Previous Article | Next Article ![]()
Journal of Bacteriology, July 2004, p. 4441-4448, Vol. 186, No. 14
0021-9193/04/$08.00+0 DOI: 10.1128/JB.186.14.4441-4448.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Patrick Eichenberger, and Richard Losick*
Department of Molecular and Cellular Biology, The Biological Laboratories, Harvard University, Cambridge, Massachusetts 02138
Received 22 January 2004/ Accepted 30 March 2004
|
|
|---|
|
|
|---|
Sporulation takes place in a two-chamber sporangium consisting of a smaller cell called the forespore, which ultimately becomes the spore, and a larger cell called the mother cell, which nurtures the developing spore. Coat proteins are produced in the mother cell and are then deposited around the outer surface of the forespore, which at the time of coat assembly is entirely contained within the mother cell (as a cell within a cell) (14). Coat proteins are produced under the control of the mother-cell-specific RNA polymerase sigma factors
E and
K (14) and the DNA-binding proteins SpoIIID (10) and GerE (13, 23). Recently, additional genes in the mother cell line of gene expression have been identified by gene microarray analysis (5). In ongoing work, we have created fusions of the coding sequence for the green fluorescent protein (GFP) to large numbers of genes identified in this manner. Here we report on the subcellular localization of three previously uncharacterized proteins produced under the control of
E. These proteins are YabP, which is encoded within an operon that also encodes the previously studied YabQ protein (2), YheD, and YutH. We show here that all three of these proteins localize to the assembling coat, but the detailed pattern of subcellular localization is different for each protein and differs from that previously described for other coat proteins. We concluded that protein localization during the process of spore coat assembly is more intricate than previously appreciated.
|
|
|---|
. Plasmids used for single-recombinant integration were isolated from E. coli strain TG1, which allows for isolation of concatenated plasmids, which provide a higher transformation frequency for single-recombination integration. The parent strain for all Bacillus subtilis strains was PY79 (22). Plasmid construction. pCVO119 was synthesized by placing the multiple cloning site of pBluescript into pKL147 (12), which was done as follows. pBluescript was digested with SacI and then treated with T4 DNA polymerase as described previously (19) to produce blunt ends. The vector was then digested with XhoI, and the released 85-bp fragment was purified. pKL147 was digested with EcoRI and treated with T4 DNA polymerase to produce filled-in, blunt ends. The vector was then digested with XhoI and gel purified. The 85-bp fragment was then ligated to the pKL147 backbone to produce pCVO119.
pCVO122 was constructed by amplifying yabP and 200 bp of its promoter sequence by PCR from PY79 genomic DNA with the oligonucleotides YabP5'-3 (GGACGGATCCCGGCCAAAAGCTTGTAACGG) and YabP1-3'-3 (GGACCTCGAGTTTAAACAACTTGCTAAAAAACCC). The resulting DNA fragment was digested with BamHI and XhoI (restriction sites introduced into oligonucleotides are underlined in the sequences) and cloned in frame upstream of the gfp gene in the similarly digested vector pCVO119, forming a yabP-gfp fusion with a five-codon linker.
pCVO124 was constructed by amplifying yabQ by PCR from PY79 genomic DNA with primers YabQ5'-1 (GGACTCTAGAATGACGCTGACGACACAATTC) and YabQ3'-1 (GGACCTCGAGTCTCTTCAAAAAACGTG) by using PY79 genomic DNA as the template. The resulting fragment was digested with XbaI and XhoI and cloned in frame upstream of the gfp gene in pCVO119 digested with SpeI and XhoI to produce a yabQ-gfp fusion with a five-codon linker.
pCVO151 was constructed by amplifying the yabPQ-gfp operon, including 200 bp of upstream, promoter-containing DNA, by PCR from genomic DNA of strain CVO1026 by using primers yabP5'-3 and jd833 (GGCACGGGATCCTTATTTGTATAGTTCATCCATGC). The resulting fragment was digested with BamHI and HindIII and cloned into similarly digested amyE integration vector pDG364 (8).
pCVO152 was constructed by amplifying the yabP-gfp fusion by PCR by using oligonucleotides yabP5'-3 and jd833 from genomic DNA of strain CVO1024. The resulting fragment was digested with BamHI and HindIII and cloned into similarly digested amyE integration vector pDG364, resulting in pCVO152.
pCVO139 was constructed by cloning a BamHI-HindIII fragment from pCVO122 containing the yabP-gfp fusion into similarly digested pDG364.
pCVO197 was constructed by amplifying the coding region of yabP by PCR with oligonucleotide spoVM-RBS (GGACGCATGCGGAGGGGACAAAAATGAATTCATATTATGATCAAAAAGGTTC), which contains the sequence of the ribosome-binding site of the spoVM gene (boldface type) upstream of the yabP start codon, and oligonucleotide GFP3'-3 (GGACAAGCTTTTATTTGTATAGTTCATCCATGC). The resulting fragment was digested with HindIII and SphI and cloned into similarly digested pDR111, an amyE integration vector that contains a copy of the isopropyl-ß-D-1-thiogalactopyranoside (IPTG)-inducible Phyperspank promoter (a gift from D. Rudner), to create a Phyperspank-yabP-gfp transcriptional fusion.
pCVO293 was constructed by amplifying yheD by PCR from PY79 genomic DNA by using primers yheD5'-1 (GACGGATCCCGCTGAAGCAAGACGGACTTTTG) and yheD3'-1 (GACCTCGAGCGAAGGCCACAATGCTTCCG). The resulting DNA fragment was digested with BamHI and XhoI and cloned in frame upstream of the gfp gene into similarly digested pCVO119.
pPE61 was constructed by amplifying 569 bp of the 3' end of yutH by PCR from PY79 genomic DNA by using primers PE691 (CGTATCCCGGGATCCCCTCCGCATGAGCCATTCGATAAA) and PE692 (CGTCTAGCCCTCGAGCCTTGAGCTGCCTTTGCCGAGCCA). The resulting DNA fragment was digested with BamHI and XhoI and cloned in frame upstream into similarly digested pCVO119.
Strain construction.
CVO1024 (yabP-gfp) was produced by transforming PY79 with pCVO122. Transformants were selected for resistance to spectinomycin (Spcr). CVO1026 (yabQ-gfp) was produced by transforming PY79 with pCVO124. Transformants were selected for Spcr. CVO1084 (amyE::yabP-gfp) was produced by transforming PY79 with pCVO152 that had been linearized with KpnI. Transformants were selected for resistance to chloramphenicol (Cmr) and tested for loss of amylase activity. CVO1088 (amyE::yabPQ-gfp) was produced by transforming PY79 with pCVO151 that had been linearized with KpnI. Transformants were selected for Cmr and tested for loss of amylase activity. CVO1111 (
yabP amyE::yabPQ-gfp) was constructed by transforming strain RL2243 (yabP::spc) (6) with genomic DNA of strain CVO1088. Transformants were selected for Cmr and tested for loss of amylase activity. CVO1183 (amyE::yabP-gfp) was constructed by transforming PY79 with linearized pCVO139 and selecting transformants for Cmr. CVO1202 was constructed by transforming RL2244 (yabQ::tet) (6) with genomic DNA of strain CVO1183. Transformants were selected for Cmr and tested for resistance to tetracycline (Tcr) and loss of amylase activity. CVO1219 (amyE::PhyperspankyabP-gfp) was constructed by transforming PY79 with pCVO197 that had been linearized with BglII. Transformants were selected for Spcr and tested for loss of amylase activity. Strain CVO1233 (spoIIGB::erm amyE::PhyperspankyabP-gfp) was produced by transforming RL1061 (spoIIGB::erm) (9) with genomic DNA of CVO1219. Transformants were selected for Spcr and tested for resistance to erythromycin (Emr) and loss of amylase activity. Strain CVO1728 (yheD-gfp) was constructed by transforming PY79 with linearized pCVO293. Transformants were selected for Spcr. Chromosomal DNA from CVO1728 was used to transform RL1397 (spoIVA::neo) and RL322 (cotE::cat) (4) to Spcr to create PE556 and PE557, respectively.
PE479 was obtained by transformation of PY79 with pPE61, followed by selection for Spcr. Chromosomal DNA was prepared from strain PE479 and used to transform RL1397 (spoIVA::neo) and RL322 (cotE::cat) (4) to Spcr to create PE547 and PE500, respectively.
Measuring sporulation efficiency. Sporulation efficiency was determined with a heat resistance assay, as described previously (7). Briefly, cells were grown at 37°C in Difco sporulation medium for 24 to 30 h. The culture was serially diluted 10-fold in T base supplemented with 1 mM MgSO4 six times. Aliquots (100 µl) of the dilutions were plated on Difco sporulation medium agar plates. The dilutions were then heated at 80°C for 20 min, and 100-µl aliquots of the heat-treated dilutions were plated on agar plates containing Difco sporulation medium. The numbers of CFU were determined after overnight incubation at 37°C.
Microscopy. To prepare a culture for microscopy, a strain was grown overnight at 25°C in growth medium to an optical density at 600 nm of 0.5 to 0.7. Sporulation was induced by resuspension of the cells in Sterlini-Mandelstam medium (20) unless indicated otherwise; this time point represented h 0 of sporulation. For strains that carried a gene under the control of an inducible promoter, IPTG (Sigma, St. Louis, Mo.) was added to a concentration of 1 mM to induce gene expression at this time. Alternatively, where indicated below, 1 ml of Difco sporulation medium was inoculated with a single colony of the strain, which was grown for 16 to 18 h at 25 or 37°C (7). Cells were prepared for microscopy by centrifuging a 200-µl aliquot for about 2 min in a microcentrifuge at the times indicated below. The cells were resuspended in 10 µl of phosphate-buffered saline containing 1.5 µg of FM4-64 per ml. Three microliters was placed on a microscope slide and covered with a coverslip that had been treated for approximately 30 s with poly-L-lysine (Sigma). Cells were observed with an Olympus BX60 fluorescence microscope. Typical acquisition times ranged from 400 to 1,000 ms for GFP and were 1,000 ms for FM4-64. Images were captured and cropped by using METAMORPH software. Some additional adjustments were made with Adobe Photoshop.
Deconvolution microscopy was performed as follows. Cells of strain CVO1024 were induced to sporulate and prepared for microscopy as described above. The microscopy was carried out with an inverted DeltaVision microscope. Twenty-five images of optical sections showing fluorescence from GFP were collected with a spacing of 0.1 µm. The images were deconvolved through 15 iterations by using the DeltaVision deconvolution software.
|
|
|---|
YabQ-GFP forms a shell around the forespore. Strains CVO1084 and CVO1088 were treated with the vital membrane stain FM4-64 and examined by fluorescence microscopy. Figure 1B shows the results for strain CVO1088. The YabQ-GFP fusion protein was localized to the septum in sporangia that had undergone asymmetric division (data not shown), to the membrane migrating around the forespore in sporangia that were undergoing engulfment (Fig. 1B, cell on the right indicated by an arrow), and to the outer membrane surrounding the forespore in sporangia in which the forespore was completely pinched off as a free protoplast within the mother cell (Fig. 1B, cell on the left indicated by an arrow). The latter could be recognized by the absence of staining by FM4-64, which is membrane impermeable and hence unable to gain access to the outer forespore membrane when it is topologically isolated from the cytoplasmic membrane. Our results also show that YabQ-GFP is maintained on the spore coat during maturation and release of the mature spore (Fig. 1B, bottom panels). The localization of YabQ was investigated previously by Asai et al. (2), and our findings are in agreement with their findings.
![]() View larger version (64K): [in a new window] |
FIG. 1. Subcellular localization of YabP-GFP and YabQ-GFP. (A) Localization of YabP-GFP. Cells of strain CVO1084 (amyE::yabP-gfp) were induced to sporulate, stained with the membrane dye FM4-64, and prepared for microscopy at the times (in hours) indicated on the left as described in Materials and Methods. YabP-GFP fluorescence is shown in the left panels, and the membrane staining is shown in the right panels. The arrows and arrowheads indicate corresponding cells in adjacent panels that have almost completed engulfment and are in the process of engulfing the forespore, respectively. The asterisks indicate a cell in which engulfment of the forespore is in a very early stage but in which YabP-GFP is nonetheless already concentrated on the sporulation septum. (B) Localization of YabQ-GFP. Cells of strain CVO1088 (amyE::yabPQ-gfp) were induced to sporulate and observed at the times (in hours) indicated on the left. The cells viewed after 3 h of sporulation were additionally stained with the membrane dye FM4-64. The panels on the left show the YabQ-GFP fluorescence. The panels on the right show the membrane dye fluorescence (top) or phase microscopy (bottom) of corresponding cells. The arrows indicate corresponding cells in the left and right panels. YabQ-GFP was concentrated on the outer forespore membrane in all cells examined. (C) Localization of YabP-GFP in the absence of YabQ. Cells of strain CVO1202 (yabQ::tet amyE::yabP-gfp) were induced to sporulate and observed after 3 h. Note the bright staining of the dots and the greatly decreased staining of the outer forespore membrane compared to the staining of YabP-GFP in a wild-type background (as shown in panel A).
|
Deconvolution microscopy of sporulating cells of strain CVO1084, which allowed us to view thin (
0.1-µm) focal planes without the interfering unfocused light from other focal planes, revealed that the dots on the side of the forespore represented a ring that encircled the entire forespore (Fig. 2). YabP-GFP was detected as a single dot at focal planes that represented the region where the cell was in contact with the glass slide (Fig. 2, panel 1). At the focal planes corresponding to the middle portion of the cell (Fig. 2, panels 2 to 5), YabP-GFP staining separated into two dots that gradually spread further apart as the focal plane was moved up through the cell. At the top of the cell YabP-GFP was once again detected as a single dot (Fig. 2, panels 6 to 8).
![]() View larger version (39K): [in a new window] |
FIG. 2. Deconvolution microscopy of sporulating cells producing YabP-GFP. Cells of strain CVO1084 (amyE::yabP-gfp) were prepared for deconvolution microscopy as described in Materials and Methods. Panel 1 shows the focal plane closest to the slide, and each consecutive image is 0.1 µm farther from the slide. Note the increase in the distance between the dots of YabP-GFP fluorescence as the focal plane moved up through the first five images. The last three images show that the distance between the dots decreased until they combined to form a single dot in panel 8, which shows the region of the cell farthest from the slide.
|
Ring formation by YabP does not depend on YabQ. Because yabP and yabQ are in the same operon, it seemed possible that the localization of YabP might depend in whole or in part on YabQ. To investigate this, we introduced the yabP-gfp fusion into a strain with a deletion of the yabQ gene. The localization of YabP-GFP in the resulting strain (CVO1202) was subtly altered. Ring formation by YabP-GFP was unimpaired, but the staining around the entire outer forespore membrane was greatly diminished (Fig. 1C). We concluded that YabP-GFP is recruited to the outer forespore membrane by YabQ but that the assembly of the protein into a ring occurs in a YabQ-independent manner.
The localization pattern of several previously investigated proteins (SpoVM and the SpoIVFA-SpoIVFB-BofA complex) whose production in the mother cell is under the control of
E is known to depend on an as-yet-unidentified
E-controlled gene product(s) (15, 21). To test if the localization of YabP-GFP was similarly dependent on
E, we placed yabP-gfp under the control of the IPTG-inducible promoter Phyperspank (16) so that the protein could be synthesized in a
E-independent manner. In a strain harboring the Phyperspank-yabP-gfp construct (CVO1219), the fusion protein exhibited a normal pattern of localization (Fig. 3, top panels). However, when synthesis of YabP-GFP was induced in a strain lacking
E (CVO1233), YabP-GFP failed to localize in a specific manner (Fig. 3, bottom panels). Instead, the fusion protein exhibited a uniform pattern of fluorescence throughout the mother cell cytoplasm. Thus, at least two gene products influence the distribution of YabP-GFP: YabQ, which is required for the adherence of the fusion protein around the entire outer forespore membrane, and another, unidentified, gene product produced under the control of
E.
![]() View larger version (46K): [in a new window] |
FIG. 3. Subcellular localization of YabP-GFP requires E-directed gene expression. Transcription of yabP-gfp was induced by addition of IPTG in strains carrying the Phyperspank-yabP-gfp translational fusion in backgrounds with and without E activity (strains CVO1219 and CVO1233, respectively). Cells were prepared for microscopy 3 h after the induction of sporulation, as described in the legend to Fig. 1. The YabP-GFP fusion was recruited to the outer forespore membrane normally when it was expressed in the wild-type (WT) background from the inducible promoter (top panels). In the absence of E, YabP was not recruited to the sporulation septum, and the protein was detected throughout the cytoplasm (bottom panels).
|
E-controlled gene yheD. The gene for GFP was joined to the 3' end of the 429-codon yheD open reading frame. (Because a yheD mutation did not exhibit a conspicuous phenotype, we were unable to assess the functionality of the fusion protein.) Plasmid DNA containing the yheD-gfp fusion was integrated into the chromosome by single-reciprocal recombination at yheD, yielding strain CVO1728 in which gfp was fused to a full-length copy of yheD. YheD-GFP displayed a striking localization pattern; the fusion protein was concentrated in four dots that were juxtaposed on opposite sides of the engulfed forespore across the short axis of the sporangium (Fig. 4, top panels). We interpreted these images to indicate that YheD-GFP forms two rings that encircle the forespore. (The two rings were in every case examined in a plane perpendicular to the cytoplasmic membrane, which appeared to rule out the possibility that the two rings were part of a spiral.) In almost all sporangia examined the ring on the mother cell proximal side of the forespore was significantly brighter than the ring nearest the mother cell distal pole. Interestingly, the location of the YheD-GFP rings did not overlap the location of the YabP-GFP ring. Whereas YabP-GFP rings were located at the middle of the forespore, at its widest point, the YheD-GFP rings were offset from the middle. At later stages of sporulation, the YheD-GFP rings disappeared and YheD-GFP was redistributed to form a complete shell around the forespore (Fig. 4, bottom panels).
![]() View larger version (63K): [in a new window] |
FIG. 4. Subcellular localization of YheD-GFP. Cells of strain CVO1728 (yheD-gfp) were induced to sporulate, stained with the membrane dye FM4-64, and prepared for microscopy as described in Materials and Methods. GFP fluorescence is shown in the left panels, and membrane staining is shown in the right panels. Note the distinct subcellular localization pattern of YheD-GFP around the outer forespore membrane at 3 h after induction of sporulation (top panels). Engulfment had been completed in these cells, which greatly decreased the staining of the forespore. At 6.5 h after the induction of sporulation (bottom panels), YheD-GFP was redistributed to form a complete shell around the outer forespore membrane. The arrows indicate corresponding cells in the left and right panels.
|
![]() View larger version (64K): [in a new window] |
FIG. 5. Subcellular localization of YheD-GFP depends on SpoIVA. Cells harboring a yheD-gfp fusion were grown for 18 h in Difco sporulation medium at 37°C, stained with the membrane dye FM4-64, and prepared for microscopy as described in Materials and Methods. YheD-GFP was visualized in a otherwise wild-type derivative of strain PY79 (strain CVO1728), in a spoIVA mutant (PE556), and in a cotE mutant (PE557), as indicated on the left. Representative sporangia are shown. Staining with the membrane dye FM4-64 is shown in the right panels, and GFP fluorescence is shown in the left panels.
|
![]() View larger version (31K): [in a new window] |
FIG. 6. Subcellular localization of YutH-GFP. Cells harboring a yutH yutH-gfp fusion were induced to sporulate, stained with the membrane dye FM4-64, and prepared for microscopy as described in Materials and Methods. (A) Fluorescence micrographs of representative cells producing YutH-GFP after growth for 18 h in Difco sporulation medium at 37°C. YutH-GFP was visualized in an otherwise wild-type derivative of strain PY79 (strain PE479), in a spoIVA mutant (PE547), and in a cotE mutant (PE500), as indicated on the left. Staining with the membrane dye FM4-64 is shown in the right panels, and GFP fluorescence is shown in the left panels. (B) Time course of YutH-GFP localization during sporulation. Cells of strain PE479 were sporulated by suspension in Sterlini-Mandelstam medium at 37°C, and samples were collected at the times indicated and analyzed by fluorescence microscopy. Only representative cells are shown. Staining with the membrane dye FM4-64 is shown in the right panels, and GFP fluorescence is shown in the left panels.
|
E, 2 h after suspension in sporulation medium, a single dot of YutH-GFP fluorescence was observed close to the engulfing forespore. Once engulfment was completed at h 3, two dots were observed along the short axis of the sporangium, suggesting that YutH-GFP had formed a ring. However, in contrast to the YabP-GFP and YheD-GFP rings, which were located at or near the center of the forespore, the YutH ring was at the extreme mother-cell-proximal end of the forespore and did not completely encircle it. By h 4, the ring had become a polar cap on the mother-cell-proximal side of the forespore. By h 5, the cap had spread to encircle most of the forespore. A model summarizing the results of the subcellular localization studies in this investigation is presented in Fig. 7. We concluded that protein localization during sporulation is richer and more intricate than previously recognized. Not only do proteins coalesce into shell-like structures that encase the developing spore, but, as we now see, some proteins (i.e., YabP, YheD and YutH) (Fig. 7) assemble into distinct rings that encircle the forespore during the process of coat assembly. Presumably, these rings contribute to the assembly of the spore coat, but how they do so is not yet apparent. A similar dynamic pattern of localization has been detected for the germination protein GerQ, a spore coat protein that is also dependent on SpoIVA for proper localization. In initial stages of coat formation GerQ is detected as a single focus on the mother cell side of the forespore, and at later stages it is distributed in a homogeneous shell around the forespore (17). A principal challenge for the future is to elucidate the nature of the spatial cues that dictate the characteristic positioning of YabQ, YabP, YheD, YutH, GerQ, and other development-specific proteins during the course of spore coat morphogenesis.
![]() View larger version (22K): [in a new window] |
FIG. 7. Model of localization of YabP, YheD, and YutH during sporulation. The cartoons show the following progressive stages of sporulation (from left to right): septation, engulfment in progress, engulfment completed, and maturing forespore. YabP initially associates with the sporulation septum and then migrates with the mother cell membrane that is engulfing the forespore, where it is concentrated at its leading edge. After engulfment is complete, YabP is concentrated in a ring that surrounds the forespore at its widest point but is also present as a shell around the outer forespore membrane. The latter pattern of localization is dependent on YabQ, as shown in the cartoon depicting the YabP localization in a yabQ mutant cell. At later stages of sporulation YabP is found primarily in the cytosol of the mother cell and occasionally as a small dot on the mother cell side of the forespore. YheD is detected initially in two caps at the poles of the forespore. Ultimately, it forms a complete shell around the forespore. Recruitment of YheD to the outer forespore depends on SpoIVA. YutH is initially detected as a thin cap at the mother cell side of the forespore, which gradually covers the outer forespore membrane as sporulation progresses. Localization of YutH on the outer forespore membrane is dependent on SpoIVA. Proteins are indicated by grey shading, and membranes are indicated by black lines; the intensity of shading indicates the relative amount of protein as detected by fluorescence microscopy.
|
This work was supported by National Institutes of Health National Research Service Award GM20165 to C.V.O., a Swiss National Science Foundation postdoctoral fellowship and a Merck Core Educational Support Program to P.E., and National Institutes of Health grant GM18568 to R.L.
Present address: Department of Pathology, Northwestern University, Chicago, IL 60611. ![]()
|
|
|---|
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»