![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Antimicrobial Agents and Chemotherapy, December 2003, p. 3688-3693, Vol. 47, No. 12
0066-4804/03/$08.00+0 DOI: 10.1128/AAC.47.12.3688-3693.2003
Copyright © 2003, American
Society for
Microbiology. All Rights Reserved.
Department of Medicine, Division of General Internal Medicine,1 Department of Pathology,3 Department of Medical Microbiology, University Medical Center Nijmegen,4 Nijmegen University Center for Infectious Diseases, Nijmegen, The Netherlands2
Received 7 June 2003/ Returned for modification 20 June 2003/ Accepted 22 August 2003
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials
AND METHODSResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials
AND METHODSResultsDiscussionReferences |
|---|
The polyene compound amphotericin B has been the mainstay of antifungal therapy for critically ill patients with invasive candidiasis. However, due to its severe and dose-limiting adverse effects, alternatives to amphotericin B are being used, either alone or as combination therapy. The triazole antifungal agent fluconazole has proven to be equally effective in treating candidemia as amphotericin B (18) and has become the drug of choice for treatment of candidemia and disseminated Candida albicans infection (1, 19). Furthermore, fluconazole is considered the drug of choice for patients with peritoneal candidiasis (5), and fluconazole prophylaxis is able to prevent abdominal candidiasis in high-risk surgical patients (6). However, despite antifungal treatment, mortality remains high and additional therapy with agents that augment host defense, such as growth factor granulocyte colony-stimulating factor (G-CSF) may be of potential therapeutic benefit.
Pretreatment with
recombinant murine G-CSF (rmG-CSF) beneficially influences the course
of acute disseminated candidiasis
(11) or bacterial
peritonitis (2,
14) in mice. Combined
therapy with an antifungal agent and rG-CSF has an additive effect
against disseminated candidiasis in nonneutropenic mouse models when
compared with antifungal treatment alone
(9,
12). However, recent
findings from a randomized, double-blind study have suggested that
patients with intra-abdominal candidiasis treated with recombinant
human G-CSF may have had a less favorable outcome than patients treated
with fluconazole alone (B. J. Kullberg, K. Vandewoude, R.
Herbrecht, F. Jacobs, and P. Kujath, Abstr. 38th Intersci. Conf.
Antimicrob. Agents Chemother., p. 479, 1998). One of the
explanations for the observed trend towards worse outcome of G-CSF
treatment in intra-abdominal candidiasis may be that G-CSF
down-regulates tumor necrosis factor alpha (TNF-
) production
(7,
11,
14). Previous research by
our group has shown that the clearance of intra-abdominal abscesses
caused by C. albicans is delayed in TNF-- and
lymphotoxin- deficient mice, primarily by inducing a T helper
(Th) 2 response (24). To
explore whether G-CSF application may indeed have detrimental
consequences in intra-abdominal candidiasis, we assessed the effect of
immunomodulation with rmG-CSF in the presence or absence of
amphotericin B or fluconazole on intra-abdominal abscesses caused by
C. albicans in mice.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials
AND METHODSResultsDiscussionReferences |
|---|
Compounds. Amphotericin B was obtained as Fungizone (Bristol-Myers Squibb, Woerden, The Netherlands), containing 50 mg of amphotericin B, 41 mg of sodium deoxycholate, and 20.2 mg of sodium phosphate per vial, and was reconstituted with 10 ml of distilled water to obtain a standard solution of 5 mg/ml. Further dilutions were prepared in pyrogen-free 5% dextrose.
Fluconazole was purchased from Pfizer Nederland, as a stock solution containing 2 mg of fluconazole/ml. rmG-CSF, provided by Amgen (Thousand Oaks, Calif.) was diluted in pyrogen-free saline to obtain a final concentration of 10 µg/ml.
MICs. MICs were determined according to the M27-A broth microdilution method as described by the National Committee for Clinical Laboratory Standards (15). Briefly, C. albicans strain UC820 was cultured for 24 h at 35°C and resuspended in 0.9% NaCl. The transmission was measured (530 nm) and was set between 75 and 77% (1 x 106 to 5 x 106 CFU/ml). The suspension was further diluted with distilled water (1:10) and RPMI 1640 containing 0.165 M 3-N-morpholinepropanesulfonic acid (MOPS) (1:100, pH 7.0) to obtain a final suspension of 1 x 103 to 5 x 103 CFU/ml. One hundred microliters of the suspension was incubated for 48 h at 35°C in the presence or absence of the antifungal agents at various concentrations. The MIC of amphotericin B was defined as the lowest concentration of amphotericin B which resulted in an optically clear tube, and the MIC of fluconazole was defined as the lowest concentration of fluconazole which resulted in a turbidity reduction of 50% compared with that of the growth control as determined spectrophotometrically.
Abdominal abscess induction. C. albicans strain UC820 was inoculated into 100 ml of Sabouraud broth and cultured for 24 h at 37°C. After three washes with pyrogen-free saline by centrifugation at 1,500 x g, the number of yeast cells was counted in a hemacytometer; occasional strings of two or more yeast cells were counted as 1 CFU of C. albicans. Pyrogen-free saline was used to dilute the suspension to the requested concentration. The viability of the yeast was at least 98%, as determined by plating serial dilutions on Sabouraud dextrose agar plates. Mouse feces were ground in a tissue grinder, suspended in 0.9% pyrogen-free saline to produce a 5% (wt/vol) mixture, and sterilized in a steam autoclave (15 min, 2 bar, 120°C). The sterility of the preparation was confirmed by plating aliquots on blood agar plates. To induce abdominal abscess formation, mice received an intraperitoneal injection of 100 µl of sterile fecal suspension containing 5 x 105 live C. albicans CFU.
Treatment regimens. Antifungal treatment started on day 3 of infection, with doses of fluconazole (50 mg/kg of body weight/day) or amphotericin B (0.75 mg/kg/day) that had proven to be equally effective in reducing the number of CFU in the kidneys in a series of pilot experiments of disseminated candidiasis (data not shown). Amphotericin B, fluconazole, or a combination of both agents was administered for 5 days. Amphotericin B was given as a subcutaneous (s.c.) injection once daily. Fluconazole was administered orally every 12 h (q12h) via gavage at a dosage of 25 mg/kg/dose. Control animals received 100 µl of sterile pyrogen-free 5% dextrose s.c. once daily. In another set of experiments, the effect of rmG-CSF on Candida outgrowth from intra-abdominal abscesses was assessed. rmG-CSF was administered daily as an s.c. injection of 1,000 ng/mouse (50 µg/kg) in 100 µl of pyrogen-free saline. Injections of rmG-CSF were given according to one of the following three schedules: regimen A, rmG-CSF prophylaxis was given once 24 h prior to infection; regimen B, prolonged rmG-CSF prophylaxis was administered once daily for 4 days from day -1 through day 2 of infection; regimen C, rmG-CSF therapy was given once daily for 5 days commencing on day 4 of infection.
For assessment of antifungal treatment combined with rmG-CSF on the number of Candida CFU in abscesses, rmG-CSF was administered according to regimen A or C starting on day 3 of infection, either alone or in combination with amphotericin B or fluconazole, with the same doses and dosing schedules described above.
Outcome assessments.
The number of circulating
granulocytes was determined in blood obtained from the retro-orbital
plexus from subgroups of mice prior to or at different points in time
during infection. On different days of infection, subgroups of animals
were anesthetized with ether and bled from the retro-orbital plexus for
measurement of circulating TNF-, interleukin-1 alpha
(IL-1), IL-6, gamma interferon (IFN-
), and IL-10
concentrations. Thereafter, these mice were sacrificed, and the
abdominal cavity was explored for the presence and number of abscesses
of 
1 mm. All abscesses of 1 mm were removed,
measured, rinsed with 70% ethanol, and subsequently rinsed with
saline to remove external microorganisms. Since it was hypothesized
that rmG-CSF might alter the containment of Candida within the
abscesses and lead to spread of the infection throughout the peritoneal
cavity with subsequent hematogenous dissemination, we removed a 10- by
10-mm sample of the peritoneum for assessment of peritonitis and the
left kidney for assessment of Candida dissemination. The
tissues were homogenized in sterile saline in a tissue grinder, and
Candida blastoconidia were enumerated by plating serial
dilutions (abscesses, kidneys) or the complete homogenate (peritoneum)
on Sabouraud dextrose agar plates, as described previously
(13). The CFU were
counted after 24 h of incubation at 37°C and
expressed as log CFU/organ. Bacterial coinfection of the abscesses was
evaluated by plating aliquots on blood agar. Cultures yielded no
aerobic or anaerobic bacteria. For histology, tissue samples were fixed
in buffered formalin (4%) and embedded in paraffin. Sections
were stained with periodic acid-Schiff and hematoxylin-eosin and
examined microscopically.
Stimulation of
splenic lymphocytes.
To
determine the effect of rmG-CSF on Th cell responses, spleens of groups
of 5 mice that received either rmG-CSF therapy (1,000 ng once daily for
5 days) or control vehicle were removed on day 8 of infection. Spleen
cells were obtained by gently squeezing spleens in a sterile
200-µm-pore-size filter chamber. Microscopic examination of
Giemsa-stained cytospin preparations showed that splenocytes consisted
of 95% lymphocytes, 2% monocytes, and 3%
granulocytes. Splenocytes were washed and resuspended in RPMI-dm and
counted in a Bürker counting chamber, and the number was adjusted
to 5 x 106/ml. One milliliter of the cell suspension
was stimulated with 1 x 107 heat-killed C.
albicans UC820 blastoconidia (effector-to-target cell ratio,
2:1). The measurement of IFN- and IL-10 concentrations was
performed in supernatants collected after 48 h of incubation
at 37°C in 5% CO2 in 24-well plates (Greiner,
Alphen a/d Rijn, The
Netherlands).
Cytokine assays.
The concentrations of TNF-
and IL-1 were determined by specific radioimmunoassays, as
described previously
(16). The detection limit
with a 100-µl sample was 40 pg/ml for TNF- and 20
pg/ml for IL-1. IL-10, IFN-, and IL-6 concentrations
were determined by a commercially available enzyme-linked immunosorbent
assay kit (Biosource Europe), according to the guidelines of the
manufacturer. The detection limits were 8, 1.5, or 150 pg/ml,
respectively.
Statistical analysis. Values were expressed as means ± standard deviation or, in case of logarithmic data, as means and 95% confidence intervals (95% CIs). The differences between two groups were analyzed by the Mann-Whitney U test. For these comparisons, the level of significance was set at a P value of <0.05. For comparison of three or more groups, the data were analyzed by using the Kruskal-Wallis one-way analysis of variance. For posttest comparisons, the Bonferroni t method was used. To ascertain reproducibility, most experiments were performed at least twice, and the data represent the average results of all experiments performed.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials
AND METHODSResultsDiscussionReferences |
|---|
Intra-abdominal abscesses and antifungal therapy. All animals survived the acute phase of infection and consistently produced abscesses, which could be detected from day 3 of infection. No differences in the number of abscesses were observed between any groups on any day of infection (Table 1). On day 8 of infection, quantification of yeast cells recovered from intra-abdominal abscesses from amphotericin B-treated mice showed a significantly reduced outgrowth of C. albicans CFU compared to controls or fluconazole-treated mice (P < 0.05) (Table 1). Fluconazole treatment did not reduce the number of CFU in the abscesses compared to the controls. Combination therapy of amphotericin B with fluconazole did not further reduce outgrowth from abscesses compared to amphotericin B treatment alone. Seven days after the termination of antifungal treatment, on day 14 of infection, no differences in the number of Candida CFU in the abscesses were observed (Table 1).
|
On the third day of infection, mice that had received rmG-CSF prophylaxis had significantly fewer Candida CFU in their abscesses than untreated mice did (P < 0.05) (Table 2). Both prophylaxis groups showed a reduced outgrowth of Candida CFU compared to controls on day 9 (P < 0.05) (Table 2). Therapy with rmG-CSF that started on day 4 of infection did not influence the numbers of CFU in the abscesses on either day of infection compared to controls. The number of abscesses did not differ between the groups on any day of infection (Table 2).
|
In addition, whether rmG-CSF increases hematogenous dissemination of Candida was assessed. Again, the number of Candida CFU was around the detection limit at any point in time, and no differences in the number of C. albicans CFU, i.e., dissemination of Candida, in the kidneys were observed between groups (data not shown).
Stimulation of splenic
lymphocytes.
To assess
whether rmG-CSF induces a Th1 or a Th2 type immune response, 5
x 106 splenic lymphocytes obtained on day 9 of
infection from untreated mice and mice that received rmG-CSF therapy
were stimulated with 107 CFU of heat-killed C.
albicans in vitro. Lymphocytes of mice that were treated with
rmG-CSF produced significantly more IL-10 (901 ± 304 pg/ml)
than control lymphocytes (532 ± 263 pg/ml; P <
0.05). No significant differences in IFN- production between
rmG-CSF-treated mice (560 ± 390 pg/ml) and controls (710
± 475 pg/ml) was observed. The IL-10/IFN- ratio was
twofold higher for mice treated with rmG-CSF (1.9 ± 0.58) than
for the control group (1.0 ± 0.54; P <
0.05).
Immunomodulation combined with antifungal agents. On day 8 of infection, amphotericin B treatment significantly reduced the number of CFU compared to controls (Fig. 1; P < 0.05). Combination treatment of rmG-CSF therapy with fluconazole or amphotericin B was slightly more effective in reducing the number of CFU in abscesses, although the differences between mice treated with an antifungal agent alone and mice treated with the agent in combination with rmG-CSF therapy were not significant (Fig. 1B; P > 0.05). Histopathology of the abscesses on day 8 of infection showed that treatment with rmG-CSF either alone or in conjunction with antifungal treatment increased the number of granulocytes present in the abscesses. This was observed for both prophylactic (data not shown) and to a larger extent for therapeutically administered rmG-CSF (Fig. 2). On day 15 of infection, no differences between groups were observed, neither by histopathological evaluation nor in outgrowth (data not shown), although the combination of amphotericin B with rmG-CSF therapy (3.02 log CFU/abscesses; 95% CI, 2.43 to 3.61 log CFU/abscesses) showed a tendency to reduce the outgrowth of Candida CFU from abscesses most compared to controls (3.39 log CFU/abscesses; 95% CI, 2.92 to 3.86 log CFU/abscesses; P > 0.05). No difference in the numbers of abscesses was observed between the groups at any point in time.
|
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials
AND METHODSResultsDiscussionReferences |
|---|
After injection of a sterile fecal suspension containing viable Candida blastoconidia, abscesses developed following a period of peritonitis. Administration of rmG-CSF during either of these periods had different effects on outgrowth from abscesses; rmG-CSF prophylaxis significantly reduced the number of CFU in abscesses compared to that in abscesses of control mice, whereas administration rmG-CSF therapy induced a slight increase in the number of CFU in abscesses compared with that in abscesses of control mice. Mice that had received rmG-CSF prophylaxis had large numbers of rmG-CSF-primed polymorphonuclear leukocytes (PMN) present at the time of infection and thus an enhanced capacity to clear the injected Candida blastoconidia from their abdominal cavities compared to that of controls, as was demonstrated by a reduced outgrowth from abscesses from these mice.
Although mice that had received
prolonged rmG-CSF prophylaxis showed an increased severity of diffuse
peritonitis, the data were around the detection limit and do not seem
to be of biological relevance, since it was not accompanied with an
increased hematogenous dissemination of Candida into the
kidneys. At the time of rmG-CSF therapy, which commenced not earlier
than on day 3 of infection, abscesses were already well established,
and at this point in time, inflammatory cells other than granulocytes,
such as macrophages and lymphocytes, gradually play a central role in
host defense against Candida abscesses. G-CSF inhibits the TNF
production by macrophages and monocytes
(7,
23) and has been shown to
induce a Th2-type cytokine pattern via attenuation of monokine release
(3,
8). Furthermore,
our laboratory has previously shown that TNF is important
for the clearance of Candida from intra-abdominal abscesses,
primarily through induction of Th1 responses and enhancement of the
extracellular killing capacity of granulocytes
(24). Hence, therapeutic
administration of rmG-CSF was expected to have detrimental effects in
this model of intra-abdominal Candida abscesses. However,
rmG-CSF therapy had no adverse influence on the course of infection.
Although the IL-10/IFN- ratios of mice treated with rmG-CSF
were significantly increased compared with controls, this may not have
been of biological relevance. G-CSF not only increases the number of
PMN but it also augments the expression of adhesion molecules,
recruitment, and capacity to kill Candida blastoconidia
(25-27).
These effects of G-CSF on anticandidal activity have probably
compensated for the anti-inflammatory effect of G-CSF through its
effect on macrophages and T cells.
Intra-abdominal abscesses are difficult to treat, and therapy consists primarily of drainage of the infected cavity with or without antimicrobial treatment. To date, no guidelines for the use of antifungal agents for treatment of intra-abdominal candidiasis exist. In an international conference of investigators with extensive experience in the treatment of candidal infections, 60% of the investigators would treat patients with peritoneal candidiasis with fluconazole alone and 5% would treat patients with amphotericin B lipid formulation (5). In our model, fluconazole had no beneficial influence on the course of experimental intra-abdominal Candida abscesses in mice, whereas amphotericin B improved the clearance of Candida blastoconidia from abscesses. This is contradicts the study by Sawyer et al., which showed that fluconazole was as effective as amphotericin B in reducing the number of Candida recovered from experimental intra-abdominal abscesses (20). In that study, however, fluconazole therapy was started at the time of infection and thus represents the effect of fluconazole on Candida peritonitis rather than on well-established Candida abscesses. Indeed, fluconazole prophylaxis has proven to prevent development of abdominal candidiasis in high-risk surgical patients (6).
Simultaneous administration of combination therapy of fluconazole with amphotericin B was not as effective as amphotericin B treatment alone in reducing the number of CFU in abscesses in our experiments.
Since combination therapy of an antifungal agent with G-CSF has proven to be more efficacious than antifungal treatment alone against disseminated candidiasis (9, 12), it was hypothesized that the combination of rmG-CSF therapy with antifungal agents may have a synergistic effect, despite the observation that therapy with rmG-CSF alone was not effective in our model of intra-abdominal abscesses. The addition of rmG-CSF therapy to either amphotericin B or fluconazole did not significantly improve the outcome, although there was a slight tendency towards further reducing the number of Candida CFU per abscess. Both amphotericin B and fluconazole accumulate within PMN (17), and their concentrations might have been increased by G-CSF, which has been shown to increase the intracellular concentration of antibiotics (10). In addition, histopathological examination of abscesses obtained from mice treated with rmG-CSF showed a larger number of granulocytes than in control mice. Serving as secondary transport systems for amphotericin B or fluconazole, these cells might have contributed to achieving higher drug concentrations in the abscesses. In our experiments, however, these putative effects have not led to significant synergistic effects of rmG-CSF and antifungal agents. In conclusion, amphotericin B was effective in treating experimental intra-abdominal Candida abscesses, and concurrent administration of rG-CSF was suggested to further improve the outcome.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials
AND METHODSResultsDiscussionReferences |
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Clin. Vaccine Immunol. | Clin. Microbiol. Rev. |
|---|---|
| J. Clin. Microbiol. | ALL ASM JOURNALS |
