|Home | About | Journals | Submit | Contact Us | Français|
FtsH is an essential membrane bound protease that degrades integral membrane proteins as well as cytoplasmic proteins. We show that Mycobacterium tuberculosis (Mtb) ftsH expression levels are upregulated upon exposure to agents that produce reactive oxygen and nitrogen intermediates (ROI and RNI) and growth in macrophages. In partial support of this result is our observation that the Mtb merodiploid overexpressing ftsH shows increased resistance to ROI. ftsH transcripts levels are downregulated during stationary phase and starvation. Overexpression of ftsH delayed growth and reduced the viability of Mtb in vitro and ex vivo. Finally, we show that the intracellular levels of FtsZ, an essential cell division protein, are reduced in ftsH overexpressing strain. Together, our results suggest that Mtb FtsH is a stress response protein that promotes the pathogen’s ability to deal with ROI stress and is possibly involved in the regulation of FtsZ levels.
FtsH is a member of AAA-family ATPases and plays an important role in the protein quality control mechanisms. FtsH mediated degradation of misassembled membrane and cytoplasmic proteins is thought to be responsible for its role in the control of cellular functions including the heat shock response (reviewed in 1, 2). FtsH is also required for the proper functioning of σ54 under nitrogen limitation conditions 3. FtsH expression is upregulated in response to heat shock in several bacteria, e.g., Escherichia coli, Bacillus subtilis, Lactococcus lactis, Caulobacter cresentus and Helicobacter pylori 4–7. The FtsH protease is essential for growth in E. coli 8, 9 whereas it is dispensable in B. subtilis and C. cresentus 4, 10. However, ftsH mutants are hypersensitive to heat, salt, and defective for sporulation and possibly cell division 4,10.
Saturated transposon mutagenesis studies indicated that Mtb ftsH (ftsHTB) is an essential gene 11 that encodes a 85 kD protein, which is 15kD larger than the 70 kD E. coli counterpart. The ftsHTB can complement the E. coli ftsH phenotype and degrade heterologous substrates σ32, SecY and bacteriophage λCII protein when expressed in E. coli ftsH null mutant 12. The expression of M. smegmatis ftsH in E. coli results in growth arrest and filamentation 13. In vitro, the FtsHEC efficiently degrades FtsZ of E. coli (FtsZEC) and Mtb (FtsZTB) 14. FtsZTB, like its E. coli counterpart, is an essential cell division protein and initiates the process by localizing at the midcell sites in the form of Z-rings 15, 16. Experiments designed to test whether FtsHEC affects the stability and turnover of FtsZEC in vivo, however, failed to support the in vitro data 17. Related experiments with Mtb have not been carried out. Consequently, it is unknown if ftsHTB activity contributes to the reduction in the intracellular levels of FtsZTB in vivo. Furthermore, it is unknown if ftsHTB expression in vivo is responsive to stress. The current study addresses these issues.
Escherichia coli Top10 was used to propagate recombinant plasmids as described 15, 18. Mtb H37Rv was grown in Middlebrook 7H9 broth supplemented with oleic acid, albumin, dextrose, and sodium chloride (OADC). Transformants were selected in the same medium supplemented with agar containing Hyg (50 μg ml−1) 16. Growth conditions such as hypoxia, starvation and exposure to agents that produce ROI and NO, are essentially as described earlier 18. Mtb cultures were exposed to 500 μM DETA/NO for 16 h, 70 μM menadione or 5 mM H2O2 for 48 h. Growth was monitored by measuring the absorbance at 600 nm and viability by determining the CFU.
The 2.3 kb ftsH coding region was amplified using oligonucleotide primer pairs FtsH8 (5′ – GGAATTCCATATGAACCGGAAAAACGTGACTCG – 3′) and FtsH9 (5′ – CTAGTCTAGACTAT-CAGCCGTGGGCCGGCTTGGTC – 3′) or Fts H-A-XbaI (5′-CTAGTCTAGACTAATGAACCGGAA-AAACGTGACT -3′) and Fts H-A-NdeI (5′-GGAATTCCATATGTCAGCCGTGGGCCGGTTGG -3′). The PCR products were cloned downstream of the amidase promoter in an integrating vector, pJFR19 15 to create the sense plasmid, pACR6, and the antisense plasmid pACR33, respectively. Following confirmation of sequences, Mtb was electrotransformed with these plasmids to create ftsH overexpression strain Mtb-6 and antisense expression strain Mtb-33.
Monocyte derived macrophage cell line THP-1 cells were infected with various Mtb strains and bacterial viability was determined as described previously 15, 19. For some experiments, RNA was isolated from macrophage-grown Mtb 19.
Extraction of total RNA in RNAzol from Mtb cultures exposed to different conditions and synthesis of complementary strand DNA from mRNA specific to 16S rRNA and ftsH using reverse transcription primers and Superscript II reverse transcriptase (Invitrogen) were as described 19. Quantitative real time PCR (Taqman chemistry) was carried out in a BioRad ICycler using Taq DNA polymerase (NEB). The calculated threshold cycle (Ct) value for ftsH was normalized to the Ct value for 16S rRNA and the fold expression was calculated using the formula: Fold change = 2Δ(ΔCt) 20. No RT RNA samples were included as negative controls. Expression data are average from 3 independent RNA preparations, each reverse transcribed and quantitated by real time PCR in triplicate. Real-time PCR conditions: initial activation at 95°C for 3minutes; followed by 45 cycles of denaturation at 95°C for 20 seconds, annealing at 60°C for 30 seconds and extension at 72°C for 30 seconds.
Mtb cellular lysates were prepared by bead-beating actively growing cultures. Approximately 2 μg protein lysate per lane was loaded, resolved by SDS-PAGE and immunoblotting was carried out to detect FtsZTB, FtsHTB and SigATB using the previously described protocols 16, 21. Anti-FtsHEC antibodies were a kind gift from Dr. Teru Ogura, Kumamoto University, Japan and were used at 1:2000 dilution, whereas anti-FtsZTB and anti-SigAEcoli were used at 1: 1000 dilution 15.
Considering the roles of FtsH on the protein quality control mechanisms in other bacteria, we first evaluated by quantitative RT-PCR whether the Mtb ftsH transcript levels relative to 16S rRNA are modulated under the conditions that are relevant for Mtb survival in vivo, i.e. growth in macrophages, exposure to reactive oxygen and nitrogen intermediates, hypoxia and nutrient starvation 22, 23. The ftsH transcript levels were found to be sharply elevated during intramacrophage growth and when exposed to DETA-NO, a nitric oxide (NO) producer, but were downregulated during starvation and stationary phase and remained relatively unchanged during hypoxia (Fig. 1). A small, but consistent, upregulation of ftsH transcript levels was also observed in cultures exposed to menadione and hydrogen peroxide, both of which produce reactive oxygen intermediates (ROI) (Fig. 1). Presumably, altered expression levels of ftsH are necessary for the optimal survival of Mtb under the above growth conditions.
The above results are consistent with a notion that elevated levels of FtsH are required for the optimal survival of Mtb under the conditions that produce NO and ROI. To further evaluate the correlation between altered ftsH expression and ROI and NO stress, we created and characterized merodiploid strains overexpressing ftsH mRNA (Mtb-6) and antisense RNA (Mtb-33) (see below). As can be seen, growth in the presence of acetamide for 72 hrs led to an ~3-fold increase in the FtsH protein levels in Mtb-6 and an ~ 60% decrease in Mtb-33, relative to that of H37Rv (Fig. 2A and inset). Elevated levels of FtsH slowed the growth in broth (Fig. 2B) and reduced viability by 50% (Fig. 2B insert). However, the reduction in FtsH levels did not affect the growth and viability of Mtb (Fig. 2B).
Next, we examined the growth and viability of Mtb ftsH merodiploid strains exposed to select stress conditions, i.e. hydrogen peroxide and menadione. A significant reduction in the viability of Mtb wild type and Mtb-33 was noted at the concentrations of the agents used (Fig. 3) indicating that the activity of FtsH protein is required for survival upon exposure to agents that produce ROI. Mtb-6 strain, producing elevated levels of FtsH (Fig. 2A), showed enhanced viability under the same growth conditions indicating that increased intracellular levels of FtsH promote the ability of Mtb to resist ROI (Fig. 3). Exposure to RNI intermediates and starvation had only a marginal effect on the viability of all three strains (data not shown). These results reveal that Mtb producing elevated levels of FtsH would weather the ROI stress better than that producing the normal or reduced levels of FtsH.
We next examined the consequences of altered ftsH expression on Mtb growth and viability upon infection. As can be seen, Mtb-6 strain expressing elevated levels of ftsH showed reduced viability in THP-1, a human macrophage cell line, as compared to the wild type parent (Fig. 4). The Mtb-33 showed delayed growth during the earlier time points of infection, but reached to the same final levels as the WT indicating a defect in replication. Macrophage growth environment is rather hostile and Mtb likely encounters stresses other than the ones tested in the present study, i.e. acidic pH, limiting nutrients, host secreted cytokines and antimicrobial peptides 24. Presumably, elevated expression of ftsH resulting in excessive levels of FtsH could trigger unregulated proteolytic activity upon infection. Similarly, decreased expression of ftsH resulting in the reduced in vivo pools of FtsH could lead to the accumulation of undesirable proteins. Both scenarios could compromise the ability of Mtb to replicate in macrophages optimally. These results are consistent with a notion that optimal FtsH levels are essential for survival of Mtb in vivo (Fig. 4).
Optimal intracellular levels of the essential cell division protein FtsZ are critical for cell division and viability 16. FtsZ levels are growth-phase dependent 16, 25. To evaluate the correlation, if any, between the decreased growth of ftsH merodiploids and FtsZ levels, we determined the intracellular FtsZ levels in Mtb-6 and Mtb-33 strains by immunoblotting. Since SigA levels are known to remain unchanged under several conditions that have been tested 15, 26, we normalized the data to SigA (Fig. 5). These analyses revealed that FtsZ levels were reduced by approximately 34% in Mtb-6, but remained unaffected in Mtb-33 (Fig. 5). These results, however, do not provide direct evidence that FtsHTB protease degrades FtsZTB protein. Our earlier data showed that the FtsZ levels are similar in broth and macrophage grown bacteria 15, although the ftsH expression is shown to be upregulated during intramacrophage growth (Fig. 1). Thus, it is possible that FtsH could directly regulate the FtsZ levels by its proteolytic activity or indirectly influence its levels by degrading hitherto unidentified proteins that stabilize FtsZTB, thereby facilitating the action of other proteases on FtsZ. Nonetheless, reduction in the FtsZ levels under FtsH overproduction conditions leads to a speculation that proliferation of Mtb depends, in part, on the optimal ratio of FtsH to FtsZ proteins. Tipping that balance by elevating the FtsH levels could influence the intracellular FtsZ pools and compromise growth and viability.
This work was supported by NIH grants, AI48417, AI41406 and AI073966. We thank Dr. Teru Ogura, Kumamoto University, Japan for antibodies to FtsHEC.