|Home | About | Journals | Submit | Contact Us | Français|
Acidic phospholipids such as cardiolipin (CL) have been shown to modulate Mycobacterium tuberculosis (Mtb) DnaA interactions with ATP. In the present study, using nonyl acridine orange fluorescent dye we localized CL-enriched regions to midcell septa and poles of actively dividing cells. We also found that CL-enriched regions were not visualized in cells defective for septa formation as a consequence of altered FtsZ levels. Using Mtb cultures synchronized for DNA replication we show that CL localization could be used as a marker for cell division and cell cycle progression. Finally, we show that the localization pattern of the DnaA-green fluorescent fusion protein is similar to CL. Our results suggest that DnaA colocalizes with CL during cell cycle progression.
Mycobacterium tuberculosis (Mtb) is a slow-growing pathogenic bacterium and the causative agent of the infectious disease tuberculosis. This bacterium has a doubling time of ~ 24 hours, and the pathogen’s efficiency as an infectious agent relies in part on its ability to shift from an active growth state to a non-replicative persistent state within the host. Mtb in this persistent state maintains metabolic activity 1 and presumably, favorable conditions resulting in reactivation of the bacterium can lead to active infection 2. Such a shift between the two growth states requires tight regulation of cell cycle mechanisms; of particular importance are mechanisms governing initiation of DNA replication.
DnaA protein is the initiator of DNA replication and binds to specific nucleotide sequences within the chromosomal origin of replication or oriC 3–5. Studies with Escherichia coli DnaA revealed that it binds ATP and ADP with equal affinity, and hydrolyzes ATP. The hydrolytic product ADP associates tightly with DnaA and the ADP-bound form of DnaA is not active for oriC replication; acidic phospholipids promote dissociation of nucleotides and activation of DnaA 6. Binding of ATP-bound form of DnaA to the oriC results in opening of the duplex, subsequent assembly of other components and DNA synthesis 7–10. The DnaA protein of Mtb is essential, has been shown to bind and hydrolyze ATP, and bind to oriC 11–16. Mtb DnaA forms an oligomeric complex on the origin, mediated by its ATPase activity 15 and this is followed by strand melting 17. Mtb DnaA protein associates with membrane lipids 13, and like the E. coli counterpart, its interactions with ATP and ADP are also modulated by membrane lipids 12. However, the mechanism(s) regulating the replication initiation process in Mtb are largely unknown.
Our earlier studies using 10-nonyl acridine orange (NAO), a fluorescent dye, determined that CL-enriched regions are confined to septa and cell poles of actively growing M. smegmatis and Mtb 18. These are nascent growth zones in mycobacteria 18, 19. Independently, we showed that lysX mutant deficient in production of lysinylated phosphatidylglycerol is elongated, defective for cell division and exhibits altered NAO staining pattern 18. We infer from these results that NAO, which stains CL, could be used as a marker for evaluating the status of cell division in mycobacteria. Given the affects of membrane phospholipids on DnaA activity, it is also assumed, but not proven that Mtb DnaA interacts with acidic phospholipids in vivo. The present study is undertaken to address two independent questions: (1) whether NAO staining could be used to mark septa and cell division status of mycobacteria; (2) whether the in vivo localization of DnaA and CL are similar.
E. coli strains were grown in Luria-Bertani (LB) broth or agar media. M. smegmatis MC2155 and M. tuberculosis H37Rv strains were grown in Middlebrook 7H9 broth supplemented with albumin (5 g/L) and dextrose (2 g/L); in addition, Mtb cultures were supplemented with oleic acid (60 μL/L) and catalase (3 mg/L). E. coli recombinant strains were grown with ampicillin (amp, 50 μg/mL) or kanamycin (km, 50 μg/mL) or hygromycin (hyg, 200 μg/mL). M. smegmatis and Mtb strains were grown in media supplemented with 25 μg/mL km or 10 μg/mL hyg. Growth was monitored by reading the optical density at 600 nm. Acetamide, when required was used at a final concentration of 0.2%. M. smegmatis FtsZ-72 is a ftsZ merodiploid strain expressing ftsZ from inducible amidase promoter (Pami) integrated at attB locus 20. Addition of acetamide for 6 h leads to a 6-fold increase in FtsZ levels in this strain. M. smegmatis FZ3-78 (also referred to as ΔftsZ,Pami::ftsZ) is the ftsZ conditional expression strain which requires acetamide for normal cell division and viability 21 and removal of acetamide results in the depletion of intracellular FtsZ levels. Mtb dnaAcos115 is a cold-sensitive dnaA strain 22. In this strain, the native dnaA was deleted and was replaced by an integrated copy of dnaA carrying 3 mutations at codons 73, 151 and 215. The strain exhibits a cold sensitive phenotype and has been characterized 22.
M. smegmatis strains for overproduction and depletion of FtsZ were previously described 21. For expression of gfp-dnaA fusion, the Mtb dnaA gene was amplified using oligonucleotide primers, SK3 (5’-CGGGATCCTTGACCGATGACCCCG-3’) and MVM379 (5’-TGTTCTAGACTACTAGCGCTTGGAGCGCTGAC-3’) and cloned into the pGEM-T-easy vector. A 700-bp gfp coding region was amplified using primers MVM469 (5’-GCGACCAGTACTAAAGGAGAAGAACTTTTCACT-3’) and MVM187 (5’-CGGGATCCCTGCAGGTTGTTGTTTTTGTATAGTTCATCC-3’) and cloned at the 5’ end of the dnaA gene in pGEM-T-easy. The entire gfp:dnaA fragment was released, placed downstream of the Pami in pMG103 (kmr) 23. This construct was used to electro-transform M. smegmatis and Mtb competent cell, and kmr transformants were screened by microscopy for observation of fluorescence (see below). To create strains with gfp-dnaA as the sole source of dnaA, the same construct was also used to swap the functional copy of integrated dnaA (Pami::dnaA, hygr) in the previously constructed M. smegmatis dnaA conditional knockout strain 24. Presences of colonies on km containing agar media plates indicate the swapping of the resident Pami::dnaA (hygr) plasmid with Pami::gfp-dnaA (kmr). The kmr resistant colonies were further purified, confirmed by PCR followed by sequencing. Exponential culture of M. smegmatis ΔdnaA, Pami::gfp-dnaA strain was grown with 0.2% acetamide for various time points and cells examined by fluorescence microscopy (see below).
Use of NAO for staining CL enriched regions of Mtb was previously described 18. Briefly, NAO was added during growth of cultures for either 1 h (for M. smegmatis) or 24 h (for Mtb) and cells harvested for microscopy. M. smegmatis cells were pelleted and suspended in fresh media prior to microscopy. Mtb pellets were suspended in 4% para-formaldehyde and stored at 4ºC for 24 h before imaging. A Nikon Eclipse E600 microscope with an attached CoolSnap ES CCD camera (Photometrics) and high-pressure mercury lamp (Nikon) was used. A Nikon B2A filter set (Ex450-490/Em515) was used for GFP. NAO-stained cells were imaged using a custom NAO filter set (Ex450-490/Em645, Chroma Technology) for red fluorescence and the Nikon B2A filter set for green fluorescence. Images were analyzed using MetaMorph 6.2 software (Universal Imaging Corporation) and optimized using Adobe Photoshop 7.0.
Probability of data correlation was calculated using the Chi square test and P values of comparisons were calculated using Student’s T test.
To begin testing whether NAO staining could be used as a marker of cell division, we stained M. smegmatis and Mtb cells at different stages of growth with NAO. It is known that NAO binds to CL in membranes of mitochondria at 2: 1 ratio and creates a stacked pair that is dimer-like. This dimerization results in a shift in absorbance from 495nm to 474nm, which is accompanied by a shift in the emission from green (525nm) to red (640nm). Thus, binding of NAO to CL, but not to other acidic phospholipids, promotes a green to red shift in fluorescence [reviewed in 18]. Consistent with earlier published data, we found NAO stained regions at septa and poles of exponentially growing M. smegmatis (Fig. 1) and Mtb cells (data not shown) 18. However, the stationary phase cultures showed poor staining (Fig. 1). Next, to understand the correlation between NAO enriched regions and cell division, we stained M. smegmatis FtsZ-72 cells grown with 0.2% acetamide for 6 and 18 h and visualized by microscopy. FtsZ protein is the initiator of cell division and localizes at midcell sites as a ring called the Z-ring 25; overproduction or conditional depletion of FtsZ levels interfere with Z-ring assembly and inhibit normal septa formation 21. Growth of M. smegmatis FtsZ-72 in the presence of 0.2% acetamide for 6 h (Fig. 2A, panels iii and iv) led to increased cell length and diminished CL staining at midcell sites and poles as compared with the uninduced cells (Fig. 2A, panels iii and iv with i and ii). Following 18 hrs of induction (Fig. 2A, panels v and vi) cell length was further increased and CL staining was nearly absent; interestingly, most cells also showed defects in cell shape. CL staining was not restored when acetamide was removed and cells were grown in acetamide-free medium for 8 hrs (Fig. 2A, panels vii and viii). Depletion of FtsZ also led to cell elongation and a reduction in CL staining at midcell and poles relative to undepleted cells (Fig. 2B, compare panels iii and iv with i and ii). After 18 hrs (Fig. 2B, panels v and vi) there was an apparent lack of CL staining and cells showed extensive elongation and presence of branches and bud-like structures. Addition of acetamide to these cultures did not recover the CL staining even after 8 hrs growth (Fig. 2B, panels vii and viii). Together these results support a notion that conditions that interfere with the formation of septa would interfere with CL localization or enrichment.
In the above experiments CL-enriched regions were visualized when cell division was blocked by either the overproduction or lack of FtsZ protein. Next we visualized CL-enriched regions in synchronously replicating Mtb dnaAcos115 cells. The dnaAcos115 strain is defective for initiating new rounds of DNA replication at 30°C, a non-permissive temperature, but is proficient at initiating new rounds of active replication and cell division upon shift to 37°C, a permissive temperature 22. Mtb wild-type cells continue to replicate at both temperatures. Accordingly, wild-type and dnaAcos115 cells that were incubated at the non-permissive temperature, and at various intervals following a shift to permissive temperature, were stained with NAO. As can be seen, in both cases we observed staining at midcell sites, poles and also at quarter cell positions (Fig. 3A), although the proportions of these structures varied (Fig. 3B).
Actively growing wild-type cells at 37ºC showed more midcell staining as compared to dnaAcos115 (Fig. 3B). After 30 hrs at 30ºC, we found similar proportion of staining at midcell sites and poles in dnaAcos115 whereas relatively higher proportion of midcell staining relative to quarter position in wild-type. The observed staining pattern of wild-type cells is consistent with a notion that these cells are dividing. NAO staining was found to be diminished at 36 hrs post-shift of wild-type cells, presumably due to a decrease in growth rate of the culture as cells enter stationary phase. Unlike the wild-type cells, we found a sharp increase in midcell staining at 12 hrs and a further burst in midcell and quarter position staining at 24 hrs followed by a decrease at 36 h in dnaAcos115 (see Fig. 3B). A significant proportion of dnaAcos115 cells showed an increase in cell length at 12 and 24 hrs relative to wild-type cells (see Fig. 3A, compare wild-type images with dnaAcos115). The observed rapid burst of staining at midcell and quarter position of dnaAcos115 supports a notion that the dnaAcos115 cells are aligned and are synchronized for replication and possibly cell division.
As reviewed, membrane phospholipids, including CL, promote dissociation of bound nucleotides from DnaA 6, 12, 26, 27. These results implied but did not prove that DnaA protein associates with phospholipids. Results presented in the earlier section and elsewhere based on NAO staining revealed that CL-enriched regions are confined to midcell sites and poles 18. Thus DnaA association with CL implies that DnaA localization and CL-enriched regions overlap. To evaluate this possibility, we created a M. smegmatis strain expressing Pami::gfp-dnaA and visualized DnaA structures by fluorescent microscopy. As can be seen, GFP-DnaA was localized as distinct foci at midcell, quarter cell, and cell poles (Fig. 4A). Note this localization pattern overlaps with CL (Fig. 1). Cell length measurements revealed that 3–4 μm long cells had midcell localizations, whereas 1–2 μm and 5–6 μm cells had mostly polar localizations (Fig. 4B); the calculated Chi square value of 6.77 for these data indicates that a relationship between cell length and DnaA localization is probable. Next we wished to visualize DnaA fluorescent protein in cells stained with NAO. However, because of the technical difficulties associated with the expression and visualization of DnaA-blue fluorescent protein, we could not colocalize DnaA and CL in the same cell. Nonetheless, the similar localization patterns of DnaA and CL tend to suggest that they interact and possibly colocalize.
Next, we asked whether GFP-DnaA could serve as the sole source of DnaA. If GFP-DnaA fusion is functional, then it is likely that the foci we observed are its native localizations and are indeed related to its functional activity within the cell. With this in mind, we swapped the Pami::dnaA from a M. smegmatis dnaA conditional expression strain with Pami::gfp-dnaA; thus providing the fusion protein as the sole source of DnaA in these cells. Analysis of viability data revealed a modest reduction in viability (Fig. 5, p=0.0138 compared with wild-type; for merdodiploid p=0.987) indicating that the M. smegmatis GFP-DnaA is functional and can serve as the sole-source for DnaA. However, we did not observe any distinct GFP:DnaA localization in this strain (data not shown).
We showed that CL-enriched regions are located at midcell sites and poles of actively dividing cells; that CL-enriched regions could not be detected in cells defective for formation of septa owing to overproduction or depletion of FtsZ. CL is enriched at cell septa during cell division 28, 29 and it has been shown that CL clustering is promoted by increased cell wall curvature that results from nascent septal synthesis 30. Modulation of FtsZ levels has been shown to result in cell division arrest in mycobacteria 21. Presumably, CL clustering failed to occur in the absence of septa formation under altered FtsZ levels and is likely responsible for the lack of NAO staining (Fig. 2).
Our studies with Mtb dnaAcos115 cells revealed that NAO could be used to mark septation and cell cycle progression events. For example, we found that, while the wild-type Mtb strain showed a random distribution of staining patterns at all time points, the dnaAcos115 strain showed a significant increase in staining of both midcell and quarter-cell positions at 12 and 24 hrs following return from nonpermissive to the permissive temperature. We can deduce from these results that NAO can be used as an indicator of cell cycle events in synchronous cultures. We also found that DnaA protein localized as distinct foci at midcell, cell poles and possibly at quarter cell positions, much like that seen in B. subtilis 31 and E. coli 32. Quarter-cell staining corresponds to presumptive daughter cell positions and it is known that under certain growth conditions FtsZ-structures are visualized at these positions 19–21. Thus, DNA and CL localization at quarter-cell positions imply that they associate with these regions as soon as the nascent septal synthesis is initiated. It should be noted that one published report indicates that DnaA forms filaments, unlike the localization pattern seen here 33. The localization patterns of DnaA and CL indicate that they interact and possibly colocalize. NAO staining of synchronous populations of cells could give insights into the location of CL-enriched phospholipid domains as a function of cell cycle progression, and in turn, help to shed light on the coordinated positioning of DnaA. DnaA and CL were observed to localize at similar regions of the cell membrane, although further studies are needed to show colocalization and provide some insight as to when their interaction might occur. Together these developments may lead to a greater understanding of cell cycle events, especially those events surrounding the ability of Mtb to enter the latent state and subsequently reactivate.
This work was supported by NIH grants AI84734 and AI73966 (MM) and AI48417 (MR).
Competing interests: The authors have no conflicts of interest to declare.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.