´
M. Krátky et al. / European Journal of Pharmaceutical Sciences 53 (2014) 1–9
2
future development of anti-tuberculosis drugs should be, inter alia,
targeted to this essential molecule and its derivatives.
et al., 2012; Zhang and Mitchison, 2003). Esters have been found to
have a greater in vitro antimycobacterial activity than POA, gener-
ally assumed that it is a consequence of increased lipophilicity and
that esters, after non-enzymatic or rather enzymatic hydrolysis,
share identical mechanism of action. Despite in vitro improved
antimycobacterial activity of POA, efficacy studies in mice have
failed, presumably due to instability of the POA esters in vivo
(Zhang and Mitchison, 2003).
PZA is active only against M. tuberculosis complex organisms [M.
tuberculosis (Mtb.), Mycobacterium africanum and Mycobacterium
microti, but not Mycobacterium bovis] (Zhang and Mitchison,
2003). Despite the wide use, the mechanism of action has not been
fully elucidated for a long time; it was considered being rather
non-specific without a clear target.
PZA as a prodrug enters M. tuberculosis cell probably by passive
diffusion and it is hydrolyzed intracellularly into its active form,
pyrazinoic acid (pyrazine-2-carboxylic acid; POA), by nicotinami-
dase/pyrazinamidase (PZase). This enzyme encoded by pncA gene
converts nicotinamide into nicotinic acid, primarily. Its various
defective mutations are thought to be the main reason for PZA-
resistance (Jureen et al., 2008; Zhang and Mitchison, 2003; Zhang
et al., 2008). A ‘‘classical’’ mechanism of action involving POA prot-
onization, deprotonization and migration, was described by Zhang
and Mitchison (2003) and Zhang et al. (2003a). In this way, POA
has been proposed to collapse the proton gradient, disrupt mem-
brane potential and transport functions, acidify cytoplasm and
thereby influence vital function. POA also decreases respiratory
synthesis of ATP and its intracellular level (Lu et al., 2011).
Fatty acid synthase I (FAS I) has been suggested as a PZA target,
but a subsequent study ended negatively (Boshoff et al., 2002). By
contrast, some newer studies have found PZA, POA and its simple
esters diminishing FAS I function (Ngo et al., 2007; Zimhony
et al., 2007); recently, it has been revealed that PZA inhibits bind-
ing of NADPH competitively, whereas POA showed a greater affin-
ity for FAS I and a different binding site (Sayahi et al., 2011). The
PZA hydrolysis to POA is not required for FAS I inhibition. Interest-
ingly, alkyl pyrazinoates have been shown to be inhibitors of FAS I,
likely on the same binding site (Sayahi et al., 2012). A previously
unrecognized target of POA was identified in 2011: the ribosomal
protein S1 (RpsA) involved in protein translation and the ribo-
some-sparing process of trans-translation. This mechanism could
explain PZA activity against non-replicating mycobacteria (Shi
et al., 2011).
PZA-resistant M. tuberculosis strains possess loss of PZase activ-
ity mostly due to mutations of pncA (rarely mutations in promoter
or an undefined regulatory gene have been discussed), which is
conventionally considered being a major reason of the resistance
(Jureen et al., 2008; Zhang and Mitchison, 2003; Zhang and Yew,
2009). PZA-resistance in some strains with an optimal PZase activ-
ity can be explained by the variation of POA efflux rate due to
mutations altering the efficiency of the POA efflux pump (Zimic
et al., 2012) and also the changes in RpsA protein represent another
resistance mechanism (Shi et al., 2011).
Nontuberculous mycobacteria share mostly natural PZA-resis-
tance. In Mycobacterium kansasii, it results from the reduced PZase
activity. Additionally, there exists a weak POA efflux mechanism
(Sun and Zhang, 1999). The natural PZA-resistance in other atypical
mycobacteria such as Mycobacterium smegmatis and Mycobacte-
rium avium consists most likely in a highly active POA efflux which
abolishes its accumulation within cells at an acidic pH; their PZase
is fully functional (Sun and Zhang, 1999). A lack or a lowering of
ATP-dependent PZA uptake was proposed being an additional
factor participating in lower PZA susceptibility at some atypical
mycobacteria as well as at M. tuberculosis with acquired PZA-
resistance (Raynaud et al., 1999). That is why PZA is not usually
used in the treatment of infections caused by nontuberculous
mycobacteria.
In contrast, propyl-pyrazinoate, unlike PZA or POA, is active at
neutral pH indicating that POA esters are not only POA prodrugs,
but they likely have intrinsic antimycobacterial activity interacting
with FAS I, without necessary previous hydrolysis. Similarly, POA
esters are active against POA-resistant M. smegmatis and M. avium,
where the resistance is not caused due to defective PZase (Sayahi
et al., 2012).
Pyrazinoic acid esters, which are mostly active towards
extended spectrum of mycobacterial species as well as with im-
proved efficacy for TB strains including those with acquired PZA-
resistance, have been reported (Bergmann et al., 1996; Cynamon
et al., 1992; Cynamon et al., 1995; Seitz et al., 2002; Speirs et al.,
1995; Yamamoto et al., 1995); however, M. avium avoid uniform
susceptibility to POA esters (Cynamon et al., 1992; Speirs et al.,
1995). POA esters with higher linear alcohols were designed to
increase the hydrolytic stability in serum and mycobacterial cell
barriers penetration. These highly lipophilic esters showed signifi-
cantly a greater activity than POA or PZA against M. tuberculosis
and they were more resistant to plasma and liver hydrolysis than
short chain esters, positively correlating with increased lipophilic-
ity. The authors concluded that more hydrolysis-resistant esters
appear convenient as POA prodrugs, overcoming the limitations
}
of previously described esters (Simoes et al., 2009). On the other
side, ester of POA with protected L-serine avoided any activity
against M. tuberculosis (Pinheiro et al., 2007).
In agreement with these results, the investigation of prodrugs
has been expanded during the last decades. The prodrug design
offers the improvement of drug candidate undesired properties,
typically in means of chemical instability, poor solubility, pharma-
cokinetics, efficacy or side effects. Esters and amides are the most
common prodrug strategies used to improve the lipophilicity. Car-
boxylates are converted to the parent compounds by ubiquitous
esterases (Huttunen and Rautio, 2011) or spontaneously. Interest-
ingly, the intrinsic antimicrobial activity of the POA counterpart
alcohol or phenol may be a further advantage, while it might result
in a synergistic action of the mutual prodrug. That is why we se-
lected antimicrobially active salicylanilides (2-hydroxy-N-phe-
nylbenzamides) for the esterification of POA in this study.
Salicylanilides have just revealed many pharmacological activi-
ties; some members of this group are established in human or vet-
erinary medicine. Importantly, they are investigated for their
activity against bacteria including mycobacteria, fungi and proto-
´
zoa (Fomovska et al., 2012; Garner et al., 2011; Krátky and Vinšová,
´
2011; Krátky et al., 2012b; Lee et al., 2013). The exact mechanism
of salicylanilides action is not still fully elucidated; the brief sum-
marization of particular effects on bacterial cells is outlined in our
´
review (Krátky and Vinšová, 2011). Recently, a moderate inhibition
of mycobacterial methionine aminopeptidase and isocitrate lyase
´
´
was reported (Krátky et al., 2012b; Krátky et al., 2013), as well as
others like transglycosylase (Cheng et al., 2010) or ‘‘resurrected’’
disruption of the membrane proton gradient (Lee et al., 2013).
For salicylanilides as phenolic compounds, increased lipophilicity
and subsequent better passing through biomembranes and de-
creased cytotoxicity represent the main reasons for their esterifica-
tion. However, it is still not fully elucidates, if salicylanilide esters
act only as prodrugs releasing parent salicylanilide and acid, or if
The observation that PZA-resistant M. tuberculosis retains sus-
ceptibility to POA has led to the development of its esters. While
ionized POA does not penetrate through mycobacterial cell wall
well, its derivatives may prevent this obstacle and may circumvent
PZase deficient M. tuberculosis and nontuberculous strains (Sayahi
´
they may interact with target sites as original entities (Krátky
and Vinšová, 2011).
Krátky, Martin
