Rifamycin derivatives active against pathogenic rapidly-growing mycobacteria

Article abstract of DOI:10.1016/j.bmcl.2019.07.001

Infections due to rapidly growing mycobacteria (RGM), and in particular the RGM species Mycobacterium abscessus (Mab), are very difficult to treat and reports on novel therapeutic options are scarce. A hallmark of all pathogenic RGM species is their resis

Full text of DOI:10.1016/j.bmcl.2019.07.001

Accepted Manuscript
Rifamycin derivatives active against pathogenic rapidly-growing mycobacteria
Keith D. Combrink, Andrea Ramirez Ramos, Stephanie Spring, Sebastian
Schmidl, Kira Elizondo, Petronilo Morin, Bryant De Jesus, Florian P. Maurer
PII:
S0960-894X(19)30446-9
https://doi.org/10.1016/j.bmcl.2019.07.001
BMCL 26542
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
23 May 2019
28 June 2019
1 July 2019
Please cite this article as: Combrink, K.D., Ramos, A.R., Spring, S., Schmidl, S., Elizondo, K., Morin, P., De Jesus,
B., Maurer, F.P., Rifamycin derivatives active against pathogenic rapidly-growing mycobacteria, Bioorganic &
Medicinal Chemistry Letters (2019), doi: https://doi.org/10.1016/j.bmcl.2019.07.001
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Rifamycin derivatives active against pathogenic rapidly-growing mycobacteria
2
a,#
a
a
a,d
a
Keith D. Combrink , Andrea Ramirez Ramos , Stephanie Spring , Sebastian Schmidl , Kira Elizondo ,
a
a
b,c
Petronilo Morin , Bryant De Jesus , and Florian P. Maurer
a
Department of Biology and Chemistry, Texas A&M International University, Laredo, USA
b
National and WHO Supranational Reference Laboratory for Mycobacteria, Research Center Borstel,
Leibniz Lung Center, Borstel, Germany
c
University Medical Center Hamburg-Eppendorf, Hamburg, Germany
dPresent address: RELLIS campus, Texas A&M University, Bryan, TX, USA
#Corresponding author:
Keith Combrink
Department of Biology and Chemistry
Texas A&M International University
Laredo, TX, 78045 USA
keith.combrink@gmail.com
Abstract
Infections due to rapidly growing mycobacteria (RGM), and in particular the RGM species Mycobacterium
abscessus (Mab), are very difficult to treat and reports on novel therapeutic options are scarce. A hallmark
of all pathogenic RGM species is their resistance to the four first-line drugs used to treat infections with
Mycobacterium tuberculosis including rifampicin. This study demonstrates that modification of the
rifampicin scaffold can restore rifampicin activity against the three most commonly isolated pathogenic

RGM species including Mab. We also note that the structure-activity relationship for Mab is different as
compared to the non-pathogenic RGM species Mycobacterium smegmatis.
Main text
Rapidly growing mycobacteria (RGM) are increasingly recognized as important bacterial pathogens in
1
vulnerable patient populations. RGM infections can affect almost all organs, but are most commonly
2
linked to lung, skin and soft tissue. 80% of lung disease caused by RGM are attributed to organisms
2
belonging to the Mycobacterium abscessus (Mab) complex. The two other most frequently encountered
pathogenic RGM species are M. fortuitum (Mfo) and M. chelonae (Mch). Infections due to RGM, and in
particular Mab, are very difficult to treat and antibiotic therapy recommended by the American Thoracic
Society (ATS) gives response rates for treatment of Mab pulmonary disease below 60% as measured by
2
negative sputum cultures for more than 12 months. This unsatisfactory outcome is in part explained by
2-3
the extensive natural resistance of Mab to a wide variety of antimicrobials. In consequence, treatment
regimens are several months in duration with multiple antibiotics often resulting in frequent adverse
events.
While rifampicin (RMP, 1) is a cornerstone drug for treatment of infections caused by slowly growing
mycobacteria such as the M. tuberculosis (Mtb) complex, it has no activity against RGM.2-3 In contrast,
rifabutin (RBT), another member of the rifamycin class of antimicrobials, has been shown to have low
4
micromolar MICs against Mab as compared to RMP. Rifabutin does not appear to be a substrate for Arr
modification and we were unable to detect ribosylation of the C23 hydroxy group of rifabutin with Mab.
However, while there is less induction of hepatic microsomal enzymes with RBT than with RMP, elevated
transaminases have been reported with high-dose (600 mg/day) rifabutin treatment in combination with
macrolides and RBT. 5-6

Two mechanisms conferring rifamycin resistance have been described in mycobacteria. Most commonly,
resistance is conferred by mutations in the rpoB gene encoding the biochemical rifamycin target, the RNA
7-9
dependent polymerase subunit B. Genetic analysis of Mab type strain ATCC 19977 showed that the core
10
region of the rpoB genetic sequence is identical to that of rifampicin susceptible Mtb strains. The second
mechanism is enzymatic drug modification due to the addition of an ADP-ribosyl group to the C23 hydroxyl
group of the rifamycin scaffold (Fig. 1a-b). This is conferred by orthologues of the ADP-ribosyltransferase
Arr, which was first described in M. smegmatis (Msm) and other RGM.11-12 ADP-ribosylation is known to
13
play an important role in signaling and disease in all kingdoms of life. However, to our knowledge, ADP-
ribosylation of rifamycins is i) the only known example for modification of non-peptide substrates and ii)
the only example for ADP-ribosylation as a mechanism of antimicrobial resistance.
Figure 1: Structures of modified rifamycins. 1, rifampicin; 1a-b, ADP-ribosyl-rifampicin; 2, C25-carbamate
derivatives.
OH
HO
1
O
O
O
O
O
R
2
3
2
3
R
25
25
O
NH
O
O
OH
O
OH
25
OH
OH
OH
O
23
OH
O
O
O
O
OH OH
3
O
O
OH OH
NH
N
NH
3
NH
O
O
O
O
OH
O
N
O
O
O
N
O
OH
O
N
N
O
N
N
1
2
1
1
1
a, R = H; b, R = ADP

In a previous study with nonpathogenic RGM species M. smegmatis (Msm), we showed that addition of a
carbamate linked group of sufficient size at the C25 position of a rifamycin derivative 2 was able to block
enzymatic inactivation of RMP by ArrMsm.14 Recently, Rominski et al. prepared three of the compounds
10
from this study (2a, 2d and 2e) and evaluated them against Mab. The authors showed that the prepared
rifamycin carbamates had modest activity against the Mab type strain ATCC 19977. Interestingly, the
authors also showed that C25 modification not only increased rifamycin activity against the Mab wild type,
but also against a ΔarrMab mutant indicating that the increased activity of C25 rifamycin derivatives is only
partially due to resistance to modification by ArrMab. Lastly, the authors concluded that the studied RMP
derivatives were still partially inactivated by ArrMab. In our original work on Msm, we observed a similar
effect with our less active compounds.14 Here, we prepared seven compounds active against Msm from
our previous study and two additional new carbamate derivatives, 2c and 2f and evaluated these
compounds for activity against a panel of 39 clinical RGM isolates belonging to M. abscessus subsp.
abscessus (n = 8), M. abscessus subsp. bolletii (n = 7), M. abscessus subsp. massiliense (n = 8), as well as
Mch (n = 8), and Mfo (n = 8).
14
We synthesized compounds, 2a-b, 2d-e, 2g as previously described (Fig. 2, Table 1) as well as two new
compounds, 2c and 2f. The 2-[bis(isopropyl)amino] ethanamine derivative, 2f, was prepared using the
commercially available amine. Compound 2c was prepared from the [(3-phenyl-5-
isoxazolyl)methyl]amine which was prepared from N-Boc-propargyl amine and N-hydroxybenzene-
carboximidoyl chloride (Fig. 3). The [(3-phenyl-5-isoxazolyl)methyl]amine was coupled to the activated
rifamycin carbamate as were all other derivatives shown in Figure 2. All compounds were characterized
1
by H NMR and LC/MS and were consistent with spectral data obtained previously. For the new
1
compounds 2c and 2f, structures were confirmed by H NMR and LC/MS data and molecular formulas
were confirmed by high resolution mass spectrometry.
Figure 2: Synthesis of C25 rifamycin carbamate derivatives.

O
O
25
N
O
O
R
25
+
O
NH
N
O
OH
O
OH
OH
O
O
1) RNH , CH Cl , iPr NEt
2 2 2 2
OH
O
O
O
NH
N
2
) dil HCl (aq) THF
NH
N
O
O
O
O
O
O
O
O
O
O
2
Figure 3: Synthesis of 5-phenyl-isoxazole
O
N
OH 1) iPr NEt, THF
2
N
NH
O
NH2
+ Ph
O
2) TFA
Cl
The compounds were evaluated in a minimum inhibitory concentration assay (MIC) using standard
conditions as detailed in Clinical Laboratory and Standards Institute (CLSI) document M24-A2.15-16 MIC
values presented in Table 1 reflect the median value of the investigated clinical isolates for each (sub-)
species (Table 1).
It is interesting to note that the previously reported structure activity relationships (SARs) for Msm differ
significantly from the SARs observed in this study of RGM. For Msm, 2a and 2b were among the most
14
active compounds with MIC values of 0.06 g/ml. In contrast, both 2a and 2b had only modest activity
against Mab with MIC values of 8 and 64 g/ml respectively. In this study, compound 2g was the most
potent compound tested thus far against Mab and was also a sub-micromolar inhibitor of Msm.

Compound 2c (MIC of 0.5 g/ml) was surprisingly more potent against Mab than the other two isoxazole
derivatives 2d and 2e which had MIC’s of 8 and 64 g/ml respectively. Both 2d and 2e were sub-
micromolar inhibitors of Msm.14 Compound 2f had similar potency against Mab and Msm but was not
uniformly potent across all investigated RGM species. In addition, 2f was considerably less effective
against Mfo with an MIC of 8 mg/ml compared to a range of <0.03 to 1 g/ml for the other compounds.
Mfo was more susceptible to all the C25 carbamates tested in this study than any of the other investigated
Mycobacterium species with the exception of 2f.
We further evaluated compounds 2c, 2e and 2g in a zone of inhibition assay using Mab ATCC 19977 and
a clinical strain FM0058. For this purpose, a bacterial inoculum corresponding to a McF 0.5 turbidity
standard was streaked on a Muller Hinton II agar plate, and filter paper disks containing 25 l of the
investigated derivatives at various concentrations were placed on the agar plate using sterile forceps.17
Inhibition zone diameters were obtained following 3 days of incubation at 37°C. Compounds 2c and 2e
had similar zones of inhibition despite having considerably different MIC values. This may be reflected in
the better diffusion of 2e versus 2c into the agar. Most importantly, compound 2g showed an inhibitory
effect on Mab growth at concentrations down to 6.25 ng and essentially cleared the plate at the highest
doses. This observation corroborates the superior anti-mycobacterial activity of compound 2g in the MIC
assay.

Table 1: Minimum inhibitory concentrations in g/ml for M. abscessus, M. chelonae and M. fortuitum. Shown are median MIC values for 8
clinical isolates per (sub-)species. All concentrations are expressed in g / ml.
Compound
Substituent
Species
M. abscessus
M. abscessus
M. abscessus
M. chelonae M. fortuitum M. smegmatis14
subsp. abscessus
subsp. bolletii
subsp. massiliense
1
Rifampicin
>64
64
8
>64
16
>64
8
>64
>64
16
>64
32
0.06
0.06
NT
2
2
2
a
b
c
3
R = 4-CH OPh-
1
1
R = 4-(CH
3
)
2
NPh-
8
8
R =
0.5
0.125
0.5
0.5
0.0625
O
N
2
d
R
=
64
16
64
64
0.25
0.12
O
O
N

O
N
2
e
8
8
8
64
1
0.5
N
R =
2
2
f
0.5
0.25
1
1
8
NT
N
R =
R =
g
<0.03
0.0625
0.25
<0.03
<0.03
0.12
N
Table 2: Zones of inhibition for M. abscessus ATCC 19977 and clinical strain FM0058 with compounds 2c, 2e and 2g. All zones are reported in
millimeters measured as the diameter across the cleared zone including the filter paper disc. Each disc was loaded with 25 l of the respective
antibiotic solution. The average of three independent measurements is reported. No inhibition was observed below 3 ng for any compounds.
Inhibition Zone diameter (mm)
Compound
rifampin
2e
Bacterial Strain
2c
2g
Disk Load (ng)
ATCC1477
FM0058
ATCC1477
FM0058
ATCC1477
FM0058
ATCC1477
>30
>30
>30
24
FM0058
>30
>30
29
8
4
2
1
5
2
00
00
00
00
0
5
0
0
0
0
0
0
0
0
0
0
0
0
0
0
27
24
20
16
14
0
>30
30
25
20
9
0
0
24
20
16
12
7
0
0
>30
26
21
17
13
7
24
22
20
16
21
16
13
1
2.5
0
0

6
.25
0
0
0
0
0
0
0
11

To explore the finding of Rominski et al. that some RMP C25 derivatives were still subject to ADP-
ribosylation, we incubated the Mab type strain ATCC 19977 with compound 2g. Using LC/MS, we were
unable to find any C23 ribosylated or adenine ribosylated diphosphate derivatives. In contrast, when we
incubated Mab with 1, greater than 90% of 1 was converted to the C23 ribosylated rifampicin 1a and we
did not observe any adenine ribosylated diphosphate derivative 1b, as described by Dabbs and
Baysarowich.11-12
In the light of highly successful, rifampin-based therapies for tuberculosis, the unavailability of rifampin
3
derivatives with strong activity against RGM presents a considerable therapeutic challenge. This study
shows that rifamycin-based activity can be restored by modification of the rifamycin structure to block
Arr-mediated inactivation of rifamycin derivatives and that compounds such as 2g can be generated that
present a novel therapeutic option in antimycobacterial regimens targeting RGM. We are currently
preparing additional analogs of this initial series to improve the promising Mab activity seen with this
initial set. We are also trying to grow Mab resistant to these compounds by serial passage in media
containing sub inhibitory drug concentrations. We expect to see rpoB mutations but to date have not yet
observed any mutational resistance to these compounds. In Staphylococcus aureus, rpoB resistant
mutants can be obtained after a single passage.14
Funding
This work was in part funded by a grant of Joachim Hertz Foundation to FPM and TAMIU start-up funds
to KDC. NMR data was collected on a Bruker 400 MHz NMR purchased using NSF MRI award #1337508.
References

1
. Brown-Elliott BA, Philley JV, in Rapidly Growing Mycobacteria, Microbiology Spectrum, Schlossberg, D.
Ed. , 2017 Jan;5(1), 703-717. doi: 10.1128/microbiolspec.TNMI7-0027-2016.
. Griffith DE, Aksamit T, Brown-Elliott BA, Catanzaro A, Daley C, Gordin F, Holland SM, Horsburgh R, Huitt
2
G, Iademarco MF, Iseman M, Olivier K, Ruoss S, von Reyn CF, Wallace RJ, Jr, Winthrop K, An official
ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases.
Am. J. Respir. Crit. Care Med. 2007, 175:367–416. http://dx.doi.org/10.1164/rccm.200604-571ST.
3
. Maurer FP, Bruderer VL, Ritter C, Castelberg C, Bloemberg GV, Böttger EC, 2014. Lack of Antimicrobial
Bactericidal Activity in Mycobacterium abscessus. Antimicrob Agents Chemother, 2014, 58(7):3828-3836
http://dx.doi.org/10.1128/AAC.02448-14
4
. Aziz DB, Low JL, Wu M-L, Gengenbacher M, Teo JWP, Dartois V, Dick T, Rifabutin is active against
Mycobacterium abscessus complex. Antimicrob Agents Chemother, 2017, 61:e00155-17.
https://doi.org/10 .1128/AAC.00155-17.
5
. Perucca E, Grimaldi R, Frigo GM, Sardi A, Monig H, Ohnhaus EE, Comparative effects of rifabutin and
rifampicin on hepatic microsomal enzyme activity in normal subjects. Eur J Clin Pharmacol 1988; 34:595–
5
6
99.
. Griffith DE, Brown BA, Girard WM, Wallace RJ Jr, Adverse events associated with high-dose rifabutin in
macrolide-containing regimens for the treatment of Mycobacterium avium complex lung disease. Clin
Infect Dis 1995;21:594–598.
7
. Aubry-Damon H, Soussy C-J, Courvalin P, Characterization of Mutations in the rpoB Gene That Confer
Rifampin Resistance in Staphylococcus aureus, Antimicrobial Agents and Chemotherapy, 1998, 42(10), p.
590–2594.
2

8
. Obata S, Zwolska Z, Toyota, E, Kudo K, Nakamura A, Sawai T, Kuratsuji T, Kirikae T, Association of rpoB
mutations with rifampicin resistance in Mycobacterium avium, International Journal of Antimicrobial
Agents, 2006, 27, 32–39.
9
. Taniguchi H, Aramaki H, Nikaido Y, Mizuguchi Y, Nakamura M, Koga T, Yoshida S-I, 108 Rifampicin
resistance and mutation of the rpoB gene in Mycobacterium tuberculosis, FEMS Microbiology Letters 144
1996) 103- 108.
(
1
0. Rominski A, Roditscheff A, Selchow P, Bötger EC, Sander P, Intrinsic rifamycin resistance of
Mycobacterium abscessus is mediated by ADP-ribosyltransferase MAB_0591, 2016, J Antimicrob
Chemother. 72 (2): 376-384, http://dx.doi.org/10.1093/jac/dkw466.
1
1. Dabbs ER, Yazawa K, Mikami Y, Miyaji M, Morisaki N, Iwasaki S, Furihata K, Ribosylation by
mycobacterial strains as a new mechanism of rifampin inactivation. Antimicrob Agents Chemother. 1995;
3
1
9(4):1007–1009;
2. Baysarowich J, Koteva K, Hughes DW, Ejim L, Griffiths E, Zhange K, Junop M, Wright G, Rifamycin
antibiotic resistance by ADP-ribosylation: Structure and diversity of Arr., PNAS, 2008; 105(12):4886-91.
3. Cohen MS, Chang P, Insights into the biogenesis, function, and regulation of ADP-ribosylation, Nature
Chemical Biology, 2018, 14, 236–243.
4. Combrink KD, Denton DA, Harran S, Ma Z. Chapo K, Yan D, Bonventre E, Roche ED, Doyle TB, Robertson
1
1
GT, Lynch AS, New C25 carbamate rifamycin derivatives are resistant to inactivation by ADP-ribosyl
transferases, Bioorganic & Medicinal Chemistry Letters, 2007, 17, 522–526.
1
5. Clinical and Laboratory Standards Institute. 2011. Susceptibility testing of mycobacteria, norcadiae,
nd
and other aerobic actinomycetes; approved standard, 2 . Ed. CLSI document M24-A2. Clinical and
Laboratory Standards Institute, Wayne, PA.

1
6. Wiegand I, HiIpert K, Hancock REW, Agar and broth dilution methods to determine the minimal
inhibitory concentration (MIC) of antimicrobial substances, Nature Protocols, 2008, 3(2), 163-175.
7. A diagram of the plate layout as well as pictures of the zones of inhibition for each compound can be
found in the supplemental material.
1

Rifamycin derivatives active against pathogenic rapidly-growing mycobacteria
a,#
a
a
a,d
a
Keith D. Combrink , Andrea Ramirez Ramos , Stephanie Spring , Sebastian Schmidl , Kira Elizondo ,
a
a
b,c
Petronilo Morin , Bryant De Jesus , and Florian P. Maurer
Graphical Abstract
C25 carbamate rifamycin derivatives show potent activity against rapidly growing mycobacterium, RGM,
including Mycobacterium abscessus and are not substrates for the bacterial modifying enzyme Arr.
Rifampin has no activity against RGM and is heavily modified by RGM.
N
O
NH
O
OH
OH
O
O
OH
O
NH
N
O
O
O
O
O