Current Pharmaceutical Design, 1996, 2, 175-194
`
`175
`
`Oxazolidinone Antibacterial Agents
`
`Steven J. Brickner, Ph.D.
`
`Medicinal Chemistry, Pharmacia & Upjohn, Inc., 301 Henrietta Street, Kalamazoo,
`MI 49001, U0A
`Abstract: The oxazolidinones are a new class of synthetic antibacterial agents. These
`compounds demonstrate potent in vitro and in vivo activity against important human
`pathogens, including multiple antibiotic-resistant strains of gram positive organisms
`including the staphylococci, streptococci, and enterococci. The oxazolidinones have a
`novel mechanism of action, inhibiting bacterial protein synthesis at a very early step
`prior to initiation. Literature disclosures have described the inability to detect in vitro
`bacterial resistance development to the oxazolidinones. Only the (S)-enantiomer is active; a new synthetic
`route yielding oxazolidinones with high optical purity has been reported. This paper will review the spectrum
`of activity, mechanism of action studies, toxicity issues, and structure activity relationships of the
`oxazolidinones.
`
`Introduction
`
`The oxazolidinones are novel synthetic antibacterial agents
`that show considerable promise for the treatment of human
`infections caused by problematic multi-drug resistant and
`sensitive gram-positive bacteria and My cob a c te r i um
`tuberculosis
`[l]. If the drugs currently in development are
`. proven efficacious in man, the oxazolidinones would represent
`the first new class of antibacterial agents to be developed in over
`a decade.
`
`Following an account of the discovery and early lead
`development by DuPont that led to the 5-acetamidomethyl-3-
`aryl-2-oxazolidinones, the major focus of this review turns to an
`extensive survey of the more relevant structure-activity
`relationship [SAR] findings. In-depth discussion also centers on
`those few oxazolidinones
`that have entered
`the drug
`development process, and have proceeded at least as far as Phase
`I human clinical trials.
`
`Given that this is a burgeoning area of research, the
`relatively few publications in the open literature have been
`supplemented with selected biological data gleaned from
`published patents, patent applications, and meeting abstracts.
`Various associated compounds that are best classified as
`members of other antimicrobial drug classes, yet happen to
`contain an oxazolidinone ring substituent,0 are excluded from
`the realm of this review. Discussion will also be found
`concerning aspects of the monoamine oxidase [MAO] inhibitory
`properties of certain 5-(substituted)methyl-3-aryl-2-
`oxazolidinones1 (eg., 5-hydroxymethyl [5] or 5-aminomethyl
`groups [6]), as such compounds have considerable synthetic
`utility as chemical intermediates to the subject compounds, or
`may themselves exhibit some degree of antibacterial activity.
`
`Nomenclature
`
`The antibacterial oxazolidinone pharmacophoric template
`can be characterized as that depicted in Fig. (1), i.e., an (S)-3-
`
`0These include the nitrofuran furazolidone [2] and the 3-chloro
`oxazolidinone bactericides [3].
`1oxazolidinones having MAO inhibitory properties have been known for
`almost 40 years [ 4].
`
`aryl-5-(substituted)methyl-2-oxazolidinone. Throughout the
`review, a consistent numbering system has been employed for
`description of the various oxazolidinones examined. 2
`
`3'
`
`2'
`
`O
`
`4·0-~ tt01 R
`
`-
`6'
`5'
`Fig. (1). Oxazolidinone numbering scheme.
`
`~5 ' ,..,c..__
`N
`H
`
`CH
`3
`
`Bacterial Resistance and the Search for Novel
`Antibiotic Structural Templates
`
`)'
`
`The alarming escalation seen worldwide in the incidence of
`bacterial resistance to previously effective antibiotics [7],
`continues to provide the impetus for the medicinal chemist to
`search for entirely new classes of antibacterial agents, that can
`cure bacterial infection by novel mechanisms [8,9]. As
`numerous bacteria have increasingly evidenced the evolution of
`multiple-antibiotic resistance, health care providers have been
`seriously challenged to provide effective therapy for the often
`life-threatening infections caused by these pathogens. Numerous
`reviews have recently appeared emphasizing the extent and
`severity of the resistance problem as it exists today, and the dim
`prospects envisioned for the future [10-14].
`
`Some of the most problematic organisms of the 1990's have
`been the multi-drug resistant gram-positive bacteria. These
`include the highly virulent organism, methicillin-resistant
`Staphylococcus aureus [MRSA] [9], and penicillin- [15] and
`cephalosporin-resistant Streptococcus pneumoniae. For many
`strains of MRSA, only vancomycin is still effective [10], as
`such MRSA isolates are also found resistant to virtually all
`penicillins and cephalosporins, tetracyclines, aminoglycosides,
`lincosamides, chloramphenicol, macrolides
`[ 16], and
`quinolones [17-19].
`
`2Designation of regiochemistry of substitution about the 3-aryl ring is
`indicated by the use of primed (') numerals. Where an additional aryl ring
`is linearly appended, double-primed (") numbering is employed for that
`ring. For the sake of clarity the same system has been applied to describe
`more structurally complex, multicyclic fused-ring templates, in lieu of the
`rather abstruse conventional nomenclature. '
`
`1381-6128/96 $8.00 + .00
`
`© 1996 Bentham Science Publishers B.V.
`
`MYLAN - EXHIBIT 1021
`
`

`
`176 Current Pharmaceutical Design, 1996, Vol. 2, No. 2
`
`Steven J. Brickner
`
`mm1mum inhibitory concentrations [MICs] of 5, 2.5, and 5
`µg/mL, respectively. In lethal infection models in mice, oral
`administration [po] of 2 or 3 required a dose to protect 50% of
`the infected animals [ED50] of 29 mg/kg and 20 mg/kg vs S.
`aureus, respectively. Weak activity (ED50=63 mg/kg) was
`observed for 2 vs the gram-negative Escherichia coli.
`
`The enterococci have become increasingly of concern, in
`that over the last five years, there have been isolated problems
`of infections with vancomycin-resistant enterococci [VRE] [11].
`The VRE are also resistant to essentially all other antibiotics
`[20]. Mortality rates greater than 35% have been reported for
`patients infected with VRE [21]. A major concern is that VRE
`will transfer the vancomycin-resistance genes encoded on a
`plasmid [22], to the much more virulent organism S. aureus, as
`Noble et al. [9] have demonstrated to be feasible in an
`experimental setting.
`
`Still another problem microorganism for which resistance
`has generated considerable medical concern is multidrug(cid:173)
`tuberculosis is a
`resistant M. tuberculosis [MDRTB] [23]. M.
`highly virulent human pathogen
`that
`is
`the cause of
`approximately 3 million deaths per year worldwide [24]; strains
`resistant to isoniazid [INH] are widespread throughout the world.
`Infections with M. tuberculosis are now commonly treated in the.
`USA with a combination of INH, rifampin, and pyrazinamide
`[25].
`
`This loss of antibacterial activity among drugs once
`efficacious against these pathogens, has led to a summoning
`from n)lmerous experts [14,26] for
`the discovery and
`development of new antibiotic classes, so that health care
`practitioners will not be left bereft of effective therapeutic
`modalities [27]. The oxazolidinone antibacterial agents may, if
`demonstrated to be efficacious in man, provide an answer to this
`call as the first new class of antibiotics to be developed since the
`fluoroquinolones [ 19].
`
`Discovery of the Oxazolidinone Antimicro- ·
`bial Agents
`
`5-Halomethyl-3-Aryl-2-0xazolidinones
`
`In a 1978 U.S. patent assigned to E.I. du Pont de Nemours
`and Co., Inc., Fugitt and Luckenbaugh [28] described a series of
`racemic 5-halomethyl-3-aryl-2-oxazolidinones3 which were
`claimed to have utility for the systemic control of bacterial and
`fungal
`foliage diseases of plants. Synthesis of
`the
`oxazolidinone ring was accomplished by heating an isocyanate
`and an epihalohydrin with a lithium halide catalyst.
`
`Application to tomato plants of a 200 ppm suspension of 5-
`( chloromethyl)-3-( 4'-methylthio )phenyl-2-oxazolidinone 1, or
`its corresponding 4'-methylsulfonyl congener 2, prior to
`inoculation with either Agrobacterium
`tumefaciens or
`Xanthomonos vesticatoria, was effective in preventing the
`establishment of disease manifested as crown gall or leaf spot. 4
`
`Subsequently, a patent issued to the same inventors [32]
`which described the antibacterial activity of related 5-
`halomethyl-2-oxazolidinones against bacterial pathogens
`infective of mammals. 5-Chloromethyl oxazolidinones 2 and 3,
`and 5-bromomethyl oxazolidinone 4 demonstrated good in
`vitro activity against Staphylococcus epidennidis, with
`
`3The author presumes that these are the lead compounds reportedly
`129,30] discovered as a result of random screening.
`A Bayer A.-G. group [31] recently filed a patent application describing
`the agricultural fungicide 5-bromomethyl-3-(4'-CF3S02)phenyl-2-
`oxazolidinone.
`
`-0- J ~Jl
`NO
`MeS020N 0
`
`R
`
`- ~R
`
`-
`
`~OH
`
`1 X= Cl R=MeS
`2 X=Cl R=MeS0 2
`3 X=Cl R=F 2HCS0 2
`4 X=Br R=MeSOz
`
`5
`
`5-Hydroxymethyl-3-Aryl-2-0xazolidinones
`
`The same patent [32] also described antibacterial 5-
`hydroxymethyl-3-aryl-2-oxazolidinones and the corresponding
`esterified alcohols. Most active in vivo was the optically active
`5-(R)-hydroxymethyl-3-(4'-methylsulfonyl)phenyl-2-oxazolidi(cid:173)
`none 5, prepared via diethylcarbonate-mediated cyclization of d-
`3-( 4-methylthioanilino )-1,2-propanediol, then oxidation with
`MCPBA. The resultant sulfone 5 exhibited reasonable in vitro
`and in vivo activity vs S. aureus (Table I).
`
`(R) -5 - (Methoxymethyl) - 3 - [4' -
`S-6123 and
`(Methylsulf ony l)Pheny l] -2- Oxazo lidinone
`
`[33] disclosed a series of 5-
`In 1984, Gregory
`hydroxymethyl-, 5-acetoxymethyl-, and 5-methoxymethyl-3-
`aryl-2-oxazolidinone antibacterial agents, wherein the preferred
`compounds bore 41-(H2NS02)- or 4'-MeS02- aryl substituents.
`S.-6123, and the 5-methoxymethyl analog 6 had notable
`potency (Table I). While S-6123 was singled out for further
`study,
`relative
`to
`the
`subsequently developed 5 -
`acetamidomethyl-2-oxazolidinone congeners, it exhibited only
`weak in vitro antibacterial activity [34,35]. Nevertheless, Daly
`et al. [34] reported that upon po administration of S-6123 (at a
`level producing a serum concentration of only one-tenth the
`MIC), 20 of 23 rats survived a lethal E. coli peritonitis
`challenge. On an ED50 basis, compound 6 had better in vivo
`activity than S-6123 (Table I) [33]. For the sake of comparison
`with the corresponding 5-acetamidomethyl-2-oxazolidinone
`clinical candidate DuP 721 (vide infra), Table I also gives in
`vivo data for the much less active (±)-3-(4'-acetyl)phenyl-5-
`(7) [33]. Any further
`hydroxymethyl-2-oxazolidinone
`developmental activity with either 6 or S-6123 was not
`evidenced in the literature.
`
`S-6123 R = H 2NS0z
`R=MeCO
`7
`
`6
`
`

`
`Oxazolidi11011e Antibacterial
`
`Current Pharmaceutical Design, 1996, Vol. 2, No. 2 177
`
`Table I. Antibacterial Activity of 5-Hydroxymethyl Oxazolidinones
`
`Organism
`
`Mean MIC(µg/mL)1
`5[33]
`S-6123[33,34]
`
`Streptococci spp.
`
`Staphylococci spp.
`
`E.coli
`
`Salmonella spp.
`
`Neisseria spp.
`
`Haemophilus inf/uenzae
`
`Clostddium spp.
`
`Bacteroides spp.
`
`Fusobacterium spp.
`
`3.7
`
`3.8
`
`21.7
`
`16
`
`10.5
`
`16
`
`7
`
`3.4
`
`0.3
`
`8-32
`
`22-64
`
`30
`
`32
`
`12
`
`32
`
`17
`
`~15
`
`ED5~
`~
`
`9
`
`25
`
`S-6123
`
`17.1
`
`13.2
`
`2
`5.1
`
`11.8
`
`1
`35.8
`
`97.6
`
`S. aureus
`
`E.coli
`
`1 Mean MIC for multiple isolates
`
`MAO Inhibition
`
`In view of the pharmacological activity of a class of
`reversible and competitive MAO inhibitors [36] having the 5-
`hydroxymethyl-3-aryl-2-oxazolidinone pharmacophore ( eg.,
`antidepressant toloxatone 8 [37]), or the 5-methoxymethyl-3-
`aryl-2-oxazolidinone template (eg., cimoxatone 9 [38]), there is
`some concern whether the antibacterial 3-aryl-2-oxazolidinones
`bearing these side-chains could potentially engender undesirable
`side-effects, via inhibition of mammalian MAO-A or MAO-B
`isozymes [6,37,39].
`
`More pertinent with respect to
`the antibacterial 5-
`acetamidomethyl-3-aryl-2-oxazolidinones (vide i11fra), are the 5-
`aminomethyl-3-aryl-2-oxazolidinones that are known to inhibit
`MAO [78]. Compound 10 [79] was reported to be a reversible
`inhibitor of MAO-B. Interestingly, Silverman and Ding [79]
`showed that synthetic conversion of this primary amine to the
`corresponding secondary amine (5-methylaminomethyl) or
`tertiary amine [5-(N,N-dimethylaminomethyl)] side-chains
`transformed the oxazolidinones into irreversible inactivators of
`MAO. The irreversible [76] MAO-B inhibitor MD780236
`[5,39] is one of the most potent oxazolidinones, being selective
`for MAO-B.5 Dostert and coworkers [80] have also shown that
`10, the S-enantiomer of the primary amine derivative
`to MD 7 8 0 2 3 6, is a potent inhibitor of
`corresponding
`semicarbazide-sensitive amine oxidase. Thus, there may be
`potential side-effect implications should the antibacterial 5-
`acetamidomethy 1-2-oxazolidinone
`infra) undergo
`(vide
`metabolic amide cleavage, to generate a 5-aminomethyl-2-
`oxazolidinone.
`
`[38] have
`[76,81] and Strolin-Benedetti
`Silverman
`extensively studied the mechanism of inactivation of the
`flavoenzyme MAO by these and related oxazolidinones. In the
`case of the 5-(methylamino)methyl oxazolidinones, one
`mechanism for the inactivation of MAO-B was proposed to
`involve one-electron oxidation of the methylamine [81]. Loss
`of a proton from the amine radical cation resulted in a radical that
`can covalently link with an amino acid residue in the enzyme,
`resulting
`in enzyme-inactivation
`[76]. An alternative
`mechanism involves the oxidation of the amine to an imine (or
`aldehyde [39]), which then traps a nucleophilic amino acid
`residue [76]. For a discussion of the proposed binding
`mechanism of the 5-hydroxymethyl-3-aryl-2-oxazolidinones,
`refer to Koenig [37].
`
`,,f
`N
`
`9
`
`MD 780236 R = Me
`105-(S) R=H
`
`5-(S)-Acetamidomethyl Oxazolidinones
`
`DuPont Clinical Candidates DuP 721 and DuP
`105
`
`The first description of 5-(S)-acetamidomethyl-3-aryl-2-
`oxazolidinone antibacterial agents is found in a 1984 patent
`application [l]. In late 1987, the DuPont group [40,83]
`presented detailed information on two oxazolidinones bearing
`the (S)-5-acetamidomethyl side chain: DuP 721 6 , and DuP
`1057 . These parenterally and orally active oxazolidinones were
`identified as members of a new class of synthetic antibacterial
`agents, having a novel mechanism of action [40,41]. Only the
`5-S enantiomer of the oxazolidinones is active [29].
`
`Unlike S-6123, DuP 721 demonstrated potent in vitro and
`in vivo activities against gram-positive pathogens, including
`MRSA, Staphylococcus epidermidis, S. pneumoniae, and
`enterococci, as well as against anaerobes, and M. tuberculosis
`[42] (Table II). There was no inhibitory activity against gram(cid:173)
`negative aero bes, or Candida albicans [30,43]. In a lethal mouse
`
`5Both the R and S enantiomers have been reported to be irreversible
`inactivators of MAO-B [5,77].
`
`6 ( S)-N-[ [3-( 4-acety lpheny 1 )-2-ox o-5-oxazolidin y 1 ]methyl ]-acetamide
`7 5-( S)-N- [[3-( 4-methy lsulfiny 1 )phenyl ]-2-oxo-5-oxazolidiny l ]methyl](cid:173)
`acetamide
`
`

`
`178 Current Pharmaceutical Design, 1996, Vol. 2, No. 2
`
`Steven J. Brickner
`
`0
`
`N
`~N
`\__J
`HO_/
`
`0 ~-q-Jl O
`f/
`\ N
`O
`11
`-
`~N,.C._,CH3
`H
`
`F
`
`U-100592
`0
`
`rN-p-~ NJlO
`
`model, DuP 721 demonstrated activity (ED50=2.2 and 2.8
`mg/kg, sq and po, respectively) better than, or comparable to,
`vancomycin (sq) for MRSA. Against E. faecalis and methicillin(cid:173)
`sensitive S. aureus, Dup 721 had ED50 values of 2.5-3.1 mg/kg
`and 3.9-5.6 mg/kg, respectively, being conside1·ably more
`potent than vancomycin.
`
`0
`
`o\~~N)l_o R
`I
`HC~ \
`N..,.c.._,CH
`~H
`3
`3
`Dup721
`
`0
`
`~\ -0-'' )l_
`
`S
`I
`H3C
`
`I
`
`-
`
`\
`
`0
`O
`N
`II
`\_ I
`..,.c.._,
`~~ CH3
`Dup 105
`
`In a number of studies, the in vitro potency of DuP 721 was
`found to be somewhat less than vancomycin for most gram(cid:173)
`positive organisms, with MIC90 values of 4 µg/mL for S. aureus
`(methicillin-sensitive or resistant) [30,44], and 1-4 µg/mL for
`the streptococci and enterococci [30]. The MIC90 for Bacteroides
`fragilis was 4 µg/mL. DuP 105 was approximately 4-fold less
`active than DuP 721, and several fold less active in vivo than
`vancomycin for most organisms
`[30,41, 45-50]. The
`antibacterial spectrum of both DuP 721 and DuP 105 was
`represented as most closely resembling that of lincomycin [43].
`Both compounds were bacteriostatic [30]. Several studies
`demonstrated the inability to select for resistant variants to DuP
`721 or DuP 105 [41,43,45].
`
`A two-fold difference in activity for the two DuP 105
`diastereomers was noted, with the R-absolute configuration of
`the methylsulfinyl group being most active [51]. This R, S
`diastereomer was extremely water soluble (782 mg/mL), in
`contrast to the S,S diastereomer (112 mg/mL). The mixture of
`diastereomers (i.e., DuP 105), had a solubility of 311 mg/mL.
`
`DuP 721 [52] and DuP 105 [53] reportedly entered Phase I
`clinical trials, but the development of each was subsequently
`discontinued [54,55]. It should be noted that Piper and
`coworkers at Upjohn8 found (±)-DuP 721 to exhibit lethal
`toxicity in the rat (vide infra).
`
`Upjohn Clinical Candidates U-100592 and U-
`100766
`
`In late 1995, a team from the Upjohn Co. [56-62] reported
`on the development of two (S)-5-acetamidomethyl-2-
`oxazolidinones that were in late Phase I clinical trials and slated
`to enter Phase II studies. U-100592 9 and U-100766 10 were
`like DuP 721,
`3-(3'-fluorophenyl)-2-oxazolidinones
`that,
`exhibited in vitro and in vivo antibacterial activity comparable
`to vancomycin (Table II), but did not have the toxicity observed
`with (±)-DuP 7218 [63].
`
`\__J
`
`R
`- ~ .... C._,
`CH3
`N
`H
`
`F
`
`U-100766
`
`U-100592 and U-100766 were under development for the
`treatment of gram-positive infections caused by both sensitive
`and drug-resistant strains of staphylococci, streptococci, and
`enterococci. Formulations were available allowing both drugs to
`be administered by iv and oral routes. Zurenko et al. [57] showed
`that neither drug was cross-resistant with vancomycin for
`enterococci, nor with penicillin for pneumococci. Activity
`comparable
`to
`tetracycline was observed against
`the
`Mycoplasma spp., respiratory pathogens in swine and humans
`[64]. A postantibiotic effect vs S. aureus of 2.3 and 1.8 h was
`observed for U-100592 and U-100766, respectively [57].
`
`In several in vivo studies carried out by Ford and coworkers
`in the mouse, U -10 0 5 9 2 administered sq or po
`[58]
`demonstrated activity comparable to sq vancomycin (for eight
`MRSA strains, ED50=0.9-8.0 mg/kg), while U~l00766 was
`equivalent to, or slightly less active (ED50=2.0-15 mg/kg,
`MRSA) than vancomycin. Against the following organisms, U -
`100592 and U-100766 had respective ED50 values of 1.9 and
`4.7 mg/kg (S. epidermidis); 1.3 and 10.0 mg/kg (E. faecalis);
`12.5 and 25.0 mg/kg (vancomycin-resistant E. faeciwn); and
`1.2-11.7 and 2.5-3.8 mg/kg (penicillin- and cephalosporin(cid:173)
`resistant Streptococcus pnewnoniae).
`Table II.
`In Vitro Antibacterial Activity of Oxazolidinone Drug
`Candidates, MIC90 (µg/mL)
`
`Organism
`
`DuP7215 DuP 1055 u-1005926 U-1007666
`
`MSSA1
`MRSA2
`MSSE3
`MRSE4
`Enterococci
`S. pyogenes
`S. pnewnoniae
`Co1)'nebacteri11111 spp.
`H. influenzae
`M. catarrhalis
`B.fragilis
`gram-(+) anaerobes
`M. hyopneumoniae 7
`Mycoplasma spp.8
`
`1-4
`2-4
`NT9
`NT
`2-8
`NT
`2
`NT
`32
`NT
`8 - 16
`NT
`NT
`NT
`
`4-16
`4-16
`NT
`NT
`16
`NT
`4
`NT
`64
`NT
`16
`NT
`NT
`NT
`
`2-4
`2-4
`1
`1
`1-2
`1-2
`0.50-1
`0.50
`16
`4
`16
`0.50-2
`2.0
`16
`
`2-4
`2-4
`2
`2
`1-4
`2-4
`1-2
`0.50
`16
`4
`4
`0.50-2
`2.0
`8
`
`8unpublished data, R.C. Piper, T.F. Platte, and J.R. Palmer, The Upjohn
`Company.
`9 (S)-N-[[[3-[3-Fluoro-4-[N-1-( 4-hydroxyacetyl)-piperazinyl ]]-phenyl]-
`2-oxo-5-oxazolidinyl]methyl]acetamide
`lO ( S)-N-[[3-[3-Fluoro-4-morpholinyl]phenyl]-2-oxo-5-
`oxazolidinyl]methyl]-acetamide
`
`1 Methicillin-sensitive S. aureus; 2Methicillin-resistant S. aureus; 3Methicillin(cid:173)
`sensitive S. epidennidis; 4Methicillin-resistant S. epidennidis; 5compilation of
`MIC90 data taken from [30,41,45,46,49) 6compilation of MIC90 data taken from
`[57,129-137). 7 respiratory pathogens of swine [147); 8Human pathogens
`Mycoplasm hominis and M. p11ew11011iae [64]; 9NT=not tested (or not reported).
`
`

`
`Oxazolidinone Antibacterial
`Activity Against M.
`M. Avium
`
`tuberculosis and
`
`DuP 721 and DuP 105. DuP 105 was reported to have
`MICs of 0.3-1.25 µg/mL against M.
`tuberculosis [82]. The
`activity of DuP 721 vs 25 clinical isolates of M. tuberculosis
`compared well with that of rifampin; MICs were 0.3-1.25
`µg/mL, and 99.9% kill minimum bactericidal concentration
`[MBC] was 2.5 µg/mL [ 42]. DuP 721 was found not to be
`cross-resistant [82,84], and attempts to select a resistant mutant
`were unsuccessful [82,83]. While the drug was inactive against
`M. avium and M.
`intracellulare, atypical mycobacterial
`pathogens had MICs of 1.9-15.6 µg/mL [85]. Antagonism with
`rifampin was seen in vitro [84,85].
`
`tuberculosis, an ED50 of 13.2
`In mice infected with M.
`mg/kg (100% survival at 50 mg/kg) was observed, when (±)·
`DuP 721 was administered daily for 17 days [85]. The increase
`in survival time was inferior to similar treatment with either
`rifampin or INH.
`
`Structurally Unidentified Oxazolidinones
`
`The activity of four structurally unidentified DuPont
`oxazolidinones vs M. avium complex [MAC] isolates from HIV(cid:173)
`infected patients was reported [86]. The compounds E-3656-2, E-
`3709-5 (relation to E-3709 not identified), and E-3556-2 had
`MICs= 0.5-4.9 µg/mL, while XA-043 (a more water soluble E-
`3709-5 derivative) was less active. Evidence for cross resistance
`with amikacin was reported.
`
`U-100480, U-100766, and U-100592
`
`Barbachyn et al. [87] reported that substitution of a
`thiomorpholinyl ring for the morpholinyl ring of U -1007 66
`gave the very potent anti-tuberculosis drug U-100480. Kilburn
`et al. [87] identified U-100480 and its sulfoxide metabolite U -
`101603 as the most potent of the Upjohn oxazolidinones
`tested vs M. tuberculosis H37Rv, (MIC :<;;0.125 µg/mL, vs the
`clinical benchmark INH, MIC=0.2 µg/mL). The corresponding
`sulfone U-101244 (a minor metabolite of U-100480), U ·
`100592, and U-100766 [56] also exhibited potent activity
`(MICsos :<;; 0.50 µg/mL) vs a battery of five drug sensitive, and
`five multi-drug resistant strains of M.
`tuberculosis. In this
`battery, the racemate of 4'-indolinyl oxazolidinone U-97456
`[73,96] also demonstrated very good in vitro activity (MICso
`=1.0 µg/mL).
`
`Current Pharmaceutical Design, 1996, Vol. 2, No. 2 179
`
`Klemens and coworkers [88] reported on the po in vivo anti(cid:173)
`tuberculosis activities of U-100480 and U-100766, relative
`to INH in CD-1 mice infected with M. tuberculosis (MICs for U-
`100480, U-100766 and INH were 1, 0.5, and 0.03 µg/mL,
`respectively). Treatment for four weeks with U-100480 (at 100
`mg/kg) gave comparable activity as INH (at 25 mg/kg), whereas
`U-100766 was somewhat less active.
`
`Cynamon et al. [89] evaluated U-100480 and U-100766,
`relative to azithromycin (AZI), in beige mice infected with
`MAC (for all three drugs, the MIC=4 µg/mL). Drug was
`administered (po) for 10 days at 100 mg/kg. Spleen and lung cell
`counts showed U-100480 to be more active than U-100766,
`but less active than AZI. Upon dosing for 4 weeks, U-100480
`and AZI at 100 mg/kg showed similar activity vs MAC in the
`lungs, but AZI was more active in the spleen.
`
`Oxazolidinone Mechanism of Action
`Studies
`
`DuP-721. DuP-721 has been extensively studied in
`attempts
`to elicit
`the mechanism of action of
`the
`oxazolidinones. The drug was shown to inhibit bacterial protein
`synthesis via a new mode of action, the details of which remain
`not well defined. Extensive work by Eustice and colleagues led
`them to surmise that "DuP 721 may inhibit recognition of the 3'
`upstream ribosome-binding sequence present in natural mRNAs"
`[65], i.e., acting at the Shine-Delgarno ribosomal recognition
`site [66]. DuP 721 inhibited protein synthesis (IC50=3. 8
`µg/mL) in an OmpF-mutant E. coli strain, but not DNA or RNA
`synthesis (IC50>64 µg/mL) [65,67].
`
`found with other 5-
`As has been subsequently
`acetamidomethyl-2-oxazolidinones, DuP 721 is inactive vs
`wild-type E. coli [30] and other gram-negative aerobes. This
`resistance to DuP 721 has been attributed to the gram-negative
`cell wall outer membrane [65]. Upon finding activity against E.
`coli permeability mutants, DuPont concluded the target site of
`action was the same for both gram-negative and gram-positive
`bacteria [ 65].
`
`Addition of DuP 721 to cell-free systems (using synthetic
`or natural mRNA templates) did not inhibit the stages of protein
`synthesis from chain initiation, through elongation. DuP 721-
`growth-arre s ted
`cells were
`found
`defective
`in
`initiation-dependent peptide synthesis directed by RNA from an
`MS2 bacteriophage. Extracts from those cells could elongate a
`peptide, yet were then unable to subsequently use natural mRNA
`for initiation of protein synthesis. Hence, it was concluded that
`DuP 721 was acting at an early step of protein synthesis, one
`preceding the interaction of the fMet-tRNA and the 30S
`ribosome with the initiator codon [41,65].
`
`Similar mechanistic studies with S-6123 demonstrated
`that it also inhibited protein synthesis, but not RNA or DNA
`synthesis [34,35].
`
`Bacteriostatic vs Bactericidal Nature of
`Oxazolidinones
`
`Time-kill curve studies indicated that both DuP 721 and
`DuP 105 were bacteriostatic vs. all organisms tested, except
`the diptheroids, with MBCs of 2-4 µ'.g/mL (DuP 721) and 4-16
`
`

`
`180 Current Pharmaceutical Design, 1996, Vol. 2, No. 2
`
`Steven J. Brickner
`
`µg/mL (DuP 105) [30,41,43,45]. U-100592 and U-100766
`were bactericidal for S. pneumoniae, and bacteriostatic for most
`staphylococci and enterococci [57].
`
`Combination
`Antibiotics
`
`Therapy with
`
`other
`
`Antagonism between DuP 721 and ciprofloxacin [68] or
`norfloxacin [48] was reported for MRSA, S. epidermidis, and E.
`faecalis; in a few strains, additivity was observed. Antagonism
`between S-6123 and nalidixic acid was also reported [35].
`
`There have been no reports of observed antagonism in
`combination drug studies with either U -10 0 5 9 2 or U -
`100766. In in vivo combination therapy studies, both drugs
`were found additive with vancomycin, imipenem, gentamicin, or
`rifampin against S. aureus. Similarly, in a mixed S. aureus and E.
`the U1.,i :m drugs combined with
`coli infection model,
`aztreonam or gentamicin as well as did the comparator
`vancomycin [58].
`
`mg) to 6.9 h (2000 mg), and urinary excretion of 40-63% of
`unchanged U-100592 was observed.
`
`Studies with U-100592 dosed po for 14.25 days in healthy
`human males and females at levels of 250, 500, 1000, 1500, and
`2000 mg q.i.d. indicated the drug was well tolerated (the
`maximum tolerated dose was not established in this study) [70].
`Medical events observed were gastrointestinal symptoms
`(change in stool consistency in 80% of subjects on drug, none
`requiring discontinuation of dosing or other medication), and
`yeast superinfection (53% of subjects on U-100592). For
`subjects given doses ~4 g/day, there was observed a dose(cid:173)
`dependent decrease in the platelet count, and a slight decrease in
`reticulocyte count, both of which
`recovered upon
`discontinuation of drug treatment.
`Steady-state AUC levels
`were dose proportional and linear. The renal clearance exceeded
`the glomerular filtration rate,
`indicating probable active
`secretion of U-100592 by the renal tubules. Cmax values
`ranged from 1.6 µg/mL (250 mg dose) to 11.4 (2000 mg), and
`averaged steady state t112 values ranged from 9.3-18.1 h, across
`the 6 dosage levels (1000 mg q.i.d., t112 =12.7 h).
`
`Pharmacokinetic Studies
`
`U-100766
`
`DuP 721 and DuP 105
`
`In mice dosed sq with DuP 721, good dose proportionality
`was observed, with average maximum plasma concentration
`[Cmax] values of 9, 21, and 38 µg/mL for doses of 20, 40, and
`80 mg/kg, respectively [69]. For the mouse and rat, good oral
`absorption was observed (Cmax=15 µg/mL, at 20 mg/kg), as
`well as in the dog (Cmax= 6 µg/mL at 10 mg/kg). The half life
`[t112] in orally dosed mice was 50 min, and in rats and dogs, was
`2 h [30]. DuP 105 had a t112 in the mouse of 30 min, and 50
`mg/kg po in mice gave a Cmax of 32 µg/mL.
`
`U-100592
`
`The metabolism and pharmacokinetics of the drug candidate
`U-100592 were studied in rats and dogs [59]. The terminal t112
`was 1 h in rats and 1.9 h in dogs (iv 10 mg/kg). Oral
`bioavailability was 56% and 100%, in the rat and dog,
`respectively; a 25 mg/kg dose gave Cmax values of 2.3 µg/mL
`and 18.9 µg/mL, respectively. The volume of distribution at
`steady state in both species was about 1 L/kg. U-100592 was
`predominantly renally cleared. Radiolabelled drug studies in
`both species indicated very little metabolism; greater than 90%
`of urinary radioactivity was attributed to unchanged drug. In rats,
`the major metabolite was the sulfate conjugate, while in dogs,
`de-acetylation of the hydroxyacetyl group was observed.
`
`In safety, tolerance, and pharmacokinetic studies of single
`oral doses of U-100592 in healthy male volunteers, six
`subjects were dosed at each level of 50, 100, 200, 400, 700,
`1000, 1500, and 2000 mg [61]. The drug was well tolerated at all
`dose levels. A total of four subjects at the two highest doses had
`diarrhea; otherwise, no effects attributable to drug were seen in
`laboratory safety tests. The Cmax and area under the plasma
`concentration-time curve [AUC] values
`increased in an
`approximately proportional manner with increasing dose, with
`Cmax values of 0.16 µg/mL (50 mg), 5.73 µg/mL (1000 mg) and
`9.78 µg/mL (2000 mg). Elimination t112 ranged from 2.4 h (50
`
`For U-100766, po administration at 10 mg/kg b.i.d. gave
`Cmax values of 8 µg/mL, and ca. 35 µg/mL at a dose of 62.5
`mg/kg b.i.d. in the rat, while in the dog, 10 mg/kg b.i.d. gave
`Cmax values of ca. 6.6 µg/mL, and ca. 29 µg/mL at a dose of 40
`mg/kg b.i.d. [62].
`
`U-100480
`
`thiomorpholinyl bioisostere of U-100766, U -
`The
`100480 [72,87] (50 mg/kg po, t112=0 .. 7 h, bioavailability of
`31 ± 9 % ) was rapidly converted in rats to the sulfoxide U -
`101603, as the major metabolite, and the sulfone U-101244,
`as the minor metabolite [72]. At the above dose, the Cmax of U -
`100480 was 2 µg/mL and 7 µg/mL for the sulfoxide metabolite,
`a compound which also possessed potent activity against M.
`tuberculosis (vide supra).
`
`Toxicity Studies
`
`(±)-DuP 721
`
`Piper et al.9 at Upjohn found (±)-DuP 721 to exhibit lethal
`toxicity in rats dosed po at 100 mg/kg/day b.i.d. for 30 days
`[63,73]. The toxic effects were clinically manifested as severe
`anorexia (cachexia) in all animals, with severe morbidity and
`death observed. Histopathologic examination indicated evidence
`of bone marrow toxicity. Most of the adverse effects were
`considered to be related to severe nonspecific stress, with
`terminal circulatory failure.
`
`Pyridyl-Substituted Oxazolidinones
`
`A similar chronic toxicity profile to that seen with (±)-DuP
`721 was observed in rat studies with the 4'-(3"-pyridyl) tricyclic
`fused oxazolidinone U-92300 [63], (a compound with close
`structural analogy with E-3709 [126]). U-92300 was poorly
`tolerated in rats, when tested at 10 times the ED50.9
`
`

`
`Oxazolidinone Antibacterial
`
`Current Pharmaceutical Design, 1996, Vol. 2, No. 2 181
`
`E-3709
`
`U-100592 and U-100766
`
`The Upjohn clinical candidates U-100592 and U-100766
`demonstrated acceptable safety profiles in studies conducted on
`rats, dogs, and humans, allowing the drug development process
`to planned Phase II clinical studies (see
`to proceed
`pharmacokinetic section). The no-observed-adverse-effect-level
`in rats and dogs, dosed po for one month, was found to be 20
`mg/kg/day for U-100766 [62], and 25 mg/kg for U-100592
`[60]. Only mild adverse effects were observed upon dosing rats
`and dogs at 80 mg/kg/day for U-100592, and 50 mg/kg/day
`(rats) and 40 mg/kg/day (dogs) for U-100766.
`
`U-100480 and U-97456
`
`One-month drug safety studies in the rat9 of two Upjohn
`drugs having excellent anti-tuberculosis activity showed a
`favorable chronic toxicity profile, similar to the above drug
`candidates. The optically active thiomorpholine U-100480
`and indoline U-97456 [63] oxazolidinones were very well
`tolerated in rats (po, 50 mg/kg b.i.d.), with all drug-related
`findings considered of minor toxicological relevance.
`
`Synthetic Routes
`0xazolidino