Introduction

During the infectious process, M. tuberculosis faces many different stressful environments; therefore, a rapid adaptation to changing conditions is key for this microorganism. Among the different strategies that can be employed by the bacterium to survive and adapt itself to a hostile environment, specific transcriptional regulation mediated by sigma factors plays a relevant role. Sigma (σ) factors are small interchangeable subunits of the RNA polymerase (RNAP) holoenzyme required to initiate transcription and determine the promoter specificity of the enzyme through the recognition of specific −35 and −10 consensus sequences. They can undergo different regulatory mechanisms at transcriptional, post-transcriptional, translational and post-translational levels1. Post-translational regulation is usually due to anti-σ factors, which are proteins that specifically bind to σ factors preventing their interaction with the RNA core enzyme. Specific environmental signals can then cause disruption of such an interaction, leading to the release of the σ factor2. The genome of M. tuberculosis encodes 13 σ factors3 that represent the highest number of σ factors per Mb among obligate pathogens4.

σE is an alternative σ factor of M. tuberculosis belonging to the extra cellular function (ECF) category whose role is of prime importance for M. tuberculosis physiology and virulence5,6,7,8,9,10. In a previous work we have shown that σE-mediated transcription is inhibited upon interaction with Rv1222, a protein encoded by the gene immediately downstream of sigE11. Amino acid sequence analysis of Rv1222 revealed a HXXXCXXC motif that is typical of the zinc-associated anti-sigma factor (ZAS) family12,13. Additionally, σE and Rv1222 interaction was shown to be redox dependent, and σE could also be released from Rv1222 as a result of a PknB-dependent phosphorylation of Rv1222 under surface stress conditions14. Taken together, all these observations led to the conclusion that Rv1222 is a regulator of sigma E factor and therefore was named RseA.

However, in a recent paper Rudra and colleagues have called into question the role of Rv1222 as an anti-σE factor15. They found that the interaction between σE and Rv1222 does not inhibit open-complex formation mediated by RNAP. Rv1222 was shown to interact with M. tuberculosis RNAP core enzyme but not DNA polymerase and bind non-specifically to DNA through its C-terminal region. In in vitro transcription experiments Rv1222 was able to inhibit transcription regardless the source of RNAP, for inhibition occurred by using holo RNAP or core RNAP coming either from M. tuberculosis, Escherichia coli or Bacillus subtilis. Additionally, they observed a growth rate reduction in both Mycobacterium smegmatis and E. coli upon inducible expression of Rv1222 that reversed when the C-terminal region of Rv1222 was removed. Such a growth defect was accompanied with a reduction of the amount of total RNA. Based on these results, they proposed a novel mechanism of transcriptional regulation according to which Rv1222 inhibits transcription by anchoring RNA polymerase onto DNA and slowing down its movement along DNA during elongation.

In this paper, we have performed a series of experiments using a genetic approach to better characterize the role of Rv1222 in M. tuberculosis physiology.

Experimental Procedures

Bacterial strains, media and growth conditions

E. coli strain DH5α was grown in Luria broth (Difco) at 37 °C. When required, antibiotics were added at the following concentrations: kanamycin 50 µg/ml, hygromycin 150 μg/ml.

M. tuberculosis H37Rv was grown in either Middlebrook 7H9 liquid medium or Middlebrook 7H10 solid medium (Difco) supplemented with 0.05% Tween 80 and ADN (2% glucose, 5% bovine serum albumin, 0.85% NaCl). M. tuberculosis liquid cultures were grown in roller bottles at 37 °C. Plates were incubated at 37 °C in sealed plastic bags. When required, antibiotics were added at the following concentrations: kanamycin 20 µg/ml; streptomycin 20 μg/ml; and hygromycin 50 µg/ml.

All the strains tested in this work are listed in Table 1.

Table 1 List of M. tuberculosis strains used in this work.
Full size table

M. tuberculosis was handled and cultivated in a biosafety level 3 (BL3) laboratory.

DNA manipulation

All recombinant DNA techniques were performed according to standard procedures using E. coli DH5α as the initial host. DNA restriction and modifying enzymes were obtained from New England Biolabs and used according to the manufacturer’s recommendations. Primers used are shown in Table S1.

RNA extraction and quantitative RT-PCR

RNA extraction and quantitative reverse transcription real-time PCR (RT-PCR) were performed using Sybr Green Master Mix (Applied Biosystems) as previously described16. Results were normalized to the amount of sigA mRNA17. RNA samples that had not been reverse transcribed were included in all experiments to exclude significant DNA contamination. For each sample, melting curves were used to confirm the purity of the amplification products. Experiments were performed at least twice, starting from independent biological samples. Primers used for quantitative real-time PCR are described in Table 2.

Table 2 Oligonucleotides used for quantitative RT-PCR assays.
Full size table

Construction of a ΔsigE_Δrv1222 null mutant in Mycobacterium tuberculosis

An unmarked ΔsigE_Δrv1222 deletion mutant was constructed in M. tuberculosis according to a published method18. Two DNA regions, one upstream sigE and one downstream rv1222 were amplified by PCR and sequentially cloned into p1NIL as ScaI/HindIII and HindIII/PacI fragments respectively (Fig. 1A). The rv1221 upstream region (1012 bp) was amplified by RP1413 and RP1414 while the downstream region (1037 bp) was amplified by RP1415 and RP1416. A lacZ-sacB-hyg cassette from pGOAL19 was then introduced as a PacI fragment in the resulting vector to obtain the final suicide plasmid pSA23.1 that was electroporated into M. tuberculosis19. Transformants were selected on plates containing both kanamycin and hygromycin. The occurrence of single crossover was confirmed by PCR (data not shown). One mutant with the correct integration of pSA23.1 was grown in the absence of any drug to allow a second homologous recombination20. Recombinants were isolated as white colonies on plates containing sucrose and X-gal and the occurrence of a double crossover leading to sigE and rv1222 deletion was confirmed by PCR screening with the couple of primers RP1520/RP1521 (couple A) and RP1522/RP1523 (couple B) (Fig. 1B). These primer couples amplify two DNA regions whose length is 2552 bp and 2562 bp respectively in the wt strain, and 1247 bp and 1257 bp respectively in the correct mutant. One strain with the proper chromosomal structure was named TB340.

Figure 1
figure 1

Construction of the ΔsigE_Δrv1222 mutant in M. tuberculosis. (A) Schematic representation of the sigE-rv1222 chromosomal location, and its structure after the correct deletion. Arrows represent primers used for PCR to confirm the deletion. The black lines above and below the chromosomal maps represent the amplification products predicted for primers a/b ((RP1520/RP1521, couple A) or primers c/d (RP1522/RP1523, couple B) in the wt and mutant strain. (B) Agarose gel of PCR products obtained amplifying the genomic DNA of H37Rv and two putative mutants with primer couples A and B. One of the recombinant strain that gave products of the expected size was named TB340. M: 1Kb ladder (NEB). Uncropped image of the gel is presented in Supplementary Fig. S1.

Full size image

Complementation of M. tuberculosis TB340

To complement TB340 with sigE, a 1309 bp fragment containing the whole sigE gene and 535 bp upstream with the sigE promoter11 was amplified by RP1917 and RP1603, and then cloned into the integrative vector able to stably integrate into the single phage attB site in the M. tuberculosis genome21. The resulting plasmid, named pFRA250, was then electroporated into TB340 and transformants were selected on Km plates. The resulting strain was named TB514. To complement TB340 with both sigE and rv1222, a 1488 bp fragment containing the chromosomal region that includes the sigE promoter, sigE and rv1222 was amplified by RP1917 and RP1918 and then cloned into pMV306. The resulting plasmid was named pFRA242 and was introduced into TB340 by electroporation. Transformants were selected on Km plates. The resulting strain was named TB508.

Construction of M. tuberculosis strains over-expressing rv1222

rv1222 was amplified by primers RP1942bis and RP1943 and then cloned into the replicative plasmid pMV261 under the Phsp60 transcriptional control. The resulting plasmid, named pLCM1 was electroporated into H37Rv and TB340 to obtain TB515 and TB516 respectively.

Results

Effect of Rv1222 on M. tuberculosis growth rate

Rudra and coll15 proposed that overexpression of Rv1222 leads to a general decrease of RNA transcription and a consequent reduced growth rate, therefore hypothesizing that deletion of this gene (constitutively expressed in physiological conditions) would result in an increase of M. tuberculosis growth rate. To challenge this hypothesis we compared the growth of H37Rv, TB340 (missing both rv1222 and sigE) and TB514 (TB340 complemented with a single copy of sigE, thus missing only rv1222). As clearly shown in Fig. 2A, H37Rv and TB514 (devoid of rv1222) grew with the same rate, whereas the double mutant grew slightly slower confirming the importance of σE on the growth of M. tuberculosis9. The same data could be observed when bacteria were grown on solid media (Fig. 2B). The deletion of rv1222 thus did not lead to an increase of M. tuberculosis growth rate. Subsequently, we analyzed whether rv1222 overexpression would decrease M. tuberculosis growth rate, as already observed for M. smegmatis and E. coli by Rudra and coll.15. For that purpose, we compared the growth profile of H37Rv, ST28 (single sigE mutant) and TB515 (an H37Rv derivative where rv1222 expression was sevenfold higher than in wild type, see below). As shown in Fig. 3A, overexpression of rv1222 indeed slightly decreased M. tuberculosis growth rate whose profile was comparable to that of ST28. This could be explained by two different reasons: either (i) a decreased RNA transcription rate which would lead to a decreased growth rate, as stated by Rudra and coll.15, or (ii) a reduced activity of σE due to the inhibitory action on σE carried out by an increased amount of Rv1222 that consequently would implicate a growth rate reduction as observed for the sigE mutant. To discriminate between these two hypotheses, we overexpressed Rv1222 in TB340, which is devoid of both sigE and rv1222 (TB516). If overexpression of Rv1222 directs a reduction of the bacterial growth rate, we would expect an additional slowdown in the growth of TB516. However, as shown in Fig. 3B, Rv1222 overexpression in this strain had no effect on its growth, suggesting that the sole consequence of overexpressing Rv1222 is to inhibit σE activity, leading consequently to a slightly reduced growth rate.

Figure 2
figure 2

Growth of M. tuberculosis mutant strains in liquid and solid medium. (A) Growth curves in liquid medium. Bacteria were grown in 7H9 ADC at 37 °C in rolling bottles; OD540 was recorded at regular intervals up to 96 hrs; results represent the average of three independent experiments; (B) Growth in solid medium. Bacteria were inoculated in Middlebrook 7H10 plates and incubated at 37 °C for 21 days.

Full size image
Figure 3
figure 3

Growth curve profiles of M. tuberculosis overexpressing Rv1222. Bacteria were grown in 7H9 ADC at 37 °C in rolling bottles; OD540 was recorded at regular intervals up to 96 hrs. Results represent the average of three independent experiments.

Full size image

Deletion of rv1222 causes specific induction of genes whose expression is controlled by σE

If Rv1222 acted as a σE-specific anti-sigma factor, we would expect to detect an upregulation of those genes whose expression is regulated by σE when Rv1222 is absent. To investigate this hypothesis, we focused our attention on sigB and pspA. While sigB basal level of expression is almost totally σE dependent, pspA expression is known to be σE dependent only under stress conditions activating σE 5,9. Indeed, in a genetic background missing σE (TB340 and ST28) the expression of sigB and pspA was repressed compared to the wild type, of about 10 times and 2 times respectively (Fig. 4). As expected, when the single mutant ST28 was complemented by reintroducing a copy of sigE in its genome, sigB and pspA expression was totally restored. When the double mutant TB340 was complemented by reintroducing a single copy of both sigE and rv1222 on the genome, we obtained a partial but nevertheless clear and significant complementation of the phenotype. However, when this mutant was complemented with the sole sigE, both sigB and pspA were expressed at a higher level than the wild-type, suggesting that in the absence of Rv1222 σE was more active, despite of a 50% reduction in the amount of its transcript levels (due to partial complementation, Fig. 4). To verify whether these results were not due to an artifact, we also measured the expression of pimA (rv2610c), an essential gene required for the biosynthesis of phosphatidyl-myoinositol mannosides22 whose regulation is not controlled by σE. As shown in Fig. 4, pimA expression level did not undergo any significant variation in all of the genotypic backgrounds tested. Taken together these data indicate that the absence of Rv1222 specifically increases the activity of σE.

Figure 4
figure 4

Relative amounts of mRNA level in different H37Rv mutants. The values are expressed as the ratio between the number of cDNA copies detected by quantitative RT–PCR in samples obtained from exponentially growing cultures of the different mutants and their parental strain H37Rv. Data were normalized to the level of sigA cDNA that represented the internal invariant control. The reported values derive from two independent experiments. *P < 0.05 **P < 0.01 versus wt (Student’s t-test).

Full size image

Overexpression of Rv1222 specifically decreases expression of σE -dependent genes

To further confirm that Rv1222 specifically modulates σE activity, we compared the expression levels of a panel of genes in H37Rv and TB515 (H37Rv overexpressing rv1222). As shown in Fig. 5, rv1222 was induced sevenfold in TB515 compared to the wild type. The expression of sigH, sigL, and pimA, whose transcription does not depend on σE 23,24, was not affected by rv1222 overexpression, whereas we could observe a reduction of sigB mRNA level, consistent with the role of Rv1222 as an anti-σE factor, as the upregulation of rv1222 would reduce the activity of σE that is responsible for the basal transcriptional level of sigB.

Figure 5
figure 5

Effects of rv1222 over-expression on relative amount of mRNA levels of different genes. Values are expressed as a ratio between the number of cDNA copies detected by quantitative RT–PCR in samples obtained from exponentially growing TB515 (overexpressing rv1222) and its parental strain H37Rv. Data were normalized to the level of sigA cDNA, that represented the internal invariant control. The reported values derive from two independent experiments. **P < 0.01 versus wt (Student’s t-test).

Full size image

Discussion

Rudra and coll.15 recently challenged the hypothesis that Rv1222 represents a specific anti-σE factor and proposed a different mechanism of action through a series of in vitro experiments, according to which Rv1222 binds to both RNAP and DNA and reduces transcription rate. To corroborate this hypothesis the authors showed that overexpression of this protein in E. coli and M. smegmatis led to a reduction of total mRNA production and consequently of their growth rate, hence the proposal that the deletion of rv1222 might increase mycobacterial growth rate. In this manuscript, we described a series of genetic and physiologic experiments challenging this hypothesis.

Neither overexpression nor deletion of rv1222 have any role in modulating M. tuberculosis growth rate

Using different M. tuberculosis mutants, we first showed that an M. tuberculosis strain devoid of rv1222 does not exhibit an increase in growth rate. Then we showed that a sevenfold increase in rv1222 expression slowed down M. tuberculosis growth only in a genetic background where active σE was present. In fact, such a growth profile was similar to that observed for a sigE mutant, suggesting that such a decrease relied on an inhibitory action on σE activity caused by an increased amount of anti-sigma rather than on a direct, σE-independent, activity of Rv1222 on RNAP. This was also confirmed by RT-PCR experiments showing that Rv1222 overexpression led to a repression of sigB (whose expression is totally controlled by σE)9, but not of pimA, sigL or sigH, whose expression does not depend on σE 22,23,24. We were not able to explain the apparent discrepancy of our data with those shown by Rudra and coll., however in their experiments they used expression systems based on acetamide- or IPTG- inducible promoters (in M. smegmatis and E. coli respectively). These systems are known to produce large amounts of proteins25,26. To perform our experiments, we preferred to use a promoter that expresses the gene at “more physiological” levels (at the best of our knowledge no data reporting rv1222 inductions higher than 2.2 folds have been published). Perhaps the amount of Rv1222 obtained upon induction by Rudra and coll.15 was far above the physiological levels thus resulting in non-specific toxicity of the protein. Unfortunately, they did not report the fold induction obtained in their experiment to address this hypothesis.

Deletion or overexpression of rv1222 specifically modulate the expression of genes under σE-transcriptional regulation

As already mentioned, Rudra and coll.15 hypothesized that Rv1222 down-modulate transcription in a non-specific manner. If this were true, its overexpression or deletion of its structural gene should affect most of the M. tuberculosis gene transcription levels.

However, after overexpressing Rv1222 we could observe transcriptional repression of the σE-dependent gene sigB, but not of the σE-independent genes sigL, sigH, and pimA. Additionally, partial complementation of the repression of the σE-dependent genes pspA and sigB was obtained in the double sigE-rv1222 mutant upon reintroduction of both sigE and rv1222, whereas the reintroduction of the sole sigE in the double mutant led to increased expression of pspA and sigB, despite a reduced amount of sigE mRNA detected in this genetic background as compared to the wild-type. On the other hand, the mRNA level of the sigE independent gene pimA remained the same in all the tested strains. These observations suggested that Rv1222 is required to modulate σE activity specifically. The reduction of sigE expression in a genetic background devoid of rv1222 could be explained by the presence of some still unknown feedback control mechanisms able to down-modulate its expression to compensate the increased activity of σE caused by the loss of post-translational control.

Taken together these data strongly support the role of Rv1222 as an anti-sigma factor regulating σE and are in contrast with those published by Rudra and coll. These authors showed that Rv1222 couldn’t inhibit open complex formation in an EMSA assay. We noted however that a control without σE was lacking in their experiments in order to ensure that the shift observed was not due to non-specific binding of RNA polymerase to the probed DNA fragment. In addition, we also noted that these authors used high concentrations of RNA polymerase in their in vitro experiments, a condition that may well induce non-specific interactions of this protein with DNA. There is no doubt however, that Rv1222 inhibits in vitro transcription, regardless of what sigma factor is used, because it binds to both RNA polymerase and DNA through its C-terminal region, however our data strongly suggest that the situation is completely different in vivo, where Rv1222 is able to inhibit only σE-mediated transcription. In conclusion, our in vivo genetic analyses support a direct role for Rv1222 (RseA) as a specific anti-SigE factor and are in line with our previously published in vitro experiments11,14.