Therapeutic Effects of Benzoxazinorifamycin KRM-1648 Administered Alone or in Combination with a Half-Sized Secretory Leukocyte Protease Inhibitor or the Nonsteroidal Anti-Inflammatory Drug Diclofenac Sodium against Mycobacterium avium Complex Infection in Mice

Abstract

The effects of half-sized secretory leukocyte protease inhibitor or diclofenac sodium administered alone or in combination with the benzoxazinorifamycin KRM-1648 on the therapeutic efficacy of KRM-1648 against Mycobacterium avium complex (MAC) in mice were studied. Neither of the two anti-inflammatory drugs affected the efficacy of KRM-1648, while they exerted significant modulating effects on tumor necrosis factor alpha production by MAC-infected macrophages.

Mycobacterium avium complex (MAC) frequently causes disseminated and fatal infections in AIDS patients (3, 10). MAC is capable of multiplying and/or surviving in host macrophages (Mφs) (3), and it possesses intrinsic resistance to most antimycobacterial drugs except some macrolides and rifamycin derivatives (2, 3). Hence, MAC infections, particularly those in AIDS patients with immunodeficiency due to a severe defect in CD4⁺ T cells (3), are difficult to treat. MAC infection causes chronic inflammation and related reactions at the sites of infection, in particular, in the lungs, such as neutrophilic granulocytosis (1) and delayed-type hypersensitivity reactions that cause granuloma formation (13). These inflammatory reactions frequently cause granuloma liquefaction, resulting in cavity formation in the lungs, and moreover, they occasionally cause pulmonary emphysema (12).

Secretory leukocyte protease inhibitor (SLPI), a potent serine protease inhibitor in lungs which is secreted by bronchial and alveolar epithelial cells (24, 28), is useful for the treatment of degenerative and inflammatory diseases of the lung, including pulmonary emphysema, and some of these diseases are also associated with pulmonary mycobacterial infections (12, 33). It is thus interesting to examine the effect of SLPI on the outcome of chemotherapy of MAC-infected patients with anti-MAC drugs when SLPI is concomitantly administered with the anti-MAC agents in order to control lung injuries due to the leukocyte protease, which is produced by the neutrophils that accumulate at the sites of MAC infection. Moreover, it is also of interest to examine the effect of nonsteroidal anti-inflammatory drugs (NSAIDs) on the therapeutic efficacy of anti-MAC drugs, since NSAIDs are sometimes administered to MAC patients in order to control host inflammatory reactions elicited by MAC infection and those due to other causes.

A new rifamycin derivative, KRM-1648, is known to exhibit excellent in vitro and in vivo antimicrobial activities against MAC (29). We are now conducting studies to assess the in vivo activity of KRM-1648 against MAC infection; in particular, we are conducting studies concerning interactions between KRM-1648 and other agents, especially anti-inflammatory drugs, which are occasionally administered to control the host inflammatory responses induced by MAC infection. In the study described here we examined the effects of SLPI and an NSAID, diclofenac sodium (diclofenac Na), on the efficacy of KRM-1648 against MAC infection induced in mice. We also studied the effects of these drugs on the cellular functions of MAC-infected Mφs, including the production of cytokines and anti-MAC antimicrobial activity.

MAC N-260 and N-444, which were isolated from patients with MAC infection, were cultured in Middlebrook 7H9 broth (Difco Laboratories, Detroit, Mich.). A recombinant half-sized human SLPI (1/2 SLPI) containing the C-terminal domain (Arg⁵⁸-Ala¹⁰⁷) of SLPI was a gift from the Institute for Biomedical Research, Teijin Limited, Tokyo, Japan, and was used as the SLPI preparation for the experiments. The antiprotease activity of native SLPI is almost completely retained in 1/2 SLPI, except that the activity of the latter is more specific for elastase than for trypsin (19, 21). Human SLPI has 68% amino acid homology with mouse SLPI (16), and the human 1/2 SLPI preparation was demonstrated to be efficacious in ameliorating chemically induced pulmonary fibrosis in hamsters (20). KRM-1648 was obtained from Kaneka Corporation, Hyogo, Japan.

Intracellular growth of MAC in Mφs was measured as described previously (25). Briefly, Mφ monolayer cultures prepared by seeding 10⁶ zymosan A-induced peritoneal exudate cells from 10- to 12-week-old BALB/c mice on 16-mm culture wells (24-well flat-bottom plates; Becton Dickinson & Company, Lincoln Park, N.J.) were incubated in 0.5 ml of RPMI 1640 medium (Nissui Pharmaceutical Co., Tokyo, Japan) supplemented with 5% fetal bovine serum (FBS) (Bio Whittaker Co., Walkersville, Md.) at 37°C for 2 h in a CO₂ incubator (5% CO₂, 95% humidified air). In this study, we used zymosan A-induced Mφs, since chemically elicited Mφs mimic the Mφs populations which emigrate from the bloodstream to the sites of infection and which play important roles in the host resistance to mycobacterial infection (22, 27). After the Mφs were washed with Hanks’ balanced salt solution (HBSS) containing 2% FBS, the Mφs were incubated in 0.5 ml of the medium containing 4 × 10⁶ CFU of MAC N-260 per ml at 37°C in a CO₂ incubator for 2 h. After the MAC-infected Mφs were rinsed with 2% FBS–HBSS, MAC-infected Mφs were cultivated in 1.0 ml of the medium in the presence or absence of each test drug (1/2 SLPI or diclofenac Na) for up to 5 days. At intervals, the Mφs were lysed by a 10-min treatment with 0.07% (wt/vol) sodium dodecyl sulfate followed by neutralization with 2.2% bovine serum albumin (BSA), and the numbers of CFU in the resultant Mφ lysate were counted on 7H11 agar plates.

Cytokine production by MAC-stimulated Mφs was measured as described previously (30, 31), with slight modifications. Briefly, a Mφ monolayer culture prepared on a 14-mm plastic culture sheet (Wako Pure Chemical Industry, Osaka, Japan) was immersed in culture medium (1 ml) in a 16-mm culture well containing each test drug (1/2 SLPI or diclofenac Na) and the mixture was preincubated at 37°C for 16 h in a CO₂ incubator. Then, each Mφ monolayer sheet was transferred to a new culture well containing fresh medium (1 ml) to which the corresponding test drugs had been added. After the addition of 10⁷ CFU of MAC N-260 per ml and allowance for cell-to-cell contact of Mφs with the organisms, the Mφs were further cultivated at 37°C in a CO₂ incubator for up to 7 days. The tumor necrosis factor alpha (TNF-α), interleukin-10 (IL-10), and transforming growth factor β (TGF-β) concentrations were measured in the fluids from the 1- or 7-day cultures of the MAC-stimulated Mφs. Immulon 4 plates (Dynatech Laboratories, Chantilly, Va.) were coated with a capture antibody (Ab) for each cytokine by using rat anti-mouse TNF-α (Pharmingen Co., San Diego, Calif.), rat anti-mouse IL-10 (Genzyme Co., Cambridge, Mass.), or mouse anti-human TGF-β (also specific to mouse TGF-β) (Genzyme) Ab. After blocking of the capture Ab-coated wells with 1% BSA dissolved in phosphate-buffered saline (PBS) and subsequently in 0.1% BSA–PBS, the sample culture fluids (100 μl) of MAC-infected Mφs were poured onto the wells coated with each of the capture Abs, and the individual cytokines contained in the Mφ culture fluids were allowed to bind to the corresponding capture Abs. After rinsing of the wells with 0.1% BSA–PBS, either biotinylated rat anti-mouse TNF-α (Pharmingen) Ab, biotinylated rat anti-mouse IL-10 (Pharmingen) Ab, or chicken anti-human TGF-β Ab (R & D Systems Inc., Minneapolis, Minn.) was added as the detecting Ab and was allowed to react with the complex consisting of the corresponding cytokines and capture Abs. In the cases of the assays with TNF-α and IL-10, after binding of alkaline phosphatase-conjugated streptavidin (Life Technologies Inc., Gaithersburg, Md.) to the resultant complex consisting of the biotinylated detecting Ab, cytokine, and capture Ab, color development was performed with p-nitrophenyl phosphate tablets (Sigma Chemical Co., St. Louis, Mo.) as the substrate. In the case of the assay with TGF-β, after binding of alkaline phosphatase-conjugated rabbit anti-chicken or anti-turkey immunoglobulin G (IgG) Ab (Zymed Laboratories Inc., San Francisco, Calif.) to the complex consisting of the chicken anti-human TGF-β Ab, TGF-β, and capture Ab used for detection, color development was performed with p-nitrophenyl phosphate as the substrate.

Notably, our enzyme-linked immunosorbent assay (ELISA) system could detect only the active form TGF-β but not the latent form of TGF-β (34). Therefore, in order to estimate the concentration of whole TGF-β (active form of TGF-β plus latent form of TGF-β), test Mφ culture fluids were pretreated with 0.1 N HCl for 30 min and were subsequently neutralized with 1 N NaOH–25 mM HEPES buffer before they were tested by ELISA. By this acid treatment, the latent form of the TGF-β molecules could be converted to the active form of TGF-β (32).

The mice were experimentally infected as follows. Six-week-old female BALB/c mice infected intravenously with 10⁷ CFU of MAC organisms were given KRM-1648 finely emulsified in 0.1 ml of 2.5% gum arabic–0.2% Tween 80 once weekly (see Table 3, experiment 1) or once daily, five times per week (see Table 3, experiment 2), from day 1 for up to 8 weeks after infection. The 1/2 SLPI dissolved in saline was injected intraperitoneally to mice once weekly from day 1 for up to 8 weeks after infection. Diclofenac Na (a gift from Novartis Pharma Co., Tokyo, Japan) dissolved in saline was given to mice by gavage once daily, five times per week, from day 1 after infection for up to 8 weeks. At day 1 and week 8 after infection, the mice were killed and examined for bacterial loads in the lungs by counting the numbers of CFU in the homogenates of individual organs by using Middlebrook 7H11 agar plates.

TABLE 3.

Effects of 1/2 SLPI and diclofenac Na on mode of growth of MAC organisms in lungs of host mice and on expression of therapeutic effects of KRM-1648^a

Expt no. and drug	Dose (mg/kg)	Bacterial load in the lungs (log CFU/ organ)b
Expt 1
None		5.96 ± 0.11c
KRM-1648	20	4.73 ± 0.22d
1/2 SLPI	100	5.85 ± 0.16e
KRM-1648 + 1/2 SLPI	20/100	4.81 ± 0.07df
Expt 2
None		5.82 ± 0.05g
KRM-1648	20	4.41 ± 0.18d
Diclofenac Na	1.25	5.76 ± 0.03e
KRM-1648 + diclofenac Na	20/1.25	4.22 ± 0.04df

^aMice were infected with MAC N-444 (experiment 1) or N-260 (experiment 2), and the bacterial loads in the lungs were measured 8 weeks after infection. 

^bThe values on day 1 were 4.08 ± 0.07 and 4.32 ± 0.04 log CFU/organ in experiments 1 and 2, respectively. 

^cValues are means ± standard errors of the means (n = 4). 

^dSignificantly less than the value for untreated control mice (P < 0.01; Bonferroni’s multiple t test). 

^eThe difference from the value for untreated control mice was insignificant (P > 0.05). 

^fThe difference from the value for mice given KRM-1648 alone was insignificant (P > 0.5). 

^gValues are means ± standard errors of the means (n = 5). 

As shown in Table 1, both 1/2 SLPI and diclofenac Na had differential modulating effects on cytokine production by MAC-stimulated Mφs. First, the TNF-α production was significantly inhibited by 1/2 SLPI (P < 0.01) even at 1 ng/ml, and the most marked reduction was achieved with 10 ng of 1/2 SLPI per ml (P < 0.005). On the other hand, 1/2 SLPI treatment caused a dose-dependent increase in IL-10 production, peaking with 10 ng of 1/2 SLPI per ml (P < 0.005), while TGF-β production was not affected by such 1/2 SLPI treatment. Second, the TNF-α production was significantly upregulated due to treatment with diclofenac Na at 1 μg/ml (P < 0.005), while IL-10 production was not affected. Moreover, TGF-β production was also enhanced due to treatment with diclofenac Na at doses of 1 to 10 μg/ml (P < 0.005). In this experiment, the inhibitory effect of 1/2 SLPI on the ability of Mφs to produce TNF-α was not due to its cytotoxicity for Mφs, since the number of intact cells attached on a culture sheet after 24 h of cultivation of MAC-stimulated Mφs with or without 1/2 SLPI treatment was not significantly decreased compared to that of drug-untreated control Mφs, as follows: Mφs without drug treatment, (2.9 ± 0.1) × 10⁵ cells; Mφs treated with 10 and 100 ng 1/2 SLPI per ml, (2.8 ± 0.1) × 10⁵ and (2.9 ± 0.2) × 10⁵ cells, respectively; Mφs treated with 10 μg of diclofenac Na per ml, (2.8 ± 0.1) × 10⁵ cells. In this case, the value for the control Mφs, which did not receive MAC stimulation, was (3.4 ± 0.1) × 10⁵ cells. In addition, these drugs did not accelerate the reduction of Mφ viability during 7 days of cultivation, as follows: untreated Mφs, (2.3 ± 0.1) × 10⁵ cells; 1/2 SLPI (100 ng/ml)-treated Mφs, (2.7 ± 0.1) × 10⁵ cells; diclofenac Na (10 μg/ml)-treated Mφs, (2.6 ± 0.1) × 10⁵ cells.

TABLE 1.

Effects of 1/2 SLPI and diclofenac Na on production of TNF-α, IL-10, and TGF-β by MAC-stimulated Mφs^a

MAC infection	Mφ treatment	Concn	Cytokine production (ng/ml)b
			TNF-α	IL-10	TGF-β
−	None		<0.1	<0.1	1.75 ± 0.09
+	None		0.82 ± 0.05	0.77 ± 0.01	4.24 ± 0.06
+	1/2 SLPI	0.1 ng/ml	0.79 ± 0.12	1.10 ± 0.07c	4.02 ± 0.15
+	1/2 SLPI	1 ng/ml	0.51 ± 0.06d	1.63 ± 0.05c	4.57 ± 0.04
+	1/2 SLPI	10 ng/ml	0.41 ± 0.06e	1.92 ± 0.06c	4.48 ± 0.11
+	1/2 SLPI	100 ng/ml	0.52 ± 0.09c	1.51 ± 0.14c	4.15 ± 0.11
+	Diclofenac Na	1 μg/ml	3.17 ± 0.14e	0.64 ± 0.03	7.21 ± 0.25e
+	Diclofenac Na	10 μg/ml	1.34 ± 0.04e	0.71 ± 0.00	8.24 ± 0.18e

^aMφs were preincubated in the presence or absence of the indicated drugs for 16 h. Then, the Mφs were allowed to come into contact with the MAC organisms, and the MAC-stimulated Mφs were further cultivated in medium to which the corresponding drugs which were used for Mφ pretreatment were or were not added. The TNF-α and IL-10 concentrations in the culture fluids were measured on day 1, and the TGF-β concentrations in the culture fluids were measured on day 7. 

^bValues are means ± standard errors of the means (n = 4). 

^cSignificantly different from the control value (P < 0.05; Student’s t test). 

^dSignificantly different from the control value (P < 0.01; Student’s t test). 

^eSignificantly different from the control value (P < 0.005; Student’s t test). 

Since TNF-α is known to increase the activity of Mφs against MAC (6, 11), while IL-10 and TGF-β decrease the activity of Mφs against MAC (4, 5), it is of interest to determine the effects of these drugs on the activity of Mφs against MAC. As indicated in Table 2, we examined the effects of 1/2 SLPI and diclofenac Na on the mode of growth of MAC in Mφs. These agents slightly (statistically insignificantly) inhibited the growth of organisms in Mφs. Therefore, it appeared that the anti-MAC activity of Mφs was not substantially affected by 1/2 SLPI or diclofenac Na, although these two agents had significant modulating effects on the TNF-α or IL-10 production by MAC-infected Mφs (Table 1). Although this finding may indicate that TNF-α is not critical for the manifestation of Mφ activity against MAC, this concept has some limitations, since 1/2 SLPI-mediated inhibition and diclofenac Na-mediated enhancement of Mφ TNF-α production were partial and not total. Moreover, the possibility that, even with inhibition due to 1/2 SLPI treatment, the TNF-α concentration would suffice to trigger Mφ the anti-MAC activity of Mφs cannot be excluded. In addition, the present findings also suggest that IL-10 may not play a decisive role in the downregulation of Mφ anti-MAC activity. In separate experiments, neither 1/2 SLPI (10 ng/ml) nor diclofenac Na (10 μg/ml) inhibited the growth of extracellular MAC organisms in 7H9 medium (data not shown).

TABLE 2.

Effects of 1/2 SLPI and diclofenac Na treatment of Mφs on mode of intracellular growth of MAC organisms

Drug	Concn	CFU/100 Mφsa
		Time zero	Day 1	Day 3	Day 5
None		18.8 ± 1.2	38.2 ± 5.4	38.8 ± 4.9	46.1 ± 8.1
1/2 SLPI	1 ng/ml	NDb	30.2 ± 3.3	31.5 ± 3.3	42.6 ± 2.2
1/2 SLPI	10 ng/ml	ND	38.9 ± 1.0	25.7 ± 2.6c	43.6 ± 10.6
Diclofenac Na	1 μg/ml	ND	38.1 ± 7.0	36.5 ± 5.6	43.9 ± 2.9
Diclofenac Na	10 μg/ml	ND	37.9 ± 3.4	ND	40.1 ± 3.6

^aMAC N-260-infected Mφs were cultured in the presence or absence of the indicated drugs for up to 5 days. Values are means ± standard errors of the means (n = 3). 

^bND, not determined. 

^cOnly in this case was the P value for the difference from the control value <0.1 (Student’s t test). In the other cases, the P values were >0.25. 

Some NSAIDs, including indomethacin, aspirin, and ibuprofen, have been reported to increase the level of TNF-α production by zymosan A- or lipopolysaccharide (LPS)-stimulated Mφs (17, 18). Enhancement of Mφ TNF-α production by these NSAIDs appears to be due to their inhibition of prostaglandin (PG) synthesis, since PGs, especially PGEs, downregulate Mφ TNF-α production (14, 18). In contrast, Tenidap (3-substituted 2-oxindole), a new NSAID which also inhibits PG synthesis, has been reported to suppress Mφ TNF-α production (7). In the present study, diclofenac Na, which also inhibits PG synthesis, also enhanced the TNF-α production by MAC-stimulated Mφs. Thus, diclofenac Na appears to modulate Mφ TNF-α production through activation of PG synthesis, as in the cases of the usual NSAIDs such as indomethacin and aspirin.

It is known that IL-10 primarily mediates anti-inflammatory reactions through the suppression of Mφ activity related to inflammation, such as the production of reactive oxygen intermediates, reactive nitrogen intermediates, and some proinflammatory cytokines including TNF-α, IL-1, and IL-8 (8, 26). Since 1/2 SLPI upregulates IL-10 production by MAC-stimulated Mφs, it may be thought that 1/2 SLPI induces IL-10-mediated anti-inflammatory reactions in MAC-infected host animals. Moreover, TGF-β also mediates anti-inflammatory reactions by suppressing Mφ functions related to inflammation (4, 9, 26, 33). Because diclofenac Na upregulates the TGF-β-producing function of MAC-stimulated Mφs, it appears that in vivo the anti-inflammatory effects of this drug are partly mediated by TGF-β.

The present finding that 1/2 SLPI moderately inhibited TNF-α production by MAC-stimulated Mφs is consistent with the finding of Jin et al. (16) that overexpression of the SLPI gene in murine Mφs reduced the level of TNF-α production by LPS-stimulated Mφs, presumably by decreasing the level of NF-κB expression induced by LPS signaling. It appears that the TNF-α production by MAC-infected Mφs is mainly triggered by a mycobacterial lipoglycan, lipoarbinomannan, since no inhibitory effect of SLPI gene expression on Mφ TNF-α production is observed when Mφs are stimulated with other agents such as gamma interferon (16).

Either 1/2 SLPI or diclofenac Na may be administered to MAC-infected patients who are being treated with multidrug regimens that include KRM-1648 for the clinical management of MAC infection. It is therefore of interest to examine the effects of these anti-inflammatory drugs on the therapeutic efficacy of KRM-1648 against MAC infection. As shown in Table 3, neither 1/2 SLPI nor diclofenac Na affected the mode of progression of MAC infection in mice, as measured by the increase in bacterial load in the lungs in drug-treated mice compared to that in untreated control mice. KRM-1648 significantly decreased the bacterial loads in the lungs (P < 0.05). Notably, neither 1/2 SLPI nor diclofenac Na affected the therapeutic efficacy of KRM-1648. This finding is of importance to the management of MAC infection with regimens that include KRM-1648, since it suggests that the therapeutic efficacy of KRM-1648 is not decreased even when patients with MAC infection receive 1/2 SLPI or diclofenac Na for other purposes, such as control of degenerative and inflammatory disorders in the lungs which are caused by MAC infection itself or by other infectious agents. Since multidrug regimens that include clarithromycin, which has excellent therapeutic activity in MAC patients (15), rifamycin derivatives (rifampin, rifabutin, rifapentine, and KRM-1648), and other drugs are potent regimens that are efficacious in controlling MAC infections (2, 3, 23), it is important to examine the influence of concomitant administration of 1/2 SLPI or diclofenac Na with such anti-MAC multidrug regimens. Further studies are under way to elucidate the effects of these anti-inflammatory drugs on the therapeutic efficacies of multidrug regimens that include these anti-MAC agents.