Comparative anti-inflammatory effects of roxithromycin, azithromycin and clarithromycin.

Abstract

There are many published reports on the anti-inflammatory effects of macrolides, some dating back to the introduction of erythromycin. Macrolides have been shown to affect a number of the processes involved in inflammation, including the migration of neutrophils, the oxidative burst in phagocytes and the production of various cytokines, although the precise mechanisms are not clear. These effects have been linked to the ability of macrolides to accumulate in mammalian cells. Roxithromycin, a macrolide with better plasma concentrations and higher tissue concentrations than erythromycin, has been tested in a standard animal model used for evaluating anti-inflammatory drugs. When rats were given a prophylactic dose (20 mg/kg), roxithromycin suppressed the oedema produced by injecting carrageenin into the paw with effects almost equal to that seen with the non-steroidal anti-inflammatory drug nimesulide. Azithromycin and clarithromycin, macrolides with better pharmacokinetics than erythromycin, only showed slight anti-inflammatory effects. These results confirm that roxithromycin has anti-inflammatory properties in vivo and encourage the investigation of its mode of action.

Full text

Page 1

Journal of Antimicrobial Chemotherapy (1998) 41, Suppl. B, 47–50
Introduction
There have been numerous reports over the past 25 years
of macrolide antibiotics having anti-inflammatory activity.
They were shown to be effective in alleviating asthma as
long ago as 19701and 1974.2More recently, Miyatake et
al.3,4have shown that erythromycin had a favourable effect
in patients with bronchial asthma, while others have used
erythromycin and the newer macrolides in cases of diffuse
panbronchiolitis (DPB).5,6
Erythromycin has been used with some success for
treating acne7,8and is active in various animal tests
of inflammatory cutaneous conditions.9–11
numerous reports of macrolides affecting the migration of
neutrophils and the production of cytokines, although
these are often contradictory.12–14
Although the precise mechanisms of these anti-
inflammatory effects are not clear, a number of authors
have suggested that, in some conditions, interactions
between macrolides and white blood cells are important.15
The ability of macrolides to accumulate in mammalian
cells may be important in reducing inflammation. Three
new macrolide derivatives, azithromycin, clarithromycin
There are
and roxithromycin, which have more favourable
pharmacokinetic profiles16and better cell penetration than
erythromycin,17have been tested in a rat paw oedema
model. Their ability, when administered prophylactically,
to prevent the inflammatory response to carrageenin was
compared with that of nimesulide, a new non-steroidal
anti-inflammatory drug (NSAID).
Materials and methods
Rat carrageenin paw oedema
The rat carrageenin paw oedema model as described by
Van Arman et al.18was used. Groups of five male Wistar
rats weighing 200–250 g (Charles River, Calco, Italy) were
given water and then an oral dose of a macrolide,
nimesulide or saline. Hind paw volumes were measured
by mercury plethysmography 1 h later. After baseline
measurements of paw volume, 0.05 mL of a 1% (w/v)
solution of carrageenin in saline was injected into the
plantar region of the right hind paw. The paw volumes
were measured at hourly intervals up to 6 h. The change in
paw volume was calculated for each time point.
47
Comparative anti-inflammatory effects of roxithromycin, azithromycin
and clarithromycin
F. Scaglione* and G. Rossoni
Instituto di Farmacologia, University of Milan, Via Vanvitelli 32, Milan 29129, Italy
There are many published reports on the anti-inflammatory effects of macrolides, some dating
back to the introduction of erythromycin. Macrolides have been shown to affect a number of
the processes involved in inflammation, including the migration of neutrophils, the oxidative
burst in phagocytes and the production of various cytokines, although the precise
mechanisms are not clear. These effects have been linked to the ability of macrolides to
accumulate in mammalian cells. Roxithromycin, a macrolide with better plasma concen-
trations and higher tissue concentrations than erythromycin, has been tested in a standard
animal model used for evaluating anti-inflammatory drugs. When rats were given a
prophylactic dose (20 mg/kg), roxithromycin suppressed the oedema produced by injecting
carrageenin into the paw with effects almost equal to that seen with the non-steroidal anti-
inflammatory drug nimesulide. Azithromycin and clarithromycin, macrolides with better
pharmacokinetics than erythromycin, only showed slight anti-inflammatory effects. These
results confirm that roxithromycin has anti-inflammatory properties in vivo and encourage the
investigation of its mode of action.
*Tel: +39-2-718937; Fax: +39-2-718687.
© 1998 The British Society for Antimicrobial Chemotherapy
JAC

Page 2

F. Scaglione and G. Rossoni
Compounds
Three macrolides, namely azithromycin (Pfizer, Rome,
Italy), clarithromycin (Abbott, Rome, Italy) and roxithro-
mycin (Hoechst Marion Roussel, Milan, Italy), were used
at a single oral dose of 20 mg/kg. The anti-inflammatory
drug nimesulide (Helsinn, Biasca, Switzerland) was used
as control at a single oral dose of 4 mg/kg. All drugs were
administered 1 h before injecting carrageenin into the
paw. Carrageenin (Sigma, St Louis, MO, USA) was
prepared as a 1% (w/v) solution in saline.
Statistical analysis
The significance of differences between control and
treated groups was evaluated by one-way analysis of
variance (ANOVA) followed by Student Newman–Keuls
test. P values of ?0.05 were considered significant.
Results
The change in mean paw volume for the various groups is
shown in Figure 1. The baseline measurements for the
various groups were very similar, with mean values of
1.34–1.38 mL. In the untreated control animals, swelling
of the paw was evident by 1 h after administration of
carrageenin and reached a peak at 4 h, with an increased
volume of 0.83 mL, decreasing slowly thereafter. The
NSAID nimesulide suppressed this swelling almost
completely, the change in volume being approximately
0.1 mL. Compared with the control, azithromycin and
clarithromycin slightly reduced the swelling (giving
volumes of 0.56 and 0.49 mL, respectively), while roxi-
thromycin reduced it markedly (giving a volume of 0.25
mL), being only slightly less effective than nimesulide.
In all groups the maximum swelling was seen 4 h after
administration of carrageenin; the mean increases in paw
volumes at 4 h are presented in Figure 2. The increase in
paw volume in the group receiving roxithromycin was 0.25
mL, which compares favourably with that found for
nimesulide (0.13 mL). These values both differ sig-
nificantly from that of the control group (P ? 0.001). The
maximum increase with azithromycin and clarithromycin
was 0.56 and 0.49 mL, respectively; again each of these
values is significantly different from the paw volume
attained with the saline control (P ? 0.05).
Discussion
The anti-inflammatory properties of macrolides were first
recognized in the early 1970s, when erythromycin and
troleandomycin, for example, were shown to be effective
in the treatment of patients with asthma.1,2More recently,
erythromycin has been found to be of benefit in patients
with bronchial asthma.3,4In Japan, erythromycin, clari-
thromycin and roxithromycin have all been shown to be
effective in the treatment of patients suffering from DPB,
a chronic inflammation of the bronchioles which is more
prevalent in Japan than elsewhere.5,6Shirai et al.6found
that roxithromycin was more effective than clarithromycin
or erythromycin in the treatment of DPB. The authors
considered that these effects were not related to inter-
actions with polymorphonuclear lymphocytes (PMNs);
however, they used cells collected from normal healthy
volunteers and not patients. Kadota et al.5harvested
neutrophils from patients treated with roxithromycin, and
found a reduction in both the interleukin concentrations
produced and the number of neutrophils in bronchial
alveolar lavage (BAL) fluid.
These two studies illustrate some of the problems in
48
Figure 1. Mean change in paw volume (?S.E.M.) after carrageenin-
induced oedema; ?, control; ?, roxithromycin; ?, clarithromycin;
?, azithromycin; ?, nimesulide.
Figure 2. Mean increase in paw volume 4 h after carrageenin-
induced oedema. *, P ? 0.05; **, P ? 0.001.

Page 3

Anti-inflammatory effects of macrolides
understanding the mechanisms responsible for the anti-
inflammatory properties of macrolides. Many of the more
detailed studies that have addressed this area have given
contradictory results. For example, some authors have
found that erythromycin inhibits the chemotaxis of
PMNs, while others have found that it does not.9,19These
differences may be explained, in part, by technical
differences, as some tests were carried out in animals,19
some used cells harvested from animals treated with the
macrolide (ex vivo)12,14and some were conducted purely
in vitro.20In addition, the mechanism by which con-
ventional anti-inflammatory drugs work is complex and
not fully understood.
The ability of macrolides to accumulate in mammalian
cells16may play an important role in their anti-inflam-
matory activity, as PMNs, although important in anti-
microbial defence, also contribute to inflammation and
tissue damage. An excessive influx of PMNs produces high
localized levels of oxidants and proteolytic enzymes.
Macrolides, including roxithromycin, reduce the oxidative
burst in phagocytic cells21,22and increase intracellular
levels of cAMP,23contributing to an anti-inflammatory
effect.
Other important aspects of the inflammatory response
include the production of cytokines, and macrolides have
been reported to influence the levels of a number of
inflammatory leukotrienes. Roxithromycin, for example,
has been shown to inhibit the production of interleukin
1? and tumour necrosis factor ? (TNF-?) in human
peripheral blood monocytes in a dose-dependent
manner.24In a series of experiments using mice dosed with
roxithromycin and mitogen-activated human peripheral
leucocytes, Konno et al.13,14showed that roxithromycin
had a dose-dependent inhibitory effect on the production
of a number of interleukins and TNF-?. They concluded
that these effects would contribute to anti-inflammatory
and anti-asthmatic activity.
Conventional anti-inflammatory drugs are tested in a
range of animal models designed to demonstrate a direct
overall anti-inflammatory effect. Many rely on the in-
jection of an irritant or inflammatory substance, such
as croton oil, carrageenin, poly-L-arginine or lipopoly-
saccharide. The carrageenin rat paw oedema model,
described by Van Arman et al.,18is well established and
widely used to demonstrate the effects of anti-inflam-
matory agents. It remains a standard model for new
agents; Brand et al.25used it in their evaluation of capsaicin
analogues, while Kumakura et al.26used it to demonstrate
the effects of a prodrug of indomethacin.
In the study reported here, all three macrolides tested
were able to reduce the inflammation provoked by
carrageenin in the rat paw model. Roxithromycin reduced
the inflammation more than either azithromycin or clari-
thromycin, almost completely suppressing the oedema
produced by carrageenin. Roxithromycin was only slightly
less active than nimesulide. These results confirm the
findings of Agen et al.,15who found that roxithromycin had
only slightly less anti-inflammatory activity than the
NSAID indomethacin. In a poly-L-arginine rat paw model,
roxithromycin had greater anti-inflammatory activity,
equal to that of indomethacin, but it was less active than
indomethacin in a chronic granuloma produced by im-
planting polyester sponges intraperitoneally.15
authors suggest that the activity of roxithromycin is not
related to inhibition of prostaglandin synthesis. We have
investigated this by testing roxithromycin in another
inflammatory model in rats, using implanted sponges
soaked in carrageenin (unpublished data). In contrast with
the previous authors, we found that, although the numbers
of polymorphs and the levels of leukotriene B4were little
affected, marked dose-dependent inhibition of prosta-
glandin E2and thromboxane B2occurred. A decrease in
the volume of exudate was also produced.
Individual NSAIDs act on separate parts of the in-
flammatory cascade; for example, indomethacin and
nimesulide reduce the production of prostaglandins by in-
hibiting cyclooxygenase, while others inhibit 5-lipooxy-
genase, an enzyme that is important in leukotriene bio-
synthesis.25,27These different modes of action lead to
different responses in various models of inflammation, and
the results of such studies require careful analysis.
The clinical reports of the beneficial effects of roxi-
thromycin in inflammatory conditions, taken together
with the results reported here, confirm the marked anti-
inflammatory activity of roxithromycin. Although the anti-
inflammatory activity of roxithromycin is secondary to its
anti-infective capacity, these results suggest that further
work should be conducted to investigate the mechanism of
these effects.
The
Acknowledgement
This study was funded by Hoechst Marion Roussel, Milan,
Italy.
References
1. Itkin, I. H. & Menzel, M. L. (1970). The use of macrolide antibiotic
substances in the treatment of asthma. Journal of Allergy 45,
146–62.
2. Spector, S. L., Katz, F. H. & Farr, R. S. (1974). Troleandomycin:
effectiveness in steroid-dependent asthma and bronchitis. Journal
of Allergy and Clinical Immunology54, 367–79.
3. Miyatake, H., Suzuki, K., Taki, F., Takagi. K. & Satake, T. (1991).
Effect of erythromycin on bronchial hyperresponsiveness in
patients with bronchial asthma. Arzneimittel Forschung 41, 552–6.
4. Miyatake, H., Taki, F., Taniguchi, H., Suzuki, R., Takagi, K. &
Satake, T. (1991). Erythromycin reduces the severity of bronchial
hyperresponsiveness in asthma. Chest 99, 670–3.
5. Kadota, J., Sakito, O., Kohno, S., Abe, K., Shirai, R., Kawakami,
K. et al. (1994). Roxithromycin treatment in patients with chronic
lower respiratory tract disease—its clinical efficacy and effect
49

Page 4

F. Scaglione and G. Rossoni
on cytokine. (In Japanese) Kansenshogaku Zasshi (Journal of the
Japanese Association for Infectious Diseases) 68, 27–33.
6. Shirai, T., Sato, A. & Chida, K. (1995). Effect of 14-membered ring
macrolide therapy on chronic respiratory tract infections and
polymorphonuclear leukocyte activity. Internal Medicine34, 469–74.
7. Eady, E. A., Holland, K. T. & Cunliffe, W. J. (1982). The use of
antibiotics in acne therapy: oral or topical administration? Journal of
Antimicrobial Chemotherapy10, 89–115.
8. Scholtz, J. R. (1977). Management of acne vulgaris. Rational
Drug Therapy 11, 1–6.
9. Dalziel, K., Dykes, P. J. & Marks, R. (1987). The effect of
tetracycline and erythromycin in a model of acne-type inflammation.
British Journal of Experimental Pathology68, 67–70.
10. Tarayre, J. P., Aliaga, M., Barbara, M., Villanova, G., Ballester,
R., Tisne-Versailles, J. et al. (1987). Cutaneously applied erythro-
mycin base reduces various types of inflammatory reactions in
mouse ear. International Journal of Tissue Reactions9, 77–85.
11. Plewig, G. & Schopf, E. (1975). Anti-inflammatory effects of
antimicrobial agents: an in vivo study. Journal of Investigative
Dermatology 65, 532–6.
12. Kita, E., Sawaki, M., Mikasa, K., Hamada, K., Takeuchi S.,
Maeda, K. et al. (1993). Alterations of host response by a long-term
treatment of roxithromycin. Journal of Antimicrobial Chemotherapy
32, 285–94.
13. Konno, S., Adachi, M., Asano, K., Okamoto, K. & Takahashi, T.
(1993). Anti-allergic activity of roxithromycin: inhibition of inter-
leukin-5 production from mouse T lymphocytes. Life Sciences 52,
PL25–30.
14. Konno, S., Asano, K., Kurukawa, M., Ikeda, K., Okamoto, K. &
Adachi, M. (1994). Antiasthmatic activity of a macrolide antibiotic,
roxithromycin: analysis of possible mechanisms in vitro and in vivo.
International Archives of Allergy and Immunology 105, 308–16.
15. Agen, C., Danesi, R., Blandizzi, C., Costa, M., Stacchini, B.,
Favini, P. et al. (1993). Macrolide antibiotics as anti-inflammatory
agents: roxithromycin in an unexpected role. Agents and Actions
38, 85–90.
16. Nilsen, O. G. (1995). Pharmacokinetics of macrolides. Com-
parison of plasma, tissue and free concentrations with special
reference to roxithromycin. Infection 23, Suppl.1, S5–9.
17. Carbon, C. (1995). Clinical relevence of intracellular and extra-
cellular concentrations of macrolides. Infection 23, Suppl. 1,
S10–14.
18. Van Arman, C. G., Begany, A. J., Miller, L. M. & Pless, H. H.
(1965). Some details of the inflammations caused by yeast and
carrageenin (with appendix on the kinetics of the reaction). Journal
of Pharmacological and Experimental Therapy150, 328–34.
19. Tamaoki, J., Sakai, N., Tagaya, E. & Konno, K. (1994).
Macrolide antibiotics protect against endotoxin-induced vascular
leakage and neutrophil accumulation in rat trachea. Antimicrobial
Agents and Chemotherapy 38, 1641–3.
20. Esterly, N. B., Furey, N. L. & Flanagan, L. E. (1978). The effect
of antimicrobial agents on leukocyte chemotaxis. Journal of
Investigative Dermatology 70, 51–5.
21. Labro, M. T., el Benna, J. & Babin-Chevaye, C. (1989).
Comparison of the in-vitro effects of several macrolides on the
oxidative burst of human neutrophils. Journal of Antimicrobial
Chemotherapy 24, 561–72.
22. Hand, W. L., Hand, D. L. & King-Thompson, N. L. (1990)
Antibiotic inhibition of the respiratory burst response in human
polymorphonuclear leukocytes. Antimicrobial Agents and Chemo-
therapy 34, 863–70.
23. Aoshiba, K., Nagai A. & Konno, K. (1995). Erythromycin
shortens neutrophil survival by accelerating apoptosis. Anti-
microbial Agents and Chemotherapy39, 872–7.
24. Yoshimura, T., Kurita, C., Yamazaki, K., Shindo, J., Morishima,
I., Machida, K. et al. (1995). Effects of roxithromycin on proliferation
of peripheral blood mononuclear cells and production of lipopoly-
saccharide-induced cytokines. Biological and Pharmaceutical
Bulletin 18, 876–81.
25. Brand, L. M., Skare, K. L., Loomans, M. E., Reller, H. H.,
Schwen, R. J., Lade, D. A. et al. (1990). Anti-inflammatory pharma-
cology and mechanism of the orally active capsaicin analogs, NE-
19550 and NE-28345. Agents and Actions 31, 329–40.
26. Kumakura, S., Mishima, M., Kobayashi, S., Shirota, H., Abe, S.,
Yamada, K. et al. (1990). Inhibitory effect of indomethacin farnesil,
a novel anti-inflammatory prodrug, on carrageenin-induced
inflammation in rats. Agents and Actions 29, 286–91.
27. Blackham, A., Norris, A. A. & Woods, F. A. (1985). Models for
evaluating the anti-inflammatory effects of inhibitors of arachidonic
acid metabolism. Journal of Pharmacy and Pharmacology 37,
787–93.
50