The mar regulon specifies an intrinsic
bacterial response to antibiotics
Michael N Alekshun, PhD
Center for Adaptation Genetics and Drug Resistance, Tufts University School of Medicine, Boston, Massachusetts,
Multiple antibiotic resistance in bacteria is generally attributed to the acquisition of plasmids and/or transposons
from other resistant organisms. However, intrinsic mechanisms of resistance, such as the chromosomal multiple antibiotic
locus found in Escherichia coli
and other members of the Enterobacteriaceae (Table 1), are equally important. The E.
coli mar locus controls resistance and/or susceptibility
to many structurally unrelated compounds including antibiotics, household disinfectants, organic solvents and other
toxic chemicals (1) (Table 2).
Mar mutants were initially selected following growth on sub-inhibitory concentrations of either tetracycline or
chloramphenicol.(2) Further experiments demonstrated that these mutants were resistant not only to the selective
agents but also to penicillins, cephalosporins, puromycin, quinolones and rifampin.(2) E.
coli Mar mutants use a number of different resistance mechanisms
simultaneously to reach a state of decreased antibiotic susceptibility.
While bacteria can use acquired resistance genes to produce enzymes that modify or inactivate drugs, these mechanisms
are usually specific for a particular drug or class of antibiotics (e.g., beta-lactamases). However, microbes may
be resistant to multiple antibiotics via other mechanisms, namely increased removal from (efflux)(3) and/or decreased
penetration into (influx) (4) the cell.
The proteins responsible for engendering a Mar phenotype in E.
coli are specified by the marRAB operon. MarR represses the synthesis of this operon while MarA activates
it.(1) It is believed that MarR is inactivated following exposure of the cell to many of the chemicals to which
it will eventually become resistant. Some drugs such as oxidative stress agents, may interact directly with MarR
and inactivate the repressor while others such as tetracycline and chloramphenicol, may do so indirectly (Table
2). In both cases, the result, overexpression of the "master activator" MarA, will be the same.
Once the cell has produced an abundance of MarA, this protein activates the expression of many other genes, constituents
of the Mar regulon,(1) throughout the chromosome. While the activity of MarA might initially be perceived as promiscuous,
this global transcriptional regulator achieves two major goals. In E.
coli, MarA decreases outer membrane porin (non-specific
pathways that antibiotics use to get into the cell) synthesis to decrease antibiotic influx. It also activates
the AcrAB/TolC5 multi-drug pump to increase drug efflux. Both changes will function cumulatively to reduce the
concentration of toxic chemicals within the cell. However, some drugs may eventually gain access. Oxidative stress
agents, such as paraquat and plumbagin, generate the exceptionally reactive and the extremely damaging superoxide
ion6 (a free radical of the ubiquitous element oxygen). In order to counteract the damaging effects that superoxide
inflicts upon the cell, MarA also increases the expression of many cytoprotective enzymes.(1)
While first-step Mar mediated resistances to some antibiotics (i.e., tetracycline) are clinically significant,(1)
resistances to other drugs are at less relevant levels. With further drug contact, higher level resistance appears.
Patients who do not complete a prescribed round of chemotherapy or take antibiotics at subtherapeutic levels would
be key contributors to the development of such organisms. This situation is worrisome because Mar mutants are the
progenitors of strains that become highly resistant, through mutations elsewhere on the chromosome. This is especially
evident in the emergence of fluoroquinolone resistance where a percentage of clinically resistant strains are Mar
mutants.(7) E. coli
that constitutively express MarA reach clinical levels of fluoroquinolone resistance faster than strains that do
not.(8) Moreover, inducible resistance in Haemophilus influenzae (9) suggests that overexpression of an AcrAB-type
efflux system in H. influenzae
might facilitate the development of resistance.(10) From what is now known about the E.
coli mar system, it would not be surprising to learn that
other organisms also employ similar survival mechanisms.
- Alekshun MN, Levy SB. 1997. Antimicrob Agents Chemother
- George AM, Levy SB. 1983. J Bacteriol 155:531-540.
- Levy SB. 1992. Antimicrob Agents Chemother 36:695-703.
- Nikaido H. 1989. Antimicrob Agents Chemother 33:1831-1836.
- Okusu H, Ma D, Nikaido H 1996. J Bacteriol 178:306-308.
- Demple B. 1991. Ann Rev Genet 25:315-337.
- Oethinger M, Podglajen I, Kern WV, Levy SB 1998. Antimicrob
Agents Chemother 42:2089-2094.
- Cohen SP, McMurry LM, Hooper DC, Wolfson JS, Levy SB.
1989. Antimicrob Agents Chemother 33:1318-1325.
- Marshall B, Roberts M, Smith A, Levy SB. 1984. J Infect
- Sánchez L, Pan W, Viñas M, Nikaido H.
1997. J Bacteriol 179:6855-6857.