apra,was used to transform ET12567/pUZ8002, from which it was mobilizedinto S. coelicolor M145 or other S. coelicolor derivatives (Table1). The Apra^r exconjugants were screened for the loss of Kan^r(the kan gene is a part of cosmid 7A1 ${Delta}$ smc::apra), indicatingthat there was double-crossover allelic exchange of the smclocus.

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

TABLE 2. Oligonucleotides used in this study

An in-frame scpAB deletion was created by using the appropriatepair of primers, p_{Delspn_f} and p_{Delspn_r}. Cosmid StI51 ${Delta}$ scpAB::aprawas digested with SwaI and religated. The amp gene in the SuperCospart of cosmid StI51 ${Delta}$ scpAB was subsequently exchanged for anapra-oriT cassette, yielding StI51 ${Delta}$ scpAB amp::apra-oriT. Thisconstruct was used to transform ET12567/pUZ8002 and for conjugationinto S. coelicolor M145. The Apra^r exconjugants were screenedfor the loss of both Apra^r and Kan^r, indicating a double-crossoverallelic exchange of the ${Delta}$ scpAB locus of M145.

Purification of SMC, ScpA, and ScpB proteins. The vectors pET-21a and pGEX-6P-2 were chosen to overexpressthe recombinant SMC protein and the fusion proteins glutathione-S-transferase(GST)-ScpA and GST-ScpB. Cosmid 7A1, which contains the entirecoding region of the S. coelicolor smc gene, was used for subcloningthe smc gene into pET-21a. The 5' end (1,135 bp) of the smcgene was PCR amplified using the appropriate pair of primers,p_{Smc_5_fw} and p_{Smc_5_rv} (Table 2). The resulting DNA fragmentwas digested with NdeI and XhoI and inserted into the NdeI-and XhoI-restricted pET-21a. Afterward, the rest of the smcgene was subcloned into the expression vector as follows. AnXhoI restriction site was created directly before the stop codonof smc on cosmid 7A1 by using PCR targeting directly, and thena 2,456-bp fragment of smc was cut out and subcloned into NcoI-and XhoI-restricted pET-21a carrying smc (1,135 bp). The S.coelicolor His₆-SMC protein was isolated as described previously(43).

The scpA and scpB genes were PCR amplified with the appropriatepairs of primers, p_{ScpA_fw} and p_{ScpA_rv} and p_{ScpB_fw} and p_{ScpB_rv}(Table 2). The PCR products were cloned into pGEM-T Easy (Promega)and later recloned as BamHI-XhoI fragments into pGEX-6P-2. Allthe PCR-derived clones were analyzed by DNA sequencing to checktheir fidelity. The fusion proteins GST-ScpA and GST-ScpB werepurified on glutathione-Sepharose 4B as described earlier (27).For removal of the GST part, the bound fusion proteins weretreated with PreScission protease (Amersham-Pharmacia-Biotech)and ScpA or ScpB protein was released from the beads. The purifiedSMC, ScpA, and ScpB proteins were more than 95% pure, as judgedby sodium dodecyl sulfate-polyacrylamide gel electrophoresis(SDS-PAGE) analysis (see Fig. S1A in the supplemental material).

Preparation of antisera. Antisera were obtained by immunization of rabbits with the purifiedSMC and ScpA proteins mixed with the adjuvant monophosphoryllipid A (a derivative of lipid A from Hafnia alvei PCM 1200)(5). Serum samples were taken 10 days after the second boosterinjection. Cellular particles were removed by centrifugation,and the serum was stored at –20°C. Anti-SMC and anti-ScpAantibodies were purified by solid-phase chromatography (14).

SDS-PAGE and Western blotting. Proteins were separated by SDS-PAGE (10% or 12% acrylamide)as described by Laemmli (24) and transferred to a nitrocellulosemembrane. The membrane was blocked for 1 h at room temperaturewith TBST (10 mM Tris-HCl, pH 8.0, 100 mM NaCl, 0.05% Tween20) containing 3% bovine serum albumin (BSA) and subsequentlyincubated with polyclonal anti-SMC antibody. Afterwards themembrane was incubated with a goat anti-rabbit secondary antibodyconjugated with alkaline phosphatase. The membrane was stainedwith 5-bromo-4-chloro-3-indolylphosphate and nitroblue tetrazolium.

Dot blot overlay assay. SMC or ScpB protein was spotted onto the nitrocellulose membranestrips. After drying, the strips were placed in a blocking solution(0.5% BSA, 0.25% gelatin, 0.5% Nonident P-40, 100 mM NaCl, 50mM Tris-HCl, pH 7.5) and subsequently incubated with ScpA protein.The strips were exhaustively washed with TBST buffer containing3% BSA. Afterwards, ScpA protein was detected by a polyclonalanti-ScpA antibody as described above.

Microscopy. Strains used for fluorescence microscopic observations wereinoculated into the acute-angle junction of coverslips insertedat an angle of 45° into minimal medium containing 1% mannitol(23) and cultured for 24 to 72 h. The staining procedures weredescribed previously (21, 35). Briefly, mycelium was fixed for10 min with a paraformaldehyde-glutaraldehyde mixture, digestedfor 2 min with 2 mg ml^–1 lysozyme, washed with phosphate-bufferedsaline (PBS), and blocked with 2% BSA. For immunostaining, sampleswere incubated with purified antibody against SMC (1:2,500 dilution)overnight, washed six times with PBS, and then incubated for1 h with secondary antibody (antirabbit) conjugated with AlexaFluor 546. For DNA staining, samples were incubated with 0.1to 1 µg ml^–1 4',6'-diamidino-2-phenylindole (DAPI;Molecular Probes), and for cell wall visualization with 10 µgml^–1 wheat germ agglutinin (WGA)-Texas Red or Alexa Fluor350 conjugate (Molecular Probes). After five washes with PBS,the coverslips were mounted in Vectashield (Vector Laboratories)antifade reagent. Fluorescence microscopy was carried out usinga Zeiss Axio Imager Z1 microscope equipped with a 100x objective.The relative intensities of the ParB foci (maximal pixel valuewith the background signal subtracted) in images of aerial hyphaeof the M145 ${Delta}$ smc parB-egfp and M145 parB-egfp (expressing enhancedgreen fluorescent protein) strains taken with the same camerasettings were measured.

RESULTS

Top

RESULTS

Identification of the Streptomyces coelicolor smc, scpA, and scpB homologues. Sequence analysis identified smc, scpA, and scpB genes in allthe Streptomyces chromosomes sequenced so far (http://www.genomesonline.org/).The smc, scpA, and scpB genes encode putative proteins withcalculated molecular masses of 128, 30, and 24 kDa, respectively.Sequence analysis of the S. coelicolor chromosome indicatesthat the smc gene is presumably independently transcribed, whilescpA and scpB may be cotranscribed as part of a larger clusterof genes (scpA being the first gene of the operon, with scpBand four other genes, encoding putative pseudouridine synthase,putative DNA hydrolase, putative glycohydrolase, and a hypotheticalprotein, downstream of this gene) (1). The SMC, ScpA, and ScpBproteins from S. coelicolor share extensive homology with chromosomallyencoded orthologues from other bacteria (sequence similaritiesto those of B. subtilis are 47%, 47%, and 44%, respectively).All the SMC signature motifs (Walker A and B, ABC signature)and domain organization (coiled-coil domains, hinge) are conservedin the predicted Streptomyces SMC protein.

In order to test whether ScpA interacts with SMC/ScpB, as thecorresponding proteins do in B. subtilis, a dot blot overlayexperiment was performed (see Fig. S1B in the supplemental material).Purified SMC and ScpB proteins (see Materials and Methods andsee Fig. S1A in the supplemental material) were spotted ontoa nitrocellulose membrane, followed by incubation with ScpA.The membrane was then washed extensively and probed with anti-ScpAantibodies. Immunoreactive spots were observed for the two proteinsbut not for the negative controls BSA or binding buffer alone,indicating that ScpA interacts with SMC and ScpB. These interactionspresumably exist in vivo as well.

Deletion of smc or scpAB genes causes aberrant condensation and missegregation of chromosomes. In B. subtilis and C. crescentus, deletion of the smc gene isnot lethal, but it does cause defects in DNA compaction andchromosome segregation. In order to evaluate the importanceof the smc and scpAB genes for viability, growth, and chromosomesegregation in Streptomyces, deletion mutants were constructed.

Since the smc gene is supposed to be independently expressed,the insertion-deletion mutant was constructed by replacing theentire gene with the apramycin resistance cassette (see Materialsand Methods). The deletion was confirmed by Southern hybridization(data not shown) and Western blotting (Fig. 1B). The resulting ${Delta}$ smc mutant strain, S. coelicolor, showed no obvious defect ingrowth or morphology when cultured in liquid or on solid medium.However, visualization of the DNA (DAPI stained) and cell walls(stained with WGA conjugated with Texas red) of sporulatinghyphae revealed missegregation of chromosomes; approximately8% of prespore compartments were anucleate. Since both the smcand partitioning gene (parA and parB) deletion mutants havedefects in chromosome segregation, we tested the phenotypesof double mutants ${Delta}$ smc ${Delta}$ parA and ${Delta}$ smc ${Delta}$ parB. The double mutantsshowed no obvious defects in morphology. The number of anucleateprespores resulting from ${Delta}$ smc ${Delta}$ parB mutations was approximatelyequal to the sum of the anucleate prespores observed with thetwo single mutations (Table 3). Interestingly, the total amountof anucleate prespores resulting from the double ${Delta}$ smc ${Delta}$ parA mutationwas approximately the same as in the ${Delta}$ parA mutant strain (Table3).

View larger version (26K):
[in this window]
[in a new window]

FIG. 1. Influence of smc or scpAB deletion on chromosome condensation and segregation. (A) Examples of images showing hyphae with cell walls stained with WGA (red) and DNA stained with DAPI (blue). Anucleate compartments are indicated by white arrows. Scale bar, 5 µm. (B) Western blotting. Total proteins (20 µg) were separated by SDS-PAGE, and then SMC protein was identified by using antibodies raised against the recombinant SMC protein. Lanes: 1, wild type; 2, ${Delta}$ smc mutant. Molecular sizes in kDa are shown on the left. (C) DNA signal intensity in spore compartments: fluorescent intensity of DNA stained with DAPI, presented in arbitrary units, was measured along a line drawn through the middle region of the spore compartments. (D) Percentage of prespores exhibiting different degrees of DNA compaction (length of nucleoid in prespores). wt, wild type; h, hour.

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

TABLE 3. Segregation defects in ${Delta}$ smc and ${Delta}$ scpAB strains

Deletion of smc also perturbed chromosome condensation in comparisonwith the chromosome condensation of the wild type; the ${Delta}$ smc mutant'snucleoids filled almost all of a prespore's volume and did notstain very brightly with DAPI (Fig. 1A, C, and D), whereas thenucleoids of the wild type became smaller and stained very brightlyduring the formation of prespores (Fig. 1A, C, and D). Thissuggests that the nucleoids of the ${Delta}$ smc mutant are less condensedduring the early stages of sporulation (the smaller volume ofmature spores made it difficult to assess chromosome condensationat later stages).

The scpA and scpB genes are supposed to be part of a largeroperon. Indeed, disruption of the scpAB genes with the insertionof the Apra cassette caused markedly delayed growth (data notshown), suggesting perturbation in the transcription of genesthat are located downstream of scpAB (including sco1768, encodinga putative pseudouridine synthase). To avoid polar effects ofscpAB deletion on these genes, a strain carrying a nonpolar,unmarked, in-frame deletion of scpAB was constructed (see Materialsand Methods). The scpAB unmarked deletion was verified by Southernhybridization. The S. coelicolor ${Delta}$ scpAB mutant showed a phenotypesimilar to that of the S. coelicolor ${Delta}$ smc mutant, including aberrationsin chromosome condensation and segregation (Fig. 1A and C),although slightly fewer anucleate prespores were observed (6.3%).

In conclusion, these observations show that the SMC, ScpA, andScpB proteins are involved in DNA condensation and influencechromosome segregation during sporulation.

In aerial hyphae, SMC forms irregularly spaced foci before septation. In Streptomyces, dozens of linear chromosomes have to condenseand segregate accurately and synchronously during sporulationto ensure that each spore receives a single copy of the chromosome.

To localize SMC protein in the hyphae of a sporulating colony,rabbit polyclonal antibodies raised against the purified SMCprotein (see Materials and Methods) were used in immunofluorescencemicroscopy. The affinity-purified anti-SMC antibodies reactedspecifically with a protein of approximately 130 kDa that waspresent in a whole-cell extract of wild-type S. coelicolor,but they did not detect any protein when the S. coelicolor ${Delta}$ smcmutant strain was tested (Fig. 1B). In agreement with the Westernblot analysis, immunofluorescent signals (foci) were observedin the wild-type strain but not in the smc deletion mutant (Fig.2). In the vegetative mycelium, sporadic immunostaining of SMCwas observed (Fig. 2), while in the aerial hyphae, the numberof foci increased as hyphae became close to full-length, reachinga maximum level before septation. At this stage, SMC proteinwas visualized as clusters of irregularly spaced foci. The abundanceof foci was approximately seven foci/10 µm (Fig. 2). Theywere observed over uncondensed DNA; however, when the DNA wascondensed in prespores, SMC foci were no longer detectable (Fig.2). The foci disappeared after septum formation (see Fig. 4and 6).

View larger version (32K):
[in this window]
[in a new window]

FIG. 2. Immunolocalization of SMC protein in the vegetative and aerial mycelium of the wild-type strain and in the ${Delta}$ smc mutant. Immunolabeled SMC (red), cell wall stained with WGA (gray), and DNA stained with DAPI (blue) in the mycelium of strain grown for 24 to 48 h on solid minimal medium. Scale bar, 5 µm. wt, wild type; h, hour.

View larger version (37K):
[in this window]
[in a new window]

FIG. 4. Localization of SMC and ParB proteins in the aerial hyphae. (A) Immunolocalization of SMC protein in a strain expressing ParB-EGFP fusion protein. The strain was grown for 44 h on solid minimal medium. SMC immunofluorescence is visible as red, ParB-EGFP as green, and cell walls in grey scale. The alphanumerical abbreviations indicate developmental stages as shown in panel B. Scale bar, 5 µm. (B) Developmental stages. Table shows percentages of aerial hyphae falling into specific developmental subclasses (100% = 102 apical compartments). Boldface indicates the most frequently observed stages.

View larger version (33K):
[in this window]
[in a new window]

FIG. 6. Localization of SMC and FtsZ in the aerial hyphae. (A) Immunolocalization of SMC protein in a strain expressing FtsZ-EGFP fusion protein. The strain was grown for 44 h on solid minimal medium. SMC immunofluorescence is visible as red, FtsZ-EGFP as green, and cell walls in grey scale. The alphanumerical abbreviations indicate developmental stages as shown in panel B of the figure. Scale bar, 5 µm. (B) Developmental stages. Table showing percentages of aerial hyphae falling into specific developmental subclasses. (100% = 105 apical compartments). Boldface indicates the most frequently observed stages.

Correlation of SMC localization with replication and segregation. In rod-shaped bacteria dividing by binary fission, chromosomereplication, condensation, and segregation take place simultaneously;the nascent origin regions are condensed and segregated towardthe cell poles soon after replication starts (26, 36, 40). InStreptomyces, several dozen chromosomes need to be condensedand uniformly segregated along the hyphal tip, ensuring thateach prespore compartment receives a single copy (21). In orderto correlate SMC localization with particular cell-cycle events(replication and segregation) and stages of aerial hyphal development,SMC protein was immunolocalized in strains which allow the monitoringof replication (DnaN-EGFP) (33) and formation of partitioningcomplexes (ParB-EGFP) (19).

SMC and DnaN-EGFP foci formations are presumably spatially restricted. Rapid growth of aerial hyphae is accompanied by intensive replicationof the chromosomes, such that 50 or more nonsegregated chromosomesmay be present in one long apical compartment (33). To correlatethe appearance of SMC foci with ongoing replication and aerialhyphal maturation, the SMC protein was immunostained. The examinationof more than 100 apical compartments (n = 112) of the strainexpressing DnaN-EGFP revealed that SMC foci did not generallycolocalize with replisomes (Fig. 3). Instead, replisomes and/orSMC protein were frequently detected as separated arrays offoci representing, respectively, zones of intensive replicationand/or condensation (Fig. 3). Thus, intensive replication andthe formation of SMC foci appear to be mutually exclusive, presumablydue to temporal separation of both processes, with the formerpreceding the latter.

View larger version (10K):
[in this window]
[in a new window]

FIG. 3. Immunolocalization of SMC protein in the aerial hyphae of a strain expressing DnaN-EGFP fusion protein. The strain was grown for 66 h on solid minimal medium. SMC immunofluorescence is visible as red, DnaN-EGFP as green, and cell wall in grey scale. Scale bar, 5 µm. h, hour.

ParB and SMC cooperate at the stage of chromosome segregation. In young aerial hyphae, ParB-EGFP is visible as irregularlyspaced and rather weak foci. However, during subsequent stepsof aerial hyphal development, ParB forms bright and regularlydistributed foci (19). They appear before DNA condensation anddisassemble after septum formation. To define the timing ofSMC foci formation with respect to ParB complexes, a substantialnumber of presporulation aerial hyphae were examined (n = 102).Apical compartments containing ParB-EGFP foci were classifiedas described earlier (Fig. 4B) (21). It was found that SMC fociappeared before the assembly of regular ParB complexes (Fig.4; B1S1, 21%). During later developmental stages, the numberof SMC foci increased (B1S2, 26%), reaching a "maximum" levelwhen the regular ParB complexes were formed (B2S2, 14%). Interestingly,SMC foci disappeared after septum formation, while ParB fociwere still visible (Fig. 4; B3S0 14%). Taken together, SMC isobserved as irregularly spaced foci which accompany but do notcolocalize with ParB complexes in aerial hyphae about to undergosporulation. SMC foci disappeared earlier than ParB complexes.

Deletion of smc influences the formation of ParB segregation complexes. To determine whether SMC protein affects the positioning and/orformation of segregation complexes, we analyzed ParB localizationin an smc deletion strain. The smc deletion was introduced intoa strain which expressed ParB-EGFP protein instead of the wild-typeParB. The resulting strain, S. coelicolor parB-egfp ${Delta}$ smc, wasused to monitor microscopically the formation of ParB-EGFP complexes.As in the wild type, we observed arrays of regularly spacedParB complexes, but they appeared smaller and less intenselyfluorescent; the average signal intensity of the ParB foci inthe ${Delta}$ smc mutant was about 70% of that in the wild type (see Materialsand Methods and Fig. 5). The results indicate that SMC proteinmay facilitate the formation of large segregation complexes.

View larger version (18K):
[in this window]
[in a new window]

FIG. 5. Influence of smc deletion on ParB complex formation. (A) Examples of merged images showing ParB-EGFP (green) and cell walls stained with WGA (gray). Scale bar, 5 µm. (B) Frequency of occurrence of ParB foci with indicated signal intensity (in arbitrary units) in aerial hyphae of strain M145 parB-egfp and parB-egfp ${Delta}$ smc mutants.

SMC foci appear shortly before FtsZ spiral formation. During sporulation, the key cell division protein FtsZ formsspiral-shaped filaments along hyphae that are gradually replacedby regularly spaced rings (8). In order to correlate SMC proteinlocalization more precisely in relation to particular stagesof cytokinesis, we immunostained SMC protein in strain K202,in which, in addition to FtsZ encoded by the normal chromosomalcopy, FtsZ-EGFP fusion protein is expressed from an integratedplasmid (8). Several compartments had SMC signals but no FtsZ,implying that SMC aggregation marginally preceded FtsZ aggregation(Fig. 6B; Z0S1 7.0%, and Z0S2, 5.8%). Compartments exhibitingFtsZ signals were grouped into three subclasses as describedpreviously (21), and the abundance of SMC foci in each subclasswas estimated (Fig. 6B). A substantial number of compartmentscontained both SMC foci and FtsZ spirals (Z1S2, 26.7%). SMCprotein was also observed in compartments with some spiralsand some irregular Z-rings (Z2S2, 17.4%) and in compartmentswhere regular Z-rings were already present (Z3S2, 10.5%). Thefoci and the Z-rings disappeared rapidly after septum formation(sS0, 13.9%).

DISCUSSION

Top

DISCUSSION

In eukaryotes, major events of the chromosome cycle, i.e., replication,cohesion, condensation, and segregation, are temporally separatedand occur at discrete stages of the cell cycle. In contrast,replication, condensation, and chromosome segregation take placesimultaneously in rod-shaped bacteria. Not much is known aboutthe coupling of replication, condensation, and segregation inmultinucleoid compartments of filamentous multicellular bacteria,such as Streptomyces. Here we studied the influence of smc,scpA, and scpB deletions on the condensation of Streptomyceschromosomes and attempted to relate this process to their replicationand segregation during the differentiation of sporulating aerialhyphae.

Genes encoding the SMC, ScpA, and ScpB proteins are presentin the sequenced Streptomyces species S. coelicolor, S. avermitilis,and S. scabies. As for other bacteria, the deletion of smc revealedthat the product of this gene is not essential to S. coelicolor.However, smc deletion affects DNA condensation (Fig. 1) andcorrect chromosome segregation during sporulation (Table 3).The scpAB deletion mutant displays a phenotype similar to thatof ${Delta}$ smc (Table 3). In addition, we demonstrated in vitro thatthe ScpA protein interacts with the ScpB and SMC proteins. Thus,the ScpA, ScpB, and SMC proteins presumably act in concert invivo. Probably, SMC/ScpA/ScpB protein complexes interact randomlywith DNA, with subsequent interactions between complexes bringingabout condensation.

Developmental-stage-specific assembly of irregularly spaced SMC complexes in aerial mycelium. In Streptomyces hyphae, linear chromosomes are uncondensed duringboth vegetative and aerial growth and remain so until sporulationseptation is completed. Using in situ immunodetection, we demonstratedthat SMC foci occur sporadically in vegetative hyphae. In contrast,irregularly spaced foci are detectable over uncondensed DNAin aerial hyphae; their number increases during aerial hyphaldevelopment, reaching a maximum before septation. At this stage,the number of SMC foci was about equal to or slightly exceededthe number of ParB complexes (Fig. 4). Since each ParB complexis bound to a single chromosome (21), we can assume that statisticallythere is one SMC focus per nucleoid. In contrast to ParB complexes,the SMC foci are not regularly positioned. Like other SMC proteins,S. coelicolor SMC protein presumably does not exhibit any sequencespecificity. Thus, we assume that SMC molecules bind randomlyto a few distinct regions and, by bringing them into close proximity,form a condensation center, i.e., a condensome. The foci presumablyreflect ongoing DNA condensation. Indeed, the lack of SMC proteincauses inefficient condensation of chromosomes in prespores(Fig. 1).

The presence of SMC foci is presumably a direct consequenceof increased SMC protein production (see Fig. S2 in the supplementalmaterial) in aerial hyphae; smc transcription is probably developmentallycontrolled. This is consistent with the idea that there is aneed for a greater amount of SMC protein at the time of sporulation;several dozen chromosomes have to be condensed simultaneouslyin order to fit into tiny prespore compartments. SMC complexesare transiently localized at the time of sporulation, when DNAhas to be compacted and segregated. Thus, SMC, like FtsZ (8)and ParAB (19, 20), seems to belong to the larger group of proteinswhose expression is highly induced in response to the requirementof aerial hyphal maturation. The SMC foci, like FtsZ and ParAB,disappear after septation. In B. subtilis, when the cells enterthe stationary phase, SMC protein is subject to induced proteolysis(28). A future question will be to identify the agents of thispresumed proteolysis during spore maturation, as well as theirmechanisms of activation.

Efficient chromosome segregation requires the joint action of ParB and SMC. In addition to the effect of smc and scpAB deletion on DNA condensation,our observations suggest a role for SMC/ScpA/ScpB complexesin accurate chromosome segregation: the elimination of any ofthese genes resulted in the relatively frequent formation ofanucleate prespores (about 7.5%) compared with their rate offormation in the wild-type strain (1.4%). Moreover, in the ${Delta}$ smcmutant, the ParB foci were smaller and less intensive, althoughthe ParB-EGFP protein was produced at about wild-type levels(data not shown). In S. coelicolor, ParB binds to numerous parSsites scattered over a 400-kbp chromosomal segment containingoriC and assembles into a large nucleoprotein complex that isnecessary for proper chromosome segregation (18, 19). Probably,condensation brought about by SMC/ScpA/ScpB protein complexesis needed for the effective formation of segregosomes, perhapsby bringing parS sequences closer together by condensing theDNA around the oriC region. The resulting nucleoprotein complexesmay enhance ParB oligomerization. Condensomes and segregosomesdo not colocalize, but they appear at nearly the same time,at the beginning of aerial hyphal maturation. In contrast toB. subtilis (31), deletion of the smc gene does not influencethe positioning of the regularly spaced ParB complexes. Interestingly,in S. coelicolor the Z-rings and, consequently, septa are alsonot mispositioned in the ${Delta}$ smc mutant—measurements did notreveal a statistically significant difference in spacing betweenZ-rings and septa compared with that in the wild type (datanot shown). In S. coelicolor, ParA provides a scaffolding forthe proper distribution of ParB complexes and controls the synchronizedsegregation of several dozen chromosomes, possibly mediatinga segregation and septation checkpoint. On the basis of ourcurrent and previous results (21), we suggest that the formationof the ParA scaffolding precedes chromosome condensation andsegregation events, and therefore, deletion of the smc genein the ${Delta}$ parA mutant does not significantly increase the numberof anucleate prespores. In contrast to Streptomyces, the deletionof B. subtilis smc unmasked a role for Soj in chromosome segregation;a soj-null mutation dramatically enhanced the production ofanucleate cells in an smc-null mutant (25). In S. coelicolor,ParA probably promotes ParB binding to parS sites, possiblyby enhancing oligomerization (21), while SMC-dependent chromosomecondensation may help in clustering ParB binding sites.

Sequential stages of aerial hyphal development with respect to the replication, condensation, and segregation of chromosomes. In rod-shaped bacteria, such as B. subtilis and E. coli, SMCprotein (MukB in E. coli) is present within defined bipolarfoci, one in each cell half, while the replication machineryis located mid-cell. Newly replicated DNA, which moves fromthe central part of the cell toward opposite poles, is probablycondensed within SMC centers. In contrast, in vegetative aswell as in young and short aerial hyphae of Streptomyces, chromosomereplication is not accompanied by condensation: the chromosomesremain uncondensed until the aerial hyphae are converted intoprespores. The rapid growth of the aerial hyphae requires intensivechromosome replication, and our results suggest that at thisstage of growth there is no chromosome condensation machinery(we could observe separate zones of intensive replication andcondensomes). A decrease in replication intensity and an increasein the formation of condensomes was accompanied by the formationof segregation complexes and could be observed before Z-ringand septum formation (Fig. 6). Thus, in Streptomyces, all threeprocesses, chromosome replication, condensation, and segregation,only partially overlap during aerial hyphal maturation, andthey all precede Z-ring formation (Fig. 7).

View larger version (21K):
[in this window]
[in a new window]

FIG. 7. Sequential stages of aerial hypha development with respect to replication, condensation, and segregation of chromosomes.

Addendum. In related work, Dedrick et al. (5a) have obtained similar resultsfor the segregation phenotype of an smc mutant and SMC localization.

ACKNOWLEDGMENTS

We thank Rebekah M. Dedrick and Joseph R. McCormick for communicatingtheir results to us prior to publication and for S. coelicolorstrain RMD2 expressing SMC-EGFP. We are grateful to Keith Chaterfor his critical reading of the manuscript.

This work was supported by the Ministry of Science and HigherEducation (grant no. N301 2035 33). A.K. and J.C.-Z. gratefullyacknowledge financial support received from the Foundation forPolish Science (Start and MISTRZ Programmes).

FOOTNOTES

* Corresponding author. Mailing address: Faculty of Biotechnology, University of Wroc $l$ aw, Tamka 2, 50-137 Wroc $l$ aw, Poland. Phone: 48 71 3752502. Fax: 48 71 3752608. E-mail: zakrzewska{at}protein.pl

Published ahead of print on 17 October 2008.

Supplemental material for this article may be found at http://jb.asm.org/.

REFERENCES

Top