A derivative of the thiopeptide GE2270A highly selective against propionibacterium acnes

Article abstract of DOI:10.1128/AAC.05155-14

A chemical derivative of the thiopeptide GE2270A, designated NAI003, was found to possess a substantially reduced antibacterial spectrum in comparison to the parent compound, being active against just a few Gram-positive bacteria. In particular, NAI003 retained low MICs against all tested isolates of Propionibacterium acnes and, to a lesser extent, against Enterococcus faecalis. Furthermore, NAI003 showed a time- and dose-dependent killing of both a clindamycin-resistant and a clindamycin-sensitive P. acnes isolate. Gel shift experiments indicated that, like the parent compound, NAI003 retained the ability to bind to elongation factors Tu (EF-Tus) derived from Escherichia coli, E. faecalis, or P. acnes, albeit with reduced efficiency. In contrast, EF-Tus derived from the NAI003-insensitive Staphylococcus aureus or Streptococcus pyogenes did not bind this compound. These results were confirmed by in vitro studies using a hybrid translation system, which indicated that NAI003 can inhibit most efficiently protein synthesis driven by the P. acnes EF-Tu. P. acnes mutants resistant to NAI003 were isolated by direct plating. With one exception, all analyzed strains carried mutations in the tuf gene, encoding EF-Tu. Because of its selective effect on P. acnes in comparison to resident skin flora, NAI003 represents a promising candidate for the topical treatment of acne, which has already completed a phase 1 clinical study.

Full text of DOI:10.1128/AAC.05155-14

AAC Accepted Manuscript Posted Online 18 May 2015
Antimicrob. Agents Chemother. doi:10.1128/AAC.05155-14
Copyright © 2015, American Society for Microbiology. All Rights Reserved.
1
2
A derivative of the thiopeptide GE2270A highly selective
against Propionibacterium acnes
3
4
Attilio Fabbretti,a Cheng-Guang He,a Eleonora Gaspari,b Sonia Maffioli,b Letizia
Brandi,a Roberto Spurio,a Margherita Sosio,b Daniela Jabes,b Stefano Donadiob
5
6
Laboratory of Genetics, School of Biosciences and Veterinary Medicine, University of Camerino,
Camerino, Italya; Naicons Srl, Milano, Italyb
7
8
9
10
Address correspondence to: Attilio Fabbretti, attilio.fabbretti@unicam.it; Stefano Donadio,
sdonadio@naicons.com.
11
12 Abstract
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
A chemical derivative of the thiopeptide GE2270A, designated NAI003, was found to possess a
substantially reduced antibacterial spectrum in comparison to the parent compound, being active
against just a few Gram-positive bacteria. In particular, NAI003 retained low MICs against all
tested isolates of Propionibacterium acnes and, to a lesser extent, against Enterococcus faecalis.
Furthermore, NAI003 showed a time- and dose-dependent killing of both a clindamycin-resistant
and a clindamycin-sensitive P. acnes isolate. Gel-shift experiments indicated that, like the parent
compound, NAI003 retained the ability to bind to elongation factors (EF) Tu derived from
Escherichia coli, E. faecalis or P. acnes, albeit with reduced efficiency. In contrast, EF-Tu's derived
from the NAI003-insenstitive Staphylococcus aureus or Streptococcus pyogenes did not bind this
compound. These results were confirmed by in vitro studies using a hybrid translation system,
which indicated that NAI003 can inhibit most efficiently protein synthesis driven by the P. acnes
EF-Tu. P. acnes mutants resistant to NAI003 were isolated by direct plating. With one exception,
all analyzed strains carried mutation in the tuf gene, encoding EF-Tu. Because of its selective effect
on P. acnes in comparison to resident skin flora, NAI003 represents a promising candidate for the
topical treatment of acne, which has already completed a Phase 1 clinical study.
28
29
30
Key words
acne vulgaris; antibiotic; elongation factor Tu; resistant mutants
1

31
2

32 INTRODUCTION
33
34
35
36
37
38
39
Acne vulgaris, a complex disease of the pilosebaceous units of the face and upper trunk, is the most
common skin condition seen by physicians. Although it affects almost 100% of adolescents to
varying degrees and generally wanes as adolescence ends, the disease may persist into adulthood
and it has been estimated that more than seventeen million people in US are affected by acne (1, 2).
At least four factors contribute to the development of acne: increased sebum production by the
sebaceous glands; follicular hyperkeratinization; colonization of the sebaceous follicles with
Propionibacterium acnes; and inflammation (3, 4).
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
Inflammatory acne is the result of the host response to P. acnes, a pleomorphic, anaerobic rod
belonging to the phylum Actinobacteria. P. acnes, a component of the normal skin flora, is a
usually harmless commensal largely incapable of tissue invasion or serious infection. The organism
catabolizes sebaceous triglycerides, using the glycerol moiety as a carbon source without
catabolizing the fatty acids. While this species has been advocated to play a fundamental role in the
aetiogenesis of acne (5, 6), a significant controversy remains as to how important a role P. acnes
plays (4, 7). Recent metagenomic studies have confirmed it to be the major inhabitant of
pilosebaceous units (8). Nonetheless, P. acnes is a normal skin commensal not necessarily
associated with acne, although a correlation between disease and P. acnes phylotypes has been
proposed (8). It is also worth pointing out that Staphylococcus epidermidis can produce catabolites
that inhibit P. acnes growth (9), highlighting the importance of an appropriate balance of the
microflora for a healthy skin. On the basis of the above considerations, a reduction in the population
of P. acnes without affecting other commensal flora, may constitute a plausible approach to an
effective therapy of acne. However, no antibacterial agent possessing such an antibacterial spectrum
has been described so far.
55
56
57
58
59
60
61
62
The thiopeptide GE2270A 1, produced by the actinomycete Planobispora rosea, is extremely active
against numerous Gram-positive pathogens including methicillin-resistant strains of Staphylococcus
aureus and vancomycin-resistant Enterococcus spp. (10). It has been demonstrated that GE2270A
binds to domain II of elongation factor (EF) Tu, making contacts with residues 215-230, 256-264
and 273-277 (11). These interactions alter the conformation of EF-Tu, so as to increase its
electrophoretic mobility (12) and to inhibit EF-Tu-GTP-aminoacyl-tRNA (aa-tRNA) ternary
complex formation. As a result, GE2270A interferes selectively with the functioning of the
elongation factor in protein synthesis.
63
64
During a program aimed at generating analogs of GE2270A by semi-synthesis, we observed that
one of the derivatives, designated NAI003 2, showed a surprisingly restricted antibacterial spectrum,
3

65
66
67
mostly limited to P. acnes. Here we report the antibacterial properties of this compound, investigate
the molecular basis for its reduced activity against other Gram-positives and provide evidence that
NAI003 targets EF-Tu in P. acnes.
68 MATERIALS AND METHODS
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
Chemistry. GE2270A (5g, 3.87 mmol), prepared as described (10), was dissolved in
dioxane:water:formic acid 10:1:1 and left overnight at 80°C under stirring. After cooling, the
solvent was evaporated and concentrated to dryness, and the resulting residue was incubated for 1h
at room temperature in 0.5 M sodium carbonate. The reaction mixture was then diluted with cold
water and brought to pH 6.5 with HCl. The thiopeptides were extracted from the aqueous phase
with ethyl acetate and then precipitated from the concentrated organic phase by adding petroleum
ether. The resulting solid (1.3 g, 1 mmol) was dissolved in DMF (10 ml), and 4-amino-N-
benzylpiperidine (1.2 mmol), TEA (1.4 mmol) and di-phenylphosphorazidate (1.2 mmol) were
added at 0°C. The temperature was allowed to rise to room temperature and stirring was continued
for about 4 h. The reaction mixture was acidified to pH 3 with 1 N HCl and then diluted with water
to complete precipitation of the product. The wet solid was dried in air and then purified by flash
chromatography (silica gel 60, 230-400 mesh ASTM – Merck) eluting with 3-5% methanol in
chloroform. Fraction containing NAI003 were pooled and evaporated to dryness a white solid. 1H-
NMR dmso-d6 0.86 (d, 3H); 0.90 (d, 3H); 1.91-1.70 (n, 2H); 2.26-2.05 (m, 2H); 2.60 (s, 3H); 2.91-
2.69 (m, 4H); 3.40 (s, 3H); 3.51 (br s, 2H); 3.95-3.75 (m, 2H); 4.30 (dd, 1H); 4.99 (s, 2H); 7.41-
7.18 (m, 2H); 8.28 (s, 1H); 8.45 (s, 1H); 8.66 (s, 1H).
85
86
87
88
89
90
91
92
93
94
95
Bacterial strains and media. Bacterial strains were from public collections (ATCC) or from the
Naicons' collection of pathogenic strains (designated by L or ND prefixes). In particular, the P.
acnes strains with an 'L' or 'ND' prefix were independent clinical isolates collected before 2003 and
in 2011, respectively. Brucella Broth (BB) and Brucella Agar (BA), Middlebrook 7H9 Broth
(7H9B), Müller Hinton Broth (MHB), Todd Hewitt Broth (THB) and Wilking Chalgren Agar
(WCA) were from Difco Laboratories (Detroit, Mi, USA). Cation-adjusted MHB (CAMHB),
prepared by adding 20 mg/ml CaCl2 and 10 mg/ml MgCl2 to MHB, was used for all aerobic bacteria
except for: streptococci, which were grown in THB; Corynebacterium sp., which were grown in
CAMHB with 5% lysed horse blood; and Mycobacterium smegmatis, which was grown in 7H9B
supplemented with 0.5 g/l Tween80 and 10% OADC enrichment (BD, MD, USA). P. acnes was
grown in BB or BA supplemented with 5% lysed horse blood, 5 µg/ml hemin and 1 µg/ml vitamin
4

96
97
K1 (Supplemented BB). All liquid media contained 0.02% bovine serum albumin, unless otherwise
noted. All cultures were grown at 37ºC.
98
99
Determination of Minimal Inhibitory Concentrations (MIC). MICs were determined by the
broth microdilution methodology (13, 14); bacteria were inoculated at 5x105 CFU/ml. Assays were
performed in sterile 96-well microtiter plates with round bottom wells. Plates were read after 20-24
h (aerobic bacteria), or 72 h (Mycobacterium smegmatis; P. acnes, anaerobic atmosphere).
GE2270A and its derivatives were dissolved in dimethylsulfoxide. Clindamycin and erythromycin
(Sigma Aldrich) were dissolved in water and in 95% ethanol, respectively. Appropriate dilutions
were made with the required culture medium immediately before testing.
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
Time kill assays. P. acnes colonies from 72-h BA plates were resuspended in BB at approximately 1-
2x105 CFU/ml and incubated for 24 h in Supplemented BB supplemented. Then, 0.5 ml of the culture
were added to fresh 5 ml Supplemented BB in glass vials with plastic caps containing the desired
concentration of antibiotic. For this experiment, a series of antibiotic solutions were made at 100x the
desired final concentration in either DMSO (for NAI003) or water (for clindamycin). In all
experiments, an extra vial, additioned with just 50 µl DMSO, was used as growth control. Cultures
were incubated at 37ºC, and anaerobic conditions were achieved with 1:25 v/v Oxyrase (Oxyrase Inc.
Mansfield, OH, USA). At different time points, 0.2 ml aliquots were withdrawn, serially diluted 1:10
with 0.9% NaCl, and 25 µl of each dilution were spread onto two-four BA plates. Colonies were
counted after 48 – 72 h at 37°C under anaerobic atmosphere and average values used in time kill
curves (variations in colony numbers were usually within ±30% of average counts).
116
117
118
119
120
121
122
123
124
125
Resistance analysis. The occurrence of spontaneous resistant mutants of P. acnes ATCC 6922 was
determined by plating 0.1 ml of a stationary phase culture on WCA containing NAI003 or GE2270A
at 1 or 10 µg/ml. Plates were incubated in anaerobic atmosphere, with colonies scored after 5 and 12
days. Colonies emerging on the plates were cultured in BB, centrifuged and stored at -80ºC in
Nutrient Broth supplemented with 20% glycerol for MIC determination and analysis of the tuf genes,
encoding EF-Tu. From selected strains, a cell lysate was made by heating at 95ºC for 5 min and used
for PCR amplifying the entire tuf gene with primers 5’-GTGGCAAAGGCCAAGTTCG-3’ and 5’-
TCACTTGATGATCTTGGTGACTC-3’. PCR products were directly sequenced (Primm Srl, Milano,
Italy). The sequenced segment from the parental strain ATCC 6922 was found to be identical to the tuf
gene from P. acnes KPA171202 (15; GenBank NC_006085.1)
126
127
128
Biochemical methods. Buffers used were the following. Buffer A: 25 mM Tris-HCl pH 7.1, 5%
glycerol, 700 mM NaCl, 6 mM β-mercaptoethanol, 0.1 mM PMSF, 0.1 mM benzamidine. Buffer B:
25 mM Tris-HCl pH 7.1, 300 mM NaCl, 5% glycerol, 20 mM imidazole, 6 mM β-mercaptoethanol,
5

129
130
131
132
0.1 mM PMSF and 0.1 mM benzamidine. Buffer C: 25 mM Tris-HCl, pH 7.1, 300 mM NaCl, 5%
glycerol, 300 mM imidazole, 6 mM β-mercaptoethanol, 0.1 mM PMSF, 0.1 mM benzamidine. Buffer
D: 20 mM Tris-HCl, pH 7.1, 100 mM NH4Cl, 5% glycerol, 6 mM β-mercaptoethanol, 0.1 mM EDTA.
Buffer E: 10 mM Tris-HCl pH 7.7,10 mM Mg acetate, 60 mM NH4Cl.
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
Cloning, overexpression and purification of EF-Tu's. The strategy used to clone the tuf genes
from genomic DNA of S. aureus (ATCC 25923), E. faecalis (ATCC 29922), S. pyogenes (ATCC
700294) and P. acnes (ATCC 6919) was essentially as reported by Pham et al. (16). Briefly, four
different phosphorylated primers (Table 1) carrying NcoI and BamHI restriction sites were used in
two separate DNA amplification reactions. Equal volumes of the two PCR mixtures were combined
and subjected to 5-min denaturation at 95°C and 15-min renaturation at room temperature. Of the
resulting final products, 25% are expected to carry NcoI and BamHI cohesive ends, which can be
ligated with the appropriately restricted expression vector pETM11. The ligation products were then
transformed into E.coli BL21 strain.
E. coli BL21(DE3) pLysS cells harboring the pETM11 derivatives were grown in LB medium at
37°C up to an OD600 of 0.7, when 1 mM isopropyl-β-D-1-thiogalactopyranoside was added. After 4
h at 37°C, the cells were harvested by centrifugation at 5000 rpm for 10 min at 4°C and the pellet,
resuspended in Buffer A, was stored at -80°C. After thawing, the cells were lysed by sonication,
and the resulting cell extract was cleared by centrifugation at 13,000 rpm for 45 min at 4°C. The
cell extract was subsequently loaded on a 5- ml nickel-nitrilotriacetic acid chromatographic column
equilibrated in Buffer A. The column was initially washed with 10 bed volumes of Buffer A,
followed by an additional wash with 5 bed volumes of Buffer B. EF-Tu was eluted from the
column with Buffer C. Fractions containing EF-Tu were dialyzed to reduce salt concentration
against Buffer C containing 100 mM NaCl (without imidazole) and digested by TEV protease
(enzyme: substrate ratio 1:40) overnight at 20 °C, with 20% extra TEV protease added for
additional 3-4 h. The reaction mixture was again loaded on the Ni-NTA resin and the cleaved
protein, without the His-tag, was recovered in the flow-through of the column. EF-Tu was
concentrated by centrifugation at 4000 rpm for 60 min at 4°C in Amicon ultra-3 centrifugal filter
devices and, after dialysis at 4°C against Buffer D, it was stored at -80°C in small aliquots.
157
158
159
160
In vitro tests. 30S and 50S ribosomal subunits, 027IF2Cp(A) mRNA, EF-G and E. coli cell free
extract were prepared as described (17, 18). In vitro mRNA translation driven by 027IF2Cp(A)
mRNA was carried out as described (17). Native polyacrylamide (12%) gel electrophoresis of EF-
Tu pre-incubated with 1 mM GTP in the presence of increasing concentrations (0.1, 0.5, 1, 5, 10, 50
6

161
162
and 100 µM) of GE2270A or NAI003 was performed in Tris-Glycine buffer (pH 8.3) as described
(11).
163
164
165
166
167
168
169
Poly(U) hybrid translation system. Poly(U) translation was carried out in 30 µl of Buffer E
containing 3 mM phosphoenolpyruvate, 0.05 µg/ml pyruvate kinase, 1 mM GTP, 0.15 µg/μl
poly(U), 10 µM [3H]-Phe-tRNA and 0.2 µM of E. coli 30S, 50S and EF-G. Twenty pmol of purified
EF-Tu (from E. coli, S. aureus, E. faecalis, S. pyogenes or P. acnes) were added to the mix after a 5
min incubation at room temperature in the presence of increasing concentrations of GE2270A or
NAI003. After 30 min at 37°C the level of poly(U) translation was quantified from the amount of
acid-insoluble [3H]Phe-tRNA incorporated.
170
171 RESULTS
172
173
174
175
Synthesis of NAI003. The oxazolidine-linked L-Ser-L-Pro-NH2 side chain of GE2270A
undergoes, on acid treatment, an N-O acyl shift forming a diketopiperazine ester that can be easily
removed by mild basic treatment (19, 20). The newly formed carboxylic acid 2 can thus be
effectively amidated, as for example with 4-amino-N-benzylpiperidine to afford NAI003 (Fig. 1).
176
177
178
179
180
181
182
183
184
185
186
187
Antibacterial spectrum of NAI003. The parent compound GE2270A is highly active against
Gram-positive bacteria but not effective against Gram-negative species(10). Table 2 shows the
activity of GE2270A against Staphylococcus, Streptococcus and Enterococcus spp., with MICs in
the range of ≤0.015–0.25, 0.06–2, and 0.008–0.015 µg/ml, respectively. The acid derivative 2
maintains substantially the same spectrum as GE2270A, although MICs were 32 to 64 fold and 4 to
16 fold higher against staphylococci and streptococci, respectively (Table 2). Surprisingly, NAI003
had lost potency against most of the tested strains: measurable MICs were observed against
Enterococcus spp., with MICs in the 0.5–16 µg/ml range, and against Mycobacterium smegmatis.
Interestingly, in the latter case, compounds 1, 2 and 3 were found to be equally active, with MICs
of 4–8 µg/ml (Table 2). It should be noted that all four Corynebacterium strains tested in this
experiment were equally insensitive to the three compounds, suggesting that this genus may be
insensitive to this class of thiopeptides (Table 2).
188
189
190
191
192
We then wondered whether, among the pathogenic Actinobacteria, some species might have
retained sufficient sensitivity to NAI003. Preliminary experiments indicated that P. acnes was
sufficiently sensitive to this compound to warrant further investigation. The results of these
evaluations are reported in Table 3, where it can be seen that Propionibacterium spp., including P.
acnes, were highly sensitive to NAI003, with MIC values in the range of 4-250 ng/ml. In contrast to
7

193
194
195
196
197
most bacterial species of Table 2, P. acnes appear to be equally sensitive to all the three
thiopeptides tested in this study (Table 3). Among the tested P. acnes isolates, one strain was highly
resistant to clindamycin and erythromycin, an increasingly frequent phenotype among P. acnes
isolates (21). As expected from their different target, NAI003 and GE2270 retained activity against
this strain.
198
199
200
201
Killing effect on P. acnes. To our knowledge, a single study has been performed to evaluate the
effect of topical treatment on the number of P. acnes on the forehead of healthy volunteers: a
reduction of approximately one log of CFU was observed after one or two weeks of daily
applications of 1% clindamycin (22, 23).
202
203
204
205
206
207
208
209
210
211
212
We thus wondered whether NAI003 had any effect in vitro on the number of viable P. acnes cells in
comparison with clindamycin. To this end, we tested the effect of the two antibiotics when added to
exponentially-growing cells. The results are illustrated in Fig. 2A. Under the experimental
conditions used, P. acnes reached stationary phase (108 CFU/ml) after approximately 48 h, with an
apparent duplication time of 4-5 h. When these cultures were exposed to clindamycin at 1x, 10x or
100x the MIC, a slow decline in viable counts was observed during the first 48 h, followed by an
apparent faster decline afterwards, resulting in viable counts below the detection limit of the assay
(around 20 CFU/ml). In contrast, NAI003 killing rate was relatively constant with time, leading
after 6 days to a 2- and 3-log reduction in viable counts at 10x and 100x the MIC. In the presence of
1x the MIC, after an approximately 24-h lag, the strain was eventually able to grow, albeit at a
reduced rate (Fig. 2A).
213
214
215
216
217
218
219
220
221
222
A similar experiment was then performed with strain ND063311, which is resistant to clindamycin
(MIC=64 µg/ml) but equally sensitive to NAI003 (Table 2). While NAI003 could be used again at
1x, 10x and 100x the MIC, we could test clindamycin only at 1x (i.e. 64 µg/ml) and 4x (i.e. 256
µg/ml) the MIC. The results are shown in Fig. 2B. Neither clindamycin nor NAI003 at 1x the MIC
had a significant effect on the viable counts of the strain, but simply resulted in growth retardation
and one log lower CFU/ml at stationary phase. At higher concentrations, NAI003 was able to
decrease the number of viable counts in a dose-dependent manner, with again a 3-log reduction seen
after 6 days. Interestingly, the behaviour of NAI003 and clindamycin at intermediate concentrations
(10x and 4x the MIC, respectively) was similar, with some decrease in viable counts up to 3-4 days,
followed by a tendency of viable counts to bounce back (Fig. 2B).
223
224
In conclusion, these experiments indicate that NAI003 at 100x the MIC (i.e. 1.5 µg/ml) can
decrease the number of viable counts to a similar extent for both a clindamycin-sensitive (strain
8

225
226
227
228
ND062711) and a clindamycin-resistant (ND063111) strain. In contrast, killing by clindamycin can
occur against the sensitive strain at lower MIC multiples and concentrations (i.e. 0.06 µg/ml) than
for NAI003. However, with the clindamycin-resistant strain tested here, 256 µg/ml clindamycin
lead to a lower redction in viable counts than 1.5 µg/ml NAI003.
229
230
231
232
NAI003 binds to and inhibits the function of EF-Tu from Escherichia coli. To confirm that the
molecular target of NAI003 is EF-Tu, the assay described by Anborgh and Parmeggiani was
initially used (12). This test takes advantage of the different mobilities of free E. coli EF-Tu-GTP
and of the EF-Tu-GTP-GE2270A complex in native polyacrylamide gel electrophoresis (12, 24).
233
234
235
236
237
238
239
240
241
242
243
The results shown in Fig. 3 indicate that NAI003 is indeed able to alter the electrophoretic mobility
of E. coli EF-Tu, thereby confirming that this elongation factor is also the target of this molecule. It
is also interesting to note that the difference in the electrophoretic mobility between free and
antibiotic-bound EF-Tu is smaller for NAI003 than for GE2270A (Fig. 3B). This finding suggests
that the interaction of the two antibiotics with EF-Tu induces different conformational changes of
the elongation factor. from the gel-shift experiments it can be surmised that low concentrations (0.5-
1 μM) of GE2270A increase the mobility of more than 50% of EF-Tu, while greater concentrations
(approx. 10 μM) of the derivative are required to shift the same amount of protein. In addition, from
the gel-shift experiments it can be observed that low concentrations (0.5-1 µM) of GE2270A
increase the mobility of EF-Tu by more than 50%, while greater concentrations (approx. 10 μM) of
the derivative are required to shift the same amount of protein.
244
245
246
247
248
249
GE2270A is known to inhibit in vitro protein synthesis in an E. coli-based cell-free system. When
the effect of NAI003 was compared to that of GE2270A in an E. coli in vitro translation system
programmed with the 027-IF2Cp(A) mRNA (17), both antibiotics proved to interfere with protein
synthesis but to a different extent: whereas GE2270A caused 50% inhibition of protein synthesis at
10 µM, a 10-fold higher concentration of NAI003 was necessary to achieve 25-30% inhibition only
(Fig. 3C).
250
251
252
253
254
255
From the results of the in vitro translation assays and from those obtained with the gel shift
experiments, it is possible to deduce that both GE2270A and NAI003 form a stable interaction with
elongation factor EF-Tu that causes an altered electrophoretic mobility of the protein in a native gel.
However, the chemical modification of GE2270A which gives rise to NAI003 reduces by almost
one order of magnitude the affinity of the latter antibiotic for EF-Tu and also reduces its efficiency
as a protein synthesis inhibitor in E. coli (Fig. 3C).
9

256
257
258
259
260
261
262
263
264
265
266
Effect of NAI003 on different EF-Tu's. To understand whether the difference in the antibacterial
spectrum of the GE2270A and NAI003 might correlate to differences in the affinity for the
corresponding EF-Tu's, in vitro experiments were carried out with EF-Tu purified from various
organisms, namely the NAI003-insenstive S. aureus and S. pyogenes, the moderately sensitive E.
faecalis and the highly sensitive P. acnes. To do so, the tuf genes encoding the EF-Tu's from these
organisms were amplified and cloned in the expression vector pETM-11, following essentially the
method described by Pham et al. (16). This procedure and the presence of an N-terminal oligo-
histidine tail afforded sufficient amounts of protein from the four microorganisms after affinity
chromatography. After purification, the oligo-histidine tail was removed by treatment with the TEV
protease and the affinities of these factors for NAI003 and for GE2270A were compared by gel shift
assays as described above.
267
268
269
270
271
272
As seen in Fig. 4, the different EF-Tu's bind GE2270A with approximately the same affinity, at
least within the resolution limits of the experimental technique used, but display appreciable
differences in their capacity to bind NAI003. Indeed, whereas the proteins from E. coli and P. acnes
can clearly bind NAI003, as judged from their different electrophoretic mobilities in the presence of
increasing concentrations of the antibiotic, the same behavior was not observed for the factors from
S. aureus and S. pyogenes, whose mobilities are hardly affected by the presence of NAI003.
273
274
275
276
277
278
279
280
281
282
283
284
285
286
The electrophoretic assay used in the experiments shown above yields qualitative but not
sufficiently quantitative data to permit an estimation of the binding constants of the antibiotics for
the various types of EF-Tu. In fact, the altered mobility is caused by the net electrical charge
displayed on the protein surface as a result of antibiotic-induced conformational. Different ligands
can induce different changes on EF-Tu and some of these alterations may not trigger variations of
the electrical charge. Furthermore, the lack of a mobility shift does not constitute an absolute
evidence for the lack of an interaction between the elongation factors and the antibiotic. For these
reasons, in vitro translation tests were carried out to detect a possible inhibition of EF-Tu function.
Since EF-Tu is involved in the elongation stage of protein synthesis, these experiments were
performed in a poly(U)-dependent polyPhe synthesis test, a translational assay that does not depend
upon the complex initiation pathway of protein synthesis (17). The tests were carried out using a
hybrid system in which ribosomes and EF-G purified from E. coli were incubated with [3H]Phe-
tRNA and a poly(U) template in the presence of an appropriate energy regenerating system and of
the appropriate EF-Tu protein.
287
288
When increasing amounts of EF-Tu originating from the selected bacteria were added to the
heterologous cell-free system, the response was similar to that observed upon the addition of the
10

289
290
291
292
293
294
295
296
297
homologous E. coli EF-Tu. The addition of 10 pmol of E. coli EF-Tu promoted the incorporation of
∼ 90 pmol of [3H]Phe in a hot-acid insoluble product while the incorporation promoted by the same
amount of EF-Tu from P. acnes, S. aureus and E. faecalis was ∼ 40 pmol. The only exception was
found with S. pyogenes EF-Tu whose efficiency was about 15% of that of E. coli (Fig. 5A). These
preliminary data indicate that the EF-Tu from the selected bacteria are compatible with the E. coli
components in the heterologous translation system. In turn, these data indicate that this type of test
can be used to identify EF-Tu inhibitors, even in the case of the less efficient S. pyogenes EF-Tu-
dependent system, and to compare the inhibition by GE2270A and by NAI003 on the different EF-
Tu's.
298
299
300
301
302
303
304
305
306
307
308
309
310
311
The graph presented in Fig. 5B shows that protein synthesis was fully inhibited by GE2270A
independent of the source of EF-Tu, with an IC50 between 0.2 and 0.5 µM. With the exception of
translation in the presence of S. pyogenes EF-Tu, at ~1 µM GE2270A inhibition reached 100%
(Fig. 5B). The results obtained in the same test with NAI003 indicate that at ~1 µM this antibiotic
caused ~25-30% inhibition when translation depended upon E. coli or E. faecalis EF-Tu, but had no
effect on the systems containing S. aureus or S. pyogenes EF-Tu, which are only slightly inhibited
at very high NAI003 concentrations (Fig. 5C). On the other hand, the translational systems
dependent on P. acnes EF-Tu prove to be the most sensitive to NAI003 inhibition, translation being
reduced by >50% at ~1 µM NAI003 (Fig. 5C). These findings are overall consistent with the
conclusions drawn from the data presented in Fig. 4 and allow for the conclusion that the
substitution of the C-terminal moiety in GE2270A with the 4-amino-N-benzylpiperidine moiety
present in NAI003 results in a weaker or different interaction of the latter molecule with EF-Tu and
a general reduction of its inhibitory activity. However, this same chemical modification resulted in a
molecule endowed with an increased selectivity in the inhibition of EF-Tu from P. acnes.
312
313
314
315
316
Resistance to NAI003 is due to mutations in P. acnes EF-Tu. The frequency of resistance to this
class of thiopeptides was evaluated in parallel experiments with NAI003 and GE2270A. When P.
acnes was plated on media containing 1 or 10 μg/ml antibiotic, colonies appeared at frequencies of
6.3x10-11 or 3.1x10-11, respectively, for GE2270A, and at frequencies of 1.5x10-10 and 9.4x10-11
,
respectively, for NAI003.
317
318
319
320
321
Twelve independent NAI003R isolates were evaluated and characterized for the possible presence of
mutations in the tuf gene, encoding EF-Tu, as well as for their susceptibility to GE2270A, NAI003
and compound 2. With one exception, all resistant colonies harbored one mutation in the tuf gene
for a total of 8 independent changes at 6 different positions (Table 4). Some of the observed
mutations (G260R and G260C) occur at a residue (G257 in E. coli and B. subtilis EF-Tu)
11

322
323
324
325
326
327
328
329
330
previously reported to confer GE2270A resistance through the G257S substitution (11, 25, 26).
Interestingly, in the 3D structure of the complex between T. termophilus EF-Tu, GTP and
GE2270A, a Gln residue (equivalent to Q99 in the P. acnes protein) makes direct contacts with the
antibiotic. Consistenly, three independent P. acnes isolates carry the Q99H mutation. Finally, two
independent mutations (N276K and N276S) were observed at an Asn residue (equivalent to N273 in
the E. coli protein) that also makes direct contacts with the antibiotic (11). When transposed to the
crystal structure of T. termophilus EF-Tu, the other observed mutations (Table 4) map to domains I
(H68Q and V81A) or II (M263T), although they do not appear to be directly involved in antibiotic
binding.
331
332
333
334
335
336
337
338
339
340
In terms of resistance levels, some mutations (i.e. V81A, G260C and G260R) confer elevate
protection to NAI003, while mutant strains harboring the other mutations exhibit intermediate
resistance (Table 4). Remarkably, despite that the different mutations afforded an increased
resistance to GE2270 or to compound 2 with respect to the parental strain, none of them was
sufficient to confer resistance levels equivanet to those achieved against NAI003 (Table 3). It
should be noted, however, that we sequence the tuf gene only and the existence of additional
mutations elsewhere in the genome cannot be ruled out. Indeed, one mutant (L1015R86) did not
show any mutation in the tuf gene, despite its high level of resistance to NAI003 (Table 4). Thus,
further studies will be necessary to understand the full range of mutations conferring resistance to
NAI003.
341
342 DISCUSSION
343
344
345
346
347
348
349
The results presented here demonstrate that NAI003 binds to the EF-Tu from P. acnes with an
affinity comparable to that of GE2270A and that its binding alters the electrophoretic mobility to
the same extent as that observed with GE2270A. Furthermore, the translation test performed
demonstrates that NAI003 inhibits in vitro P. acnes EF-Tu-dependent translation to a level similar
to that obtained with GE2270. That EF-Tu is the target of NAI003 in vivo is confirmed by the
observation that most spontaneous resistant mutants of P. acnes carry mutations in the tuf gene,
corresponding to residues already established to confer resistance to or to bind GE2270A (11).
350
351
352
353
GE2270A is a better inhibitor of P. acnes EF-Tu than its derivative NAI003. Consistently,
GE2270A MICs tend to be lower than those of NAI003. Nonetheless, we were surprised to observe
that mutations in EF-Tu that confer complete resistance to NAI003 in vivo, some of which
corresponding to residues previously demonstrated to be critical for GE2270A binding, only afford
12

354
355
partial protection to GE2270A in P. acnes. Further studies will be necessary to explain this
observation.
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
We have demonstrated that the lack of sensitivity of many Gram-positive bacteria to NAI003 is
correlated to a lower inhibition of their EF-Tu's by this antibiotic. In this respect, the observed
extent of EF-Tu inhibition (Fig. 5C) well correlates with the MIC values for the cognate
microorganism (Table 1). It should be noted that GE2270A is able to effectively inhibit all EF-Tu's
tested so far, with natural resistance documented only in the producer strain Planobispora rosea (25)
and in the Corynebacterium species tested herein (Table 2). This is consistent with the fact that EF-
Tu is a highly conserved protein with limited variation among divergent bacterial taxa. Despite
these features, GE2270A has been converted by a relatively simple chemical modification into a
derivative with a dramatically reduced antibacterial spectrum. This has resulted by the fortuitous
occurrence of two factors: the highly reduced susceptibility of EF-Tu's from the three Firmicutes
specis analyzed herein; and the lack of permeability of E. coli (and presumably of other Gram-
negative bacteria) to this class of thiopeptides, which protects an otherwise partially sensitive EF-Tu.
In light of these results it would be interesting to conduct further experiments aimed at elucidating
the 3D structure of the P. acnes EF-Tu complexed with NAI003 and compare it to the EF-Tu–
GE2270A complex. This would provide insights into the nature of the different EF-Tu
conformations induced by the two antibiotics and enable predictive models of which bacterial
strains may be sensible to NAI003 or to other GE2270A derivatives.
373
374
375
376
377
378
NAI003 was found to be equally active against 21 independent P. acnes, including one highly
resistant to clindamycin and erythromycin (Table 3). Additional testing confirmed that this
GE2270A derivative retains activity against several independent clindamycin- and erythromycin-
resistant P. acnes clinical isolates (D.J. et al., unpublished data). This observation is not surprising,
since clindamycin and erythromycin resistance in P. acnes mostly involves mutation in 23S RNA
(27–29), a site unrelated to the interaction of EF-Tu with the ribosome.
379
380
381
382
383
384
Antibacterial agents with an extremely narrow spectrum hold the promise of curing infections with
minimal disturbance of commensal flora. This might be particularly true for topical applications,
where a diverse and body-specific flora is expected to contribute to skin health (30-32). The
compound, previously designated BIK 0376, has already completed a clinical study in healthy
volunteers as a topical application (32, 33). A clinical study will be necessary to establish its
efficacy.
385
386
ACKNOWLEDGMENTS
13

387
388
389
390
This work was partially supported by a grant from Regione Lombardia (progetto ATP). We thank
Gabriella Romanò and Paola Guglierame for early contributions to this project, and Cristina Brunati
for performing the time kill assays. We are also grateful to Claudio O. Gualerzi for valuable advice
and insights into this work.
391
392
14

393
REFERENCES
394
395
1.
Bergfeld W. 1995. The evaluation and management of acne: economic considerations. J. Am. Acad.
Dermatol. 5:S52–S56.
396
397
398
2.
3.
4.
5.
Bhate K, Williams HC. 2013. Epidemiology of acne vulgaris Br. J. Dermatol. 168:474–485.
Johnson BA, Nunley JR. 2000. Topical therapy for acne vulgaris. Postgrad. Med. 107:69–80.
Shaheen B, Gonzalez M. 2013. Acne sans P. acnes. J. Eur. Acad Dermatol. Venereol. 27:1–10.
399
400
401
Webster GF, Leyden JJ, Musson RA, Douglas SD. 1985. Susceptibility of Propionibacterium
acnes to killing and degradation by human neutrophils and monocytes in vitro. Infect. Immun.
49:116–121.
402
403
404
6.
Beylot C, Auffret N, Poli F, Claudel JP, Leccia MT, Del Giudice P, Dreno B. 2013.
Propionibacterium acnes: an update on its role in the pathogenesis of acne. J. Eur. Acad. Dermatol.
Venereol. 28:271–2788.
405
406
7.
8.
Dessinioti C, Katsambas AD. 2010. The role of Propionibacterium acnes in acne pathogenesis:
facts and controversies. Clin. Dermatol. 28:2-7.
407
408
409
410
Fitz-Gibbon S, Tomida S, Chiu BH, Nguyen L, Du C, Liu M, Elashoff D, Erfe MC, Loncaric A,
Kim J, Modlin RL, Miller JF, Sodergren E, Craft N, Weinstock GM, Li H. 2013.
Propionibacterium acnes strain populations in the human skin microbiome associated with acne. J.
Invest. Dermatol. 133:2152–2160.
411
412
413
414
9.
Wang Y, Kuo S, Shu M, Yu J, Huang S, Dai A, Two A, Gallo RL, Huang CM. 2014.
Staphylococcus epidermidis in the human skin microbiome mediates fermentation to inhibit the
growth of Propionibacterium acnes: implications of probiotics in acne vulgaris. Appl. Microbiol.
Biotechnol. 98:411–424.
415
416
10. Goldstein BP, Berti M, Ripamonti F, Resconi A, Scotti R, Denaro M. 1993. In vitro antimicrobial
activity of a new antibiotic, MDL 62,879 (GE2270 A). Antimicrob. Agents Chemother. 37:741–745.
417
418
11. Parmeggiani A, Krab IM, Okamura S, Nielsen RC, Nyborg J, Nissen P. 2006. Structural basis of
the action of pulvomycin and GE2270 A on elongation factor Tu. Biochemistry. 45:6846–6857.
419
420
12. Anborgh PH, Parmeggiani A. 1991. New antibiotic that acts specifically on the GTP-bound form
of elongation factor Tu. EMBO J. 10:779–784.
421
422
423
13. National Committee for Clinical and Laboratory Standards. 2012 Methods for dilution
antimicrobial susceptibility tests for bacteria that grow aerobically - Ninth Edition: Approved
Standard M7-A9, NCCLS, Wayne, PA.
424
425
426
14. National Committee for Clinical and Laboratory Standards. 2012 Performance standards for
antimicrobial susceptibility testing - TwentySecond Informational Supplement; Approved Standards
M100-S22 NCCLS, Wayne, PA.
427
428
429
15. Brüggemann H, Henne A, Hoster F, Liesegang H, Wiezer A, Strittmatter A, Hujer S, Dürre P,
Gottschalk G. 2004. The complete genome sequence of Propionibacterium acnes, a commensal of
human skin. Science. 305:671–673.
430
431
16. Pham K, LaForge KS, Kreek MJ. 1998. Sticky-end PCR: New method for subcloning.
BioTechniques 25:206–208.
432
433
17. Brandi L, Fabbretti A, Milon P, Carotti M, Pon CL, Gualerzi CO. 2007. Methods for identifying
compounds that specifically target translation. Methods Enzymol. 431:229–267.
434
435
436
18. Milon P, Konevega AL, Peske F, Fabbretti A, Gualerzi CO, Rodnina MV. 2007. Transient
kinetics, fluorescence, and FRET in studies of initiation of translation in bacteria. Methods Enzymol.
430:1–30.
437
438
19. Tavecchia P, Gentili P, Kurz M, Sottani C, Bonfichi R, Selva E, Lociuro S, Restelli E, Ciabatti
R. 1995. Degradation studies of antibiotic MDL 62,879 (GE2270A) and revision of the structure
15

439
Tetrahedron. 51:4867–4890.
440
441
442
20. LaMarche MJ, Leeds JA, Dzink-Fox J, Gunderson K, Krastel P, Memmert K, Patane MA,
Rann EM, Schmitt E, Tiamfook S, Wang B. 2011. 4-Aminothiazolyl analogues of GE2270A:
Antibacterial lead finding. J. Med. Chem. 54:2517–2521.
443
21. Humphrey S. 2012. Antibiotic resistance in acne treatment. Skin Therapy Lett. 17:1–3.
444
445
446
22. Leyden J, Kaidbey K, Levy SF. 2001. The combination formulation of clindamycin 1% plus
benzoyl peroxide 5% versus 3 different formulations of topical clindamycin alone in the reduction of
Propionibacterium acnes. An in vivo comparative study. Am. J. Clin. Dermatol. 2:263–266.
447
448
23. Gans EH, Kligman AM. 2002. Comparative efficacy of clindamycin and benzoyl peroxide: in-vivo
suppression of Propionibacterium acnes. J. Dermatol. Treat. 13:107–110.
449
450
451
24. Anborgh PH, Parmeggiani A. 1993. Probing the reactivity of the GTP- and GDP-bound
conformations of elongation factor Tu in complex with the antibiotic GE2270 A. J. Biol. Chem.
268:24622–24628.
452
453
454
25. Sosio M, Amati G, Cappellano C, Sarubbi E, Monti F, Donadio S. 1996. An elongation factor Tu
(EF-Tu) resistant to the EF-Tu inhibitor GE2270 in the producing organism Planobispora rosea.
Mol. Microbiol. 22:43–51.
455
456
457
458
26. Zuurmond AM, Martien de Graaf J, Olsthoorn-Tieleman LN, van Duyl BY, Mörhle VG,
Jurnak F, Mesters JR, Hilgenfeld R, Kraal B. 2000. GE2270A-resistant mutations in elongation
factor Tu allow productive aminoacyl-tRNA binding to EF-Tu.GTP.GE2270A complexes. J. Mol.
Biol. 304:995–1005.
459
460
461
27. Ross JI, Eady EA, Cove JH, Ratyal AH, Cunliffe WJ. 1997. Clinical resistance to erythromycin
and clindamycin in cutaneous propionibacteria is associated with mutations in 23S rRNA.
Antimicrob. Agents Chemother. 41:1162–1165.
462
463
464
465
28. Ross JI, Snelling AM, Eady EA, Cove JH, Cunliffe WJ, Leyden JJ, Collignon P, Dréno B,
Reynaud A, Fluhr J, Oshima S. 2001. Phenotypic and genotypic characterization of antibiotic-
resistant Propionibacterium acnes isolated from acne patients attending dermatology clinics in
Europe, the U.S.A., Japan and Australia. Br. J. Dermatol. 144:339–346.
466
467
468
29. Oprica C, Löfmark S, Lund B, Edlund C, Emtestam L, Nord CE. 2005. Genetic basis of
resistance in Propionibacterium acnes strains isolated from diverse types of infection in different
European countries. Anaerobe. 11:137–143.
469
30. Grice EA, Segre JA. 2011. The skin microbiome. Nat. Rev. Microbiol. 9:244–253.
470
471
472
31. Oh J, Byrd AL, Deming C, Conlan S; NISC Comparative Sequencing Program, Kong HH,
Segre JA. 2014. Biogeography and individuality shape function in the human skin metagenome.
Nature. 514:59–64.
473
474
32. Butler MS. 2005. Natural products to drugs: natural product derived compounds in clinical trials.
Nat. Prod. Rep. 22:162-195.
475
476
477
33. Donadio S, Maffioli S, Monciardini P, Sosio M, Jabes. D 2010. Sources of novel antibiotics--aside
the common roads. Appl. Microbiol. Biotechnol. 88:1261-1267.
16

478 FIGURE LEGENDS
479
Figure 1. Synthetic route to NAI003.
480
481
482
483
484
485
Figure 2. Killing kinetics of P. acnes. Effect of NAI003 (closed symbols, solid line) and
clindamycin (empty symbols, dashed line) on the viability of P. acnes, using the clindamycin-
sensitive ND062711 (panel A) and clindamycin-resistant ND06311 (panel B) isolates. Compounds
were added at 1X MIC (triangles), 10X MIC (circles), or 100X MIC (squares). In panel B only,
clindamycin was also used at 4X the MIC (i.e. 256 µg/ml). Growth controls are shown for both
panels by a thick dashed line.
486
487
488
489
490
491
492
493
Figure 3. Effect of GE2270A and NAI003 on the electrophoretic mobility of EF-Tu and on in
vitro translation. (A) Migration on native polyacrylamide gel of E. coli EF-Tu (pre-incubated with
GTP) in the presence of increasing concentrations (from left to right, 0.1, 0.5, 1, 5, 10, 50 and 100
µM) of GE2270A (lanes 2–8) or NAI003 (lanes 9–15). (B) Electrophoretic mobility difference
between EF-Tu-GTP alone (lane 1), in the presence of 1 µM GE2270A (lane 2) or of 10 µM
NAI003 (lane 3). The two arrows indicate the different migrations of EF-Tu in complex with
GE2270A or with NAI003. (C) Inhibition by GE2270A (●) or NAI003 (■) in a protein synthesis
system based on an E. coli extract programmed with 027-IF2Cpoly(A) mRNA.
494
495
496
497
Figure 4. Effect of GE2270A and NAI003 on the electrophoretic mobility of different EF-Tu's.
Migration on native polyacrylamide gels of EF-Tu from (A) E. coli, (B) S. aureus, (C) P. acnes, and
(D) S. pyogenes in the presence of increasing concentrations (1, 4, 19, 50 and 100 μM) of GE2270A
or of NAI003.
498
499
500
501
502
503
504
Figure 5. Effect of GE2270A and NAI003 on in vitro protein synthesis. (A) Poly(U) dependent
incorporation of [3H]Phe in a hot-TCA insoluble product in the presence of increasing
concentrations of different EF-Tu's The amount of Phe incorporated with E. coli EF-Tu are
indicated on the right ordinate while those obtained with EF-Tu from P. acnes, S. aureus, E. faecais,
or S. pyogenes are indicated on the left ordinate. Effect of GE2270A (B) and NAI003 (C) on in vitro
translation with different EF-Tu's. Labels for EF-Tu are as follows: E. coli (●); P. acnes (♦); E.
faecalis (■); S. aureus (▼); or S. pyogenes (▲).
505
17

506
Table 1. Oligonucleotide primers used for sticky-end PCR
tufA gene source
Sequence (5'- 3')
F: CATGGCAAAAGAAAAATTCGATCGTTC
R: CTTATTTAATGATTTCAGTAACAACGC
F: GCAAAAGAAAAATTCGATCGTTCTAAAG
R: GATCCTTATTTAATGATTTCAGTAACAACG
F: CATGGCAAAAGAAAAATTTGACCGTTC
R: CTTATTTAATGATTTCAGTAACAACGC
F: GCAAAAGAAAAATTTGACCGTTCTAAATC
R: GATCCTTATTTAATGATTTCAGTAACAACG
F: CATGGCAAAAGAAAAATACGATCGTAG
R: GATCCTTAAGCTTCGTTTCTGAAACG
F: GCAAAAGAAAAATACGATCGTAGTAAAC
R: CTTAAGCTTCGATTTCTGAAACGATACC
F: CATGGCAAAGGCCAAGTTCGAGCGG
R: GATCCTCACTTGATGATCTTGGTGACTCG
F: GCAAAGGCCAAGTTCGAGCGGACC
S. aureus
E. faecalis
S. pyogenes
P. acnes
R: CTCACTTGATGATCTTGGTGACTCGACC
507
508
509
18

510
511
Table 2. MICs (µg/ml) of GE2270A, compound 2 and NAI003 against selected Gram-positive
bacteria.a
Microorganism
GE2270A
≤0.015
≤0.015
≤0.015
0.03
Compound 2
NAI003
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
>128
0.5
Vancomycin
Staphylococcus aureus ATCC29213
Staphylococcus aureus ND041009
Staphylococcus aureus ND018107
Staphylococcus aureus ND018107
Staphylococcus aureus ND055710
Staphylococcus aureus ND055910
Staphylococcus epidermidis ND006607
Staphylococcus epidermidis ND015007
Staphylococcus epidermidis ND020708
Staphylococcus epidermidis ND021208
Staphylococcus epidermidis ND024908
Staphylococcus epidermidis ND025708
Staphylococcus epidermidis ND026508
Staphylococcus capitis ND004307
Staphylococcus capitis ND018807
Streptococcus agalactiae ND050210
Streptococcus agalactiae L310
0.5
0.5
1
1
1
0.25
0.5
0.5
1
1
0.03
1
0.03
0.5
1
≤0.015
0.125
0.03
1
2
2
0.5
2
2
0.03
2
≤0.015
0.03
1
2
1
2
0.06
4
1
0.03
2
2
0.25
8
2
2
>128
32
2
0.5
0.5
0.5
0.5
0.25
0.5
>128
2
2
Streptococcus pyogenes L1303
0.5
Streptococcus pyogenes L800
0.5
4
Streptococcus pneumoniae ATCC49619
Enterococcus faecalis VanS L559
Enterococcus faecalis VanA L560
Enterococcus faecium VanS L568
Enterococcus faecium VanA L569
Mycobacterium smegmatis mc2155
Micrococcus luteus L108
0.06
0.5
n.t.
n.t.
n.t.
n.t.
8
0.008
0.008
0.015
0.015
4
0.5
16
4
>128
n.t.
0.125
1
4
≤0.015
>128
>128
>128
>128
1
>128
>128
>128
>128
>128
Corynebacterium jeikeium L2508
Corynebacterium jeikeium L2645
Corynebacterium sp. L689
>128
>128
>128
>128
0.5
1
Corynebacterium sp. L691
0.5
512
513
a n.t., not tested
19

514
Table 3. MICs (µg/ml) of GE2270A, compound 2 and NAI003 against P. acnes.a
Microorganism
GE2270A
compound 2
NAI003
Erythromycin Clindamycin
Propionibacterium acnes ND062411
Propionibacterium acnes ND062511
Propionibacterium acnes ND062611
Propionibacterium acnes ND062711
Propionibacterium acnes ND062811
Propionibacterium acnes ND062911
Propionibacterium acnes ND063011
Propionibacterium acnes ND063111
Propionibacterium acnes ND063211
Propionibacterium acnes ND063311
Propionibacterium acnes L627
≤0.0018
≤0.0018
≤0.0018
≤0.0018
≤0.0018
≤0.0018
≤0.0018
≤0.0018
≤0.0018
0.0009
≤0.015
0.008
0.03
0.06
0.06
0.06
0.06
0.03
0.03
0.06
0.03
0.03
n.t.
0.015
0.015
0.06
0.03
0.06
0.06
0.03
0.03
0.06
0.06
0.06
0.06
>128
n.t.
0.06
0.125
0.06
0.06
0.06
0.03
0.5
0.015
0.015
0.03
0.007
0.015
0.015
0.015
0.06
0.06
0.06
64
n.t.
Prppionibacterium acnes ATCC6919
Ppropionibacterium acnes ATCC06922
Propionibacterium acnes ATCC25746
Propionibacterium acnes L1559
Propionibacterium acnes L1560
Propionibacterium acnes L1561
Propionibacterium acnes L1562
Propionibacterium acnes L1563
Propionibacterium acnes L1564
Propionibacterium acnes L1565
n.t.
0.06
n.t.
n.t.
0.03
n.t.
0.13
n.t.
n.t.
0.03
n.t.
0.25
n.t.
n.t.
0.008
n.t.
0.06
n.t.
n.t.
0.015
n.t.
0.06
n.t.
n.t.
0.015
n.t.
0.06
n.t.
n.t.
0.008
n.t.
0.06
n.t.
n.t.
0.008
n.t.
0.06
n.t.
n.t.
0.015
n.t.
0.06
n.t.
n.t.
0.015
n.t.
0.06
n.t.
n.t.
515
516
a n.t., not tested
20

517
Table 4. Genotypes and phenotypes of P. acnes NAI003R mutants.
MIC (µg/mL)
NAI003 GE2270A
EF-Tu
EF-Tu
Strain code tuf mutationa
ATCC6922
substitutionb
domainc
compound 2
-
-
H68Q
V81A
Q99H
Q99H
Q99H
G260R
G260C
M263T
N276S
N276S
N276K
none
-
0.015
0.007
0.015
≤ 0.06
0.015
0.007
0.003
0.125
0.125
≤ 0.06
0.007
0.007
0.03
≤0.06
L1015R13 C207A
L1015R51 T245C
I
I
I
I
I
8
32
8
0.5
1
L1015R6
L1015R5
L1015R7
G300T
G300T
G300T
1
2
0.5
0.25
0.25
2
1
L1015R24 G781C
L1015R8 G781T
II
II
II
II
II
II
>128
32
4
L1015R96 T791C
L1015R73 A830G
L1015R72 A830G
L1015R21 C831G
4
1
0.125
0.25
0.25
4
2
1
L1015R86
-
128
0.25
none
518
519
520
521
522
a Nucleotide numbering is based on tuf sequence from P. acnes KPA171202 (15)
b Amino acid numbering is based on EF-Tu sequence from P. acnes KPA171202 (15)
c As defined by Parmeggiani et al. (11).
21