Cell Host & Microbe
CarD Tricks and Magic Spots:
Mechanisms of Stringent Control in Mycobacteria
Lynn E. Connolly1,2,3and Jeffery S. Cox1,2,*
1Department of Microbiology and Immunology
2Program in Microbial Pathogenesis and Host Defense
3Department of Medicine, Division of Infectious Diseases
University of California, San Francisco, San Francisco, CA 94143, USA
Global reprogramming of bacterial gene expression in response to nutritional stress, the stringent response,
is well studied in E. coli. Now Stallings et al. report that Mycobacterium tuberculosis employs a different
strategy involving the general transcription factor CarD for growth control and persistence in response to
stresses encountered during infection.
Tuberculosis (TB) continues to be a major
cases and 1.7 million deaths per year
(Dye, 2006). A hallmark of Mycobacterium
tuberculosis pathogenesis is the ability of
bacilli to entera slow-growing or nonrepli-
cative state, leading to a latent infection
nutrients, oxygen, and iron is thought to
restrict bacterial replication during in-
fection (Boshoff and Barry, 2005). For
example, M. tuberculosis bacilli have the
exceptional ability to persist for decades
in vitro under starvation conditions and
can revive upon addition of nutrients.
This remarkable ability of M. tuberculosis
to persist in the face of nutritional and
immune stresses renders tuberculosis
difficult to treat with current antibiotics,
which generally require some bacterial
growth to exert their killing effects. Thus,
elucidating the mechanisms of persis-
tence is of great interest in the field and
critically important to the global control
of this deadly disease.
Our understanding of bacterial res-
ponses to starvation stress has largely
been informed by studies in Escherichia
coli. In response to amino acid depriva-
tion, E. coli cells inhibit expression of sta-
ble RNAs required for protein synthesis,
rRNA and tRNA, and stimulate transcrip-
tion of other operons, including those
required for the biosynthesis and trans-
port of certain amino acids (Srivatsan
latory cascade are the ‘‘magic spot’’ star-
referred to collectively as (p)ppGpp. In
(p)ppGpp synthesis is catalyzed by the
tantly, the transcriptional changes in
response to increased (p)ppGpp are not
mediated by DNA-binding transcription
factors but, rather, by direct alteration of
the stability of the RNA polymerase
(RNAP) complex at regulated promoters.
(p)ppGpp alone is not sufficient to mete
out these changes but works along with
the general transcription factor DksA,
which interacts directly with RNAP to
exert its effects on gene expression.
promoters and amino acid biosynthetic
promoters allow for DksA-modified RNAP
to favor one over the other. Although the
stringent response has been studied
array of stresses activate (p)ppGpp syn-
thesis, including deprivation of phospho-
rous, iron, and fatty acids (Srivatsan and
In a recent study published in Cell, Stal-
lings et al. set out to identify genes whose
transcription was induced in response to
double-strand DNA breaks in the chro-
mosome of Mycobacterium smegmatis, a
nonpathogenic mycobacterium distantly
related to M. tuberculosis. A general tran-
scription factor carD was among the most
highly induced genes under these condi-
tions, and the authors show that carD is
that induce DNA damage, but also in
response to oxidative damage and nutri-
tional limitation. The authors show that
carD is an essential gene and thus use
conditional alleles under the control of
regulated promoters for their studies. As
expected, CarD depletion renders the
bacteria sensitive to DNA damage, oxida-
tive stress, and starvation. Together,
these findings suggest that CarD is a key
player in the general stress response.
The clue that CarD is involved in strin-
studies of cells depleted of the protein.
Stallings et al. show that levels of stable
rRNA and mRNAs encoding ribosomal
protein subunits are strongly induced in
the CarD-depleted strain compared to
the wild-type, even under normal growth
conditions. As in E. coli, rRNA transcrip-
tion is repressed under starvation condi-
tions in wild-type M. smegmatis, but not
in the CarD depletion strain, even though
(p)ppGpp still accumulated in the mutant.
Therefore, CarD appears to control the
stringent response in M. smegmatis in
a fashion analogous to that of DksA in
E. coli, even though these two proteins
share no sequence similarity. Astonish-
ingly, despite the apparent differences
in these proteins, the carD gene from
M. tuberculosis can functionally comple-
ment an E. coli dksA mutant.
Despite the similarity in their function,
the molecular mechanism of CarD likely
differs from that of DksA. Structural and
biochemical studies suggest that DksA
interacts with RNAP in the secondary
channel, the same portal used by the
structurally similar transcriptional elonga-
tion factors GreA/B (Haugen et al., 2008).
DksA is thought to act primarily by affect-
ing specific kinetic steps during initiation
of transcription. CarD, however, likely
binds to a different region of RNAP via
a domain similar to the RNAP-binding
Cell Host & Microbe 6, July 23, 2009 ª2009 Elsevier Inc.
domainofthetranscriptionrepaircoupling Download full-text
factor (TRCF). TRCF functions to remove
stalled RNAP at sites of lesions in the
ences RNAP activity is unknown. Given
the differences in the mode of interaction
with RNAP between these two proteins,
it seems likely that CarD will alter RNAP
activity via a mechanism distinct from
that of DksA.
The authors show that, in M. tubercu-
losis, depletion of CarD leads to bacterial
killing in a mouse model of infection.
Although there is growing evidence that
at least limited replication occurs during
chronic infection and thus any essential
gene may be required for persistence
(Gandotra et al., 2007; Gill et al., 2009),
this result underscores the importance of
stringent control of bacterial growth in
the host (Dahl et al., 2003). In addition,
CarD is vital for bacterial resistance to
oxidative stress and DNA damage, two
further stresses likely encountered during
infection. Although these results may not
be surprising, they highlight the impor-
tance of the general bacterial stress
response during chronic infection and
support the notion that the CarD-RNAP
interaction may be a viable target for ther-
Why is CarD essential in mycobacteria?
E. coli mutants deficient
(p)ppGpp synthesis, or both are viable
in rich media, as are relA mutants of
M. tuberculosis and M. smegmatis (Dahl
et al., 2003; Stallings et al., 2009). One
possible explanation is that the stringent
response is not essential for growth, but
CarD performs an additional, (p)ppGpp-
independent role in mycobacterial cells.
Alternatively, the stringent response itself
may be essential, but mycobacterial
(p)ppGpp synthases other than RelA are
able to support low-level concentrations
survival. Interestingly, in Bacillus subtilis,
which also has CarD, but not DksA, the
carD gene is dispensable for growth (Ko-
bayashi et al., 2003). This suggests that
mycobacteria may be uniquely suscep-
tible to stresses countered by CarD. For
example, in contrast to B. subtilis, which
encodes 10 rRNA operons scattered
throughout the genome, most mycobac-
teria encode only one or two such
operons. The authors suggest that, in
the absence of CarD, stalled RNAP
complexes may accumulate at these few
sites in the genome, effectively blocking
DNA replication and repair. Indeed, this
as DksA and CarD is to remove stalled
transcription complexes that arise as
a result of DNA damage (Trautinger et al.,
prokaryotic genomes show that both
diverse bacterial taxa. Indeed, some
genomes encode both dksA and carD,
supporting the notion that, despite their
functional similarities, the two factors are
not redundant. Organisms ranging from
Actinomycetes to Cyanobacteria encode
only carD homologs, indicating that CarD-
mediated regulation of the stringent re-
sponse is likely widespread in the micro-
bial world. Finally,
including important human pathogens,
lack both dksA and carD homologs
despite encoding (p)ppGpp synthases.
Therefore, other mechanisms to read out
(p)ppGpp await discovery.
Boshoff, H.I., and Barry, C.E., III. (2005). Nat. Rev.
Microbiol. 3, 70–80.
Dahl, J.L., Kraus, C.N., Boshoff, H.I., Doan, B.,
Foley, K., Avarbock, D., Kaplan, G., Mizrahi, V.,
Rubin, H., and Barry, C.E., III. (2003). Proc. Natl.
Acad. Sci. USA 100, 10026–10031.
Dye, C. (2006). Lancet 367, 938–940.
Gandotra, S., Schnappinger, D., Monteleone, M.,
Hillen, W., and Ehrt, S. (2007). Nat. Med. 13,
Gill, W.P., Harik, N.S., Whiddon, M.R., Liao, R.P.,
Mittler, J.E., and Sherman, D.R. (2009). Nat. Med.
Haugen, S.P., Ross, W., and Gourse, R.L. (2008).
Nat. Rev. Microbiol. 6, 507–519.
Kobayashi, K., Ehrlich, S.D., Albertini, A., Amati,
G., Andersen, K.K., Arnaud, M., Asai, K., Ashikaga,
S., Aymerich, S., Bessieres, P., et al. (2003). Proc.
Natl. Acad. Sci. USA 100, 4678–4683.
Srivatsan, A., and Wang, J.D. (2008). Curr. Opin.
Microbiol. 11, 100–105.
Stallings, C.L., Stephanou, N.C., Chu, L., Hochs-
child, A., Nickels, B.E., and Glickman, M.S.
(2009). Cell 138, 146–159.
Trautinger, B.W., Jaktaji, R.P., Rusakova, E., and
Lloyd, R.G. (2005). Mol. Cell 19, 247–258.
Cell Host & Microbe
Cell Host & Microbe 6, July 23, 2009 ª2009 Elsevier Inc.