Full text of DOI:10.1093/jac/47.5.581

Journal of Antimicrobial Chemotherapy (2001) 47, 581–587
JAC
The activity of 14-hydroxy clarithromycin, alone and in combination
with clarithromycin, against penicillin- and erythromycin-resistant
Streptococcus pneumoniae
Steven J. Martin*, Cory G. Garvin, Christopher R. McBurney and Eric G. Sahloff
The Infectious Diseases Research Laboratory at The University of Toledo, College of Pharmacy,
2801 W. Bancroft Street, Toledo, OH 43606, USA
There are no data regarding the activity of clarithromycin’s active metabolite, 14-hydroxy
clarithromycin, against penicillin-intermediate, penicillin-resistant or erythromycin-resistant
Streptococcus pneumoniae. Agar dilution MICs were determined for clarithromycin,
14-hydroxy clarithromycin (henceforth called ‘metabolite’), azithromycin, erythromycin and
clarithromycin/metabolite (2:1 and 1:1 ratio) against 24 penicillin-intermediate and 14 penicillin-
resistant strains, including 13 erythromycin-resistant clinical strains and one ATCC strain of
S. pneumoniae. The interaction between clarithromycin and its metabolite was determined
using an agar chequerboard assay against all isolates, and time–kill tests were performed
against five penicillin-intermediate (macrolide-susceptible) and five penicillin-resistant
(two macrolide-resistant) strains of S. pneumoniae using all antibiotics alone at simulated
peak serum concentrations, and clarithromycin/metabolite in a 2:1 ratio (physiological). MICs
were as follows: clarithromycin, 0.008–>64 mg/L; metabolite, 0.015–>64 mg/L; erythromycin,
0.015–>64 mg/L; azithromycin, 0.125–>64 mg/L; clarithromycin/metabolite (1:1 and 2:1 com-
binations), 0.001–>64 mg/L. The MIC of the clarithromycin/metabolite combination was one or
more tube dilution lower than the MIC of clarithromycin in 28 of the isolates tested. In chequer-
board testing, 13 strains (seven erythromycin susceptible and six erythromycin resistant)
demonstrated synergy, 18 additivity and seven indifference. In time–kill testing, bacterial
eradication below detection limits occurred with clarithromycin and metabolite in seven of
10 organisms. The combination of parent and metabolite was more rapidly bactericidal than
clarithromycin alone in six of the seven isolates (P = 0.026). The metabolite has potent activity
against S. pneumoniae and enhances the activity of the parent compound against this
organism. The metabolite’s activity must be considered in evaluating clarithromycin in vitro to
avoid underestimation of clarithromycin’s activity against the pneumococcus.
resistance mechanisms are acquired determinants.5 It is dif-
ficult to predict macrolide susceptibility solely by penicillin
Introduction
resistance.
Macrolide resistance in Streptococcus pneumoniae is in-
creasing.1 Macrolide resistance can be mediated by ribo-
somal modification or active drug efflux.2,3 The former
mechanism involves the erm(AM) gene and the latter
involves the mef(E) gene.2,4 Ribosomal modification is
associated with high-level resistance to macrolides, linco-
samides and streptogramins (MLSB-type resistance pat-
tern), often producing MICs of ꢀ64–128 mg/L.2,5 Efflux is
associated with low-level resistance to 14- and 15-membered
ring macrolides only (M-type resistance pattern).3,5 Both
Clarithromycin is extensively metabolized in the liver by
oxidative and hydrolytic mechanisms.6 An active metabol-
ite, 14-hydroxy clarithromycin (henceforth called ‘metabol-
ite’, accounts for approximately 20% of the parent drug’s
metabolism.6 Neither parent nor metabolite is extensively
bound to plasma proteins.6 The active metabolite provides
the majority of the compound’s antibacterial effects against
Haemophilus influenzae.7 Although few data are available,
it appears that the metabolite is also active against
*Corresponding author. Tel: ꢁ1-419-530-1964; Fax: ꢁ1-419-530-1950; E-mail: smartin2@pop3.utoledo.edu
581
© 2001 The British Society for Antimicrobial Chemotherapy

S. J. Martin et al.
penicillin-susceptible, erythromycin-susceptible S. pneumo- 25 erythromycin-susceptible and 13 erythromycin-resistant
niae.7–11 Its activity against penicillin-intermediate, peni- isolates by combining the parent compound and metabol-
cillin-resistant or erythromycin-resistant S. pneumoniae is ite, each at concentrations of 0.001–64 mg/L.13 Interaction
not known.
between the two compounds was determined by calculat-
The objective of this work was to characterize more ing the fractional inhibitory concentration (FIC) index.13
fully the potency of the metabolite against penicillin- and FIC indices were interpreted as follows: ꢃ0.5, synergy;
erythromycin-resistant S. pneumoniae, and to investigate ꢀ0.5–1, additive; ꢀ1–4, indifference; and ꢀ4, antagon-
whether this metabolite has a role in the parent com- ism.14
pound’s activity against these organisms in vitro. We
Standardized time–kill assays were performed in a CO2
investigated the in vitro activity of the metabolite alone and environment using 10 representative pneumococcal strains
in combination with clarithromycin against several clinical (two erythromycin resistant). The macrolides were each
strains of penicillin-intermediate, penicillin-resistant and tested alone at simulated physiological peak serum con-
erythromycin-resistant S. pneumoniae. Interaction was centrations: clarithromycin, 2.6 mg/L; metabolite, 0.8 mg/L;
tested by investigating combination MICs, agar dilution azithromycin, 0.4 mg/L; and erythromycin, 2.0 mg/L.15–17
chequerboards and time–kill assays.
Clarithromycin and metabolite were also tested in com-
bination. All work was performed in duplicate. There was
excellent correlation between duplicate colony counts, so
results presented are mean values.
Materials and methods
The pneumococci tested included 38 clinical isolates and
one ATCC control strain (S. pneumoniae 49619). These
consisted of 24 penicillin-intermediate (MIC 0.1–1.0 mg/L)
Results
and 14 penicillin-resistant S. pneumoniae (MIC ꢂ 2.0 Metabolite alone was as potent as the parent compound
mg/L). Twenty five of these isolates were erythromycin against S. pneumoniae (Tables I and II). For erythromycin-
susceptible (MIC ꢃ 0.25 mg/L) and 13 erythromycin resist- susceptible strains, the MIC90s of clarithromycin, metabol-
ant (MIC ꢂ 1 mg/L). MICs of erythromycin, azithromycin ite and erythromycin were 0.125 mg/L. The MIC90 of
and penicillin were also determined, and erythromycin and azithromycin was 0.5 mg/L. The MIC90s of the clari-
azithromycin were included in chequerboard and time–kill thromycin/metabolite 1:1 and 2:1 ratio were 0.06 mg/L; this
testing for comparison. Clarithromycin and metabolite was one tube dilution lower than the MIC of the parent
were obtained from Abbott Laboratories (Abbott Park, compound alone.
IL, USA). Erythromycin, azithromycin and penicillin V
potassium were obtained from the United States Phar- the macrolides were ꢀ64 mg/L, but MIC90s for organisms
macopoeia (Rockville, MD, USA). expressing the mef(E) gene (erythromycin, 8 mg/L; azithro-
For 13 erythromycin-resistant strains, the MIC90s of all
Agar dilution MICs were determined according to mycin, 32 mg/L; clarithromycin, 4 mg/L; metabolite, 8 mg/L;
the standards of the NCCLS with incubation in 5% CO2. clarithromycin/metabolite 1:1, 4 mg/L; and clarithromycin/
Final antibiotic concentrations tested were 0.001–64 mg/L. metabolite 2:1, 4 mg/L) were lower than MICs for those
S. pneumoniae ATCC 49619 and S. aureus ATCC 29213 expressing the erm(AM) gene (ꢀ64 mg/L for all).
were included for quality control. All procedures were per-
The clarithromycin/metabolite combination in a 2:1
formed in duplicate. MICs were determined for each agent ratio was more potent than clarithromycin (having a lower
individually and for clarithromycin and metabolite in a 2:1 MIC) for 25 strains of pneumococcus tested (19 with a
and 1:1 ratio.
decrease of one tube dilution and six with a decrease of two
Isolates that were not susceptible to erythromycin (MIC or more tube dilutions). The combination in a 1:1 ratio was
ꢂ 1 mg/L) were screened for the presence of known MLSB more potent than clarithromycin alone (having a lower
resistance genes by polymerase chain reaction (PCR) ampli- MIC) for 28 strains of pneumococcus tested (17 with a
fication with the following gene-specific primers: erm(AM): decrease of one tube dilution and 11 with a decrease of two
upper primer (5ꢄ position 974), TCAACCAAATAAT tube dilutions). In five of the 13 erythromycin-resistant
AAAACAA, lower (3ꢄ position 1311), AATCCTTCTT isolates the clarithromycin MIC was reduced by one or
CAACAATCAG; mef(E): upper (5ꢄ position 206), more tube dilutions by combination with the metabolite
ATGCAGACCAAAAGCCACCAT, lower (3ꢄ position (Table II).
439), GCCATAGACAAGACCATCGC.12 Crude lysates
Synergy was demonstrated against 13 strains and addi-
of the strains were prepared and stored at 4°C until tested. tive effects against 18 strains of S. pneumoniae. Against
One microlitre of lysate was used in a 25 ꢅL amplification erythromycin-resistant organisms, three of eight erm(AM)-
reaction (PCR Supermix; Gibco-BRL, Bethesda, MD, producing strains demonstrated synergy, as compared with
USA). The presence of the erm(AM) and mef(E) genes three of five mef(E)-producing strains. Against erythro-
was identified in 13 organisms.
mycin-susceptible isolates, seven of 25 strains demon-
Agar chequerboard testing was performed using the strated synergy, 16 additive effects and two indifference.
582

14-Hydroxy clarithromycin activity against S. pneumoniae
Table I. Agar dilution MICs and clarithromycin/metabolite chequerboard FICs for 25 erythromycin-susceptible
strains of S. pneumoniae
MIC (mg/L)
clarithromycin ꢁ metabolite
Organism erythromycin azithromycin clarithromycin metabolite
1:1 ratio
2:1 ratio
penicillin FICa
P10
P11
C10
C11
C12
U10
C13
I10
0.015
0.06
0.06
0.06
0.06
0.5
0.25
0.5
0.5
0.5
0.125
0.5
0.5
0.25
0.5
0.5
0.125
0.5
0.5
0.5
0.25
0.5
0.5
0.5
0.25
0.5
0.5
0.008
0.06
0.06
0.015
0.06
0.004
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.06
0.06
0.06
0.06
0.06
0.001
0.008
0.03
0.03
0.03
0.03
0.06
0.03
0.03
0.03
0.03
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.001
0.25
2
0.015
0.015
0.015
1
1
0.25
1
0.75
0.75
0.625
0.75
0.5
0.75
0.625
0.75
0.75
0.5
0.75
0.75
0.75
1
0.5
0.5
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.06
0.125
0.125
0.06
0.06
0.06
0.125
0.06
0.06
0.06
0.06
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.06
0.06
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
0.125
P12
I11
1
2
C14
C15
T10
C16
I12
P13
P14
U11
I13
C17
I14
P15
P16
C18
I15
0.125
0.25
0.06
0.5
2
4
1
4
2
0.5
0.5
0.75
0.75
0.75
0.75
0.75
1
2
0.125
0.125
2
0.25
0.5
0.25
0.002
0.025
aFractional inhibitory index for clarithromycin/metabolite chequerboard, calculated as A/MICA ꢁ B/MICB, where A is the concentration of drug A
in a tube that is the lowest inhibitory concentration in its row; MICA is the MIC of drug A alone for the organism; B and MICB are the equivalent
data for drug B.10
Representative time–kill curves for macrolide-resistant seven of the 10 organisms (seven of the eight erythromycin-
isolates are shown in Figures 1 and 2. For the macrolide- susceptible strains and neither of the two erythromycin-
susceptible isolates (data not shown), clarithromycin and resistant ones). The time required for bactericidal activity
its 14-hydroxy metabolite alone and in combination were to be manifested was determined for these seven isolates.
bactericidal and produced colony counts below the lower The combination of parent and metabolite was more
limit of detection at 24 h. Against the macrolide-resistant rapidly bactericidal than clarithromycin alone in six of the
[mef(E)], penicillin-resistant isolate (Figure 1), clarithro- seven isolates (P ꢆ 0.026). Metabolite was more rapidly
mycin and the combination of parent and metabolite bactericidal than clarithromycin in five of the seven iso-
reduced the colony count by approximately 2 log cfu/mL at lates, but the cumulative difference did not reach statistical
24 h, whereas growth in the presence of all other agents significance (P ꢆ 0.222).
resembled that of the control. For the macrolide-resistant
[erm(AM)], penicillin-intermediate S. pneumoniae (Figure
2), an initial decrease in colony counts was observed over
the first 12 h in the presence of clarithromycin, its metabol-
Discussion
ite, the combination and erythromycin, but in all tubes We speculated that clarithromycin’s active metabolite,
there was regrowth above baseline at 24 h.
14-hydroxy clarithromycin, may have significant activity
Clarithromycin and its metabolite completely eradicated against penicillin- and erythromycin-intermediate and
583

S. J. Martin et al.
584

14-Hydroxy clarithromycin activity against S. pneumoniae
Figure 1. Time–kill curve for macrolide-resistant [mef(E)], penicillin-resistant S. pneumoniae (P–A). ꢀ, Control; ꢁ, clarithromycin;
ꢂ, metabolite; ꢇ, clarithromycin/metabolite; ꢃ, erythromycin; ꢄ, azithromycin.
-resistant pneumococci, and contribute to the parent drug’s or greater activity against pneumococcal strains. In all but
clinical effectiveness. There are no reports of in vitro activ- the high-level macrolide-resistant isolate tested [I-B,
ity of the metabolite against these pathogens. The few data erm(AM)], the clarithromycin/metabolite combination
describing the compound’s activity against S. pneumoniae was more potent than the parent compound. In seven of the
isolates suggest that its potency is equal to or greater than 10 time–kill assays, bactericidal activity was demonstrated
that of its parent compound.7,9 Clarithromycin and its for all of the macrolides. Thus, macrolides may not be
metabolite have demonstrated synergy in combination effective against all pneumococcal isolates, but in vitro
against other bacteria, and this interaction may be respons- MICs may not be an accurate predictor of pharmaco-
ible for the seemingly few clinical failures of this drug dynamic effect.
against S. pneumoniae strains that are not susceptible to
The erm gene probably encodes high-level macrolide
penicillin.8,18,19
resistance; our work confirms that of others in this regard.20
In MIC testing, the MIC of clarithromycin was reduced In time–kill testing against an erm(AM)-expressing strain,
in ꢀ70% of the isolates when clarithromycin was combined erythromycin, and clarithromycin and metabolite either
with metabolite in a serum physiological ratio. Jones et al.18 alone or in combination reduced colony counts over the
demonstrated similar findings for the combination against first 12 h of the assay. Clarithromycin alone and in com-
Legionella spp. The results of agar chequerboard studies bination with the metabolite produced an approximately
confirmed this interaction, as 31 of 38 isolates demon- 2 log cfu/mL reduction in the macrolide-resistant strain
strated synergy or additive effects. To investigate the phar- encoded by the mef(E) gene. Erythromycin and azithro-
macodynamic nature of the interaction, we performed mycin appeared to have little activity against this strain.
time–kill assays against a variety of penicillin-intermediate, This efflux mechanism produces low- to moderate-level
-resistant and macrolide-resistant isolates. Simulated serum macrolide resistance, and organisms expressing this gene
concentrations were chosen for testing to simulate bacter- may not be uniformly affected by the macrolides.
aemia in vivo. We chose a 2:1 ratio of parent to metabolite
We have demonstrated that 14-hydroxy clarithromycin
to allow comparison of our results with those of other has activity against S. pneumoniae in vitro and that it con-
investigators.7,18 Chu et al.15 demonstrated that in healthy tributes to the activity of clarithromycin in MIC and
young and elderly volunteers the ratio of parent to metabol- time–kill studies. Further work on the activity of this
ite Cmax and Cmin serum concentrations after five doses of combination against mef(E)- and erm(AM)-expressing
clarithromycin (500 mg bd) ranged from 2.46:1 to 3.65:1 macrolide-resistant pneumococci is warranted. This work
and from 1.92:1 to 2.28:1, respectively. One would expect supports the argument that the activity of active metabol-
that higher concentrations of these agents, as might be ites must be taken into consideration in determining the
present in epithelial lining fluid, would produce similar antimicrobial activity of parent drugs.
585

S. J. Martin et al.
Figure 2. Time–kill curve for macrolide-resistant [erm(AM)], penicillin-intermediate S. pneumoniae (I-B). Symbols as in Figure 1.
6. Langtry, H. D. & Brogden, R. N. (1997). Clarithromycin. A review
of its efficacy in the treatment of respiratory tract infections in
Acknowledgements
immunocompetent patients. Drugs 53, 973–1004.
The authors gratefully acknowledge the technical assist-
7. Hardy, D. J., Hensey, D. M., Beyer, J. M., Vojtko, C., McDonald,
E. J. & Fernandes, P. B. (1988). Comparative in vitro activities of
new 14-, 15-, and 16-membered macrolides. Antimicrobial Agents
and Chemotherapy 32, 1710–9.
ance of Dr Robert Flamm and Jill Beyer. This work was
supported by a grant from Abbott Laboratories. The results
were presented, in part, at the Thirty-eighth Interscience
Conference on Antimicrobial Agents and Chemotherapy,
1998, San Diego, CA, and the Twenty-first International
Congress of Chemotherapy, 1999, Birmingham, UK.
8. Bedos, J.-P., Azoulay-Dupuis, E., Vallee, E., Veber, B. &
Pocidalo, J.-J. (1992). Individual efficacy of clarithromycin (A-56268)
and its major human metabolite, 14-hydroxy clarithromycin
(A-62671), in experimental pneumococcal pneumonia in the mouse.
Journal of Antimicrobial Chemotherapy 29, 677–85.
References
9. Logan, M. N., Ashby, J. P., Andrews, J. M. & Wise, R. (1991). The
in-vitro and disc susceptibility testing of clarithromycin and its
14-hydroxy metabolite. Journal of Antimicrobial Chemotherapy 27,
161–70.
1. Campbell, G. D. & Silberman, R. (1998). Drug-resistant Strepto-
coccus pneumoniae. Clinical Infectious Diseases 26, 1188–95.
2. Leclerq, R & Courvalin, P. (1991). Bacterial resistance to
macrolide, lincosamide, and streptogramin antibiotics by target
modification. Antimicrobial Agents and Chemotherapy 35, 1267–72.
10. Bergman, K. L., Olsen, K. M., Peddicord, T. E., Fey, P. D. &
Rupp, M. E. (1999). Antimicrobial activities and postantibiotic effects
of clarithromycin, 14-hydroxy clarithromycin, and azithromycin in
epithelial cell lining fluid against clinical isolates of Haemophilus
influenzae and Streptococcus pneumoniae. Antimicrobial Agents
and Chemotherapy 43, 1291–3.
3. Sutcliffe, J., Tait-Kamradt, A. & Wondrack, L. (1996). Strepto-
coccus pneumoniae and Streptococcus pyogenes resistant to
macrolides but sensitive to clindamycin: a common resistance pat-
tern mediated by an efflux system. Antimicrobial Agents and
Chemotherapy 40, 1817–24.
11. Gu, J.-W., Scully, B. E. & Neu, H. C. (1991). Bactericidal
activity of clarithromycin and its 14-hydroxy metabolite against
Haemophilus influenzae and streptococcal pathogens. Journal of
Clinical Pharmacology 31, 1146–50.
4. Tait-Kamradt, A., Clancy, A., Cronan, M., Dib-Hajj, F., Wondrack,
L., Yuan, W. et al. (1997). mefE is necessary for erythromycin-
resistant M phenotype in Streptococcus pneumoniae. Antimicrobial
Agents and Chemotherapy 41, 2251–5.
12. Shortridge, V. D., Flamm, R. K., Ramer, N., Beyer, J. & Tanaka,
S. K. (1996). Novel mechanism of macrolide resistance in Strepto-
coccus pneumoniae. Diagnostic Microbiology and Infectious
Diseases 26, 73–8.
5. Oster, P., Zanchi, A., Cresti, S., Lattanzi, M., Montagnani, F.,
Cellesi, C. et al. (1999). Patterns of macrolide resistance determin-
ants among community-acquired Streptococcus pneumoniae 13. Eliopoulos, G. M. & Moellering, R. C. (1996). Antimicrobial
isolates over a 5-year period of decreased macrolide susceptibility combinations. In Antibiotics in Laboratory Medicine, 4th edn,
rates. Antimicrobial Agents and Chemotherapy 43, 2510–2.
(Lorian,V., Ed.), pp. 330–96. Williams & Wilkins, New York.
586

14-Hydroxy clarithromycin activity against S. pneumoniae
14. Anonymous. (1995). Instructions to authors. Antimicrobial
Agents and Chemotherapy 39, i–xiv.
alone and in combination against Legionella species. European
Journal of Clinical Microbiology and Infectious Diseases 9, 846–8.
15. Chu, S., Wilson, D. S., Deaton, M. S., Mackenthun, A. V., 19. Martin, S., Pendland, S., Chen, C., Schreckenberger, P. &
Eason, C. N. & Cavanaugh, J. H. (1993). Single- and multiple-dose Danziger, L. (1997). In vitro activity of clarithromycin alone and in
pharmacokinetics of clarithromycin, a new macrolide antimicrobial. combination with ciprofloxacin or levofloxacin against Legionella
Journal of Clinical Pharmacology 33, 719–26.
spp: enhanced effect by the addition of the metabolite 14-hydroxy
clarithromycin. Diagnostic Microbiology and Infectious Diseases 29,
167–71.
16. Foulds, G., Shepard, R. M. & Johnson, R. B. (1990). The
pharmacokinetics of azithromycin in human serum and tissues.
Journal of Antimicrobial Chemotherapy 25, Suppl. A, 73–82.
20. Johnston, N. J., de Azavedo, J. C., Kellner, J. D. & Low, D. E.
(1998). Prevalence and characteristics of the mechanisms of
macrolide, lincosamide, and streptogramin resistance in isolates
of Streptococcus pneumoniae. Antimicrobial Agents and Chemo-
therapy 42, 2425–6.
17. Gerding, D. N., Hughes, C. E., Bamberger, D. M., Foxworth, J.
& Larson, T. A. (1996). Extravascular antimicrobial distribution and
the respective blood concentrations in humans. In Antibiotics in
Laboratory Medicine, 4th edn, (Lorian, V., Ed.), pp. 835–99. Williams
& Wilkins, Baltimore, MD.
18. Jones, R. N., Erwin, M. E. & Barrett, M. S. (1990). In vitro activ- Received 13 July 2000; returned 25 September 2000; revised 4
ity of clarithromycin (TE-031, A-67268) and 14OH-clarithromycin December 2000; accepted 24 January 2001
587