INTRODUCTION

Top Abstract Introduction Materials and methods Results Discussion References

Mycobacterium avium complex (MAC) infections are frequently encountered in patients with AIDS and in other types of immunocompromised hosts (35). Clinical management of MAC infections is difficult, since MAC organisms are resistant to common antituberculosis drugs such as isoniazid, ethambutol, pyrazinamide, and rifampin (4). Although some new drugs, including clarithromycin and rifabutin, are fairly effective in controlling MAC bacteremia in AIDS patients (4, 17), the treatment of pulmonary MAC infections is still difficult, even with the use of multidrug regimens containing these drugs (5, 17).

Recently, KRM-1648 (KRM), a new benzoxazinorifamycin with excellent anti-MAC activity, has been developed (19, 26, 31). Although one research group reported that KRM was inefficacious in controlling MAC infection induced in mice when drug treatment was initiated after the establishment of a severe infection (22), many investigations indicated that this new rifamycin derivative has potent therapeutic efficacy against MAC infections in mice and rabbits (8, 29, 31). Therefore, KRM may be useful for clinical control of intractable MAC infections in humans. Indeed, this drug is in phase I trials as a new component of multidrug regimens for the treatment of MAC infection and tuberculosis.

The Chinese medicine mao-bushi-saishin-to (MBST), which is a mixture of extracts from three medicinal herbs, mao, saishin, and hou-bushi, has long been used in Japan for treatment of the common cold (23). This drug is also efficacious in controlling perennial nasal allergy (20, 25). Indeed, it has been demonstrated to suppress experimental passive cutaneous anaphylaxis induced in rats, presumably by inhibiting histamine release from mast cells (27, 28). In Japan, MAC patients receiving KRM might use MBST for treatment of respiratory infections and nasal allergy. Thus, it is important to assess drug-to-drug interaction between KRM and MBST, which may be induced in vivo when both drugs are concomitantly administered to MAC patients. In the present study, we have examined the effects of MBST on the therapeutic effect of KRM against MAC infection induced in mice. We found that MBST did not interfere with but moderately increased the therapeutic efficacy of KRM by potentiating KRM-mediated killing and inhibition of MAC organisms residing in host macrophages (Ms).

	MATERIALS AND METHODS

Top Abstract Introduction Materials and methods Results Discussion References

Microorganisms. MAC N-444 (serovar 8), which we previously isolated from patients with MAC infection and identified as M. avium by DNA probe testing, was cultured in Middlebrook 7H9 broth (Difco Laboratories, Detroit, Mich.) supplemented with 10% (vol/vol) albumin-dextrose-catalase enrichment and 0.05% (vol/vol) Tween 80, and a bacterial suspension prepared with phosphate-buffered saline (PBS) containing 0.1% (wt/vol) bovine serum albumin was frozen at 80°C until use. The MAC organisms yielded smooth, transparent, and flat colonies on Middlebrook 7H11 (Difco) agar plates, a characteristic of virulent colonial variants of MAC.

Mice. Female BALB/c mice were purchased from Japan Clea Co., Osaka, Japan. When BALB/c mice were intravenously (i.v.) infected with 3.8 × 10⁶ CFU of M. avium N-444, progressive bacterial growth was observed at sites of infection, resulting in very heavy bacterial loads in the visceral organs at 1 year after infection: 9.2 and 8.6 log U in the lungs and spleen, respectively.

Special agents. MBST was obtained from Kotaro Pharmaceutical Co., Osaka, Japan. For preparation of MBST, a mixture of three medicinal herbs including Mao, Saishin, and Hou-Bushi was extracted with hot water (100°C) for 1 h, filtered, and then lyophilized. MBST prepared from the same batch was used throughout the experiments. MBST powder was initially dissolved in PBS and subsequently diluted with RPMI 1640 medium (Nissui Pharmaceutical Co., Tokyo, Japan) supplemented with 5% fetal bovine serum (FBS) (BioWhittaker Co., Walkersville, Md.) before use in in vitro experiments. MBST contains a number of pharmacologically active components, including mao-derived l-ephedrine and ephedran; saishin-derived methyleugenol, elemicin, l-asarinin, and hygenamine; and hou-bushi-derived aconitine, coryneine, and mesaconitine. KRM was obtained from Kaneka Corporation, Hyogo, Japan, and finely emulsified in 2.5% gum arabic-0.2% Tween 80 before use in in vivo experiments.

Experimental infection. Six-week-old BALB/c mice infected i.v. with 10⁷ CFU of MAC N-444 were given no drug or KRM finely emulsified in 0.1 ml of 2.5% gum arabic-0.2% Tween 80 and/or MBST dissolved in saline by gavage once daily five times per week from day 1 after infection for up to 8 weeks. Doses of KRM and MBST were fixed to be nearly equivalent to their clinical dosages by weight as follows: KRM, 20 mg/kg; MBST, 50 or 100 mg/kg. At day 1 and week 8, mice were sacrificed and examined for bacterial loads in the lungs by counting the CFU in the homogenates of individual organs using Middlebrook 7H11 agar plates. When healthy volunteers (53- to 80-kg body weights) were orally administered 400 mg of MBST containing 5 mg of l-ephedrine, the maximum drug concentration in serum and the area under the concentration-time curve from 0 to 24 h of l-ephedrine in the blood were estimated to be 19.5 ± 0.7 ng/ml and 192 ± 9 ng · h/ml, respectively (21). Similar pharmacokinetics of l-ephedrine has been reported for mice orally administered a crude aqueous extract of ephedra (mao, one of the major components of MBST) (16).

Intracellular growth of MAC in Ms. M monolayer cultures prepared by seeding 3 × 10⁵ zymosan A (1 mg)-induced peritoneal exudate cells of 10- to 12-week-old BALB/c mice in 6.0-mm-diameter culture wells (96-well flat-bottom plate; Becton Dickinson & Company, Lincoln Park, N.J.) were incubated in 0.2 ml of 5% FBS-RPMI medium with or without the addition of MBST at concentrations of 1 to 100 µg/ml at 37°C for 2 days in a CO₂ incubator (5% CO₂-95% humidified air). After washing with Hanks' balanced salt solution (HBSS) containing 2% FBS, the Ms were incubated in 0.1 ml of the medium containing 1.5 × 10⁶ CFU of MAC N-444 per ml at 37°C in a CO₂ incubator for 2 h. The MAC-infected Ms were then washed with 2% FBS-HBSS to remove extracellular organisms and thereafter cultivated in 0.2 ml of 5% FBS-RPMI medium in the presence (KRM, 1 µg/ml; MBST, 1 to 100 µg/ml) or the absence of each test drug for up to 7 days. At intervals, the Ms were lysed by 10 min of treatment with 0.07% (wt/vol) sodium dodecyl sulfate, and the cell lysate (0.28 ml) was mixed with 0.12 ml of PBS containing 20% bovine serum albumin to neutralize the sodium dodecyl sulfate. After collection of bacterial cells from the resultant M lysate by centrifugation at 2,000 × g for 15 min and subsequent washing of recovered bacteria with distilled water by centrifugation, the CFU were counted on 7H11 agar plates.

M production of RNI. M production of reactive nitrogen intermediates (RNI) was measured as described previously (1). Briefly, M monolayer cultures in 16-mm-diameter culture wells (24-well flat-bottom plate; Becton Dickinson) pretreated with MBST for 2 days as described above were incubated in 0.5 ml of 5% FBS-RPMI 1640 medium containing 5 × 10⁷ CFU of MAC N-444 per ml at 37°C in a CO₂ incubator for 2 h. After washing with 2% FBS-HBSS, the MAC-infected Ms were cultivated in 5% FBS-RPMI 1640 at 37°C for 24 h. Culture supernatants of the Ms were allowed to react with Griess reagent, and the nitrite content was quantitated by measuring the A₅₄₀.

IL-10 and TGF- production by Ms. The 2- or 7-day culture fluids of MAC-infected Ms with or without the MBST treatment described above were measured for interleukin-10 (IL-10) and transforming growth factor  (TGF-) concentrations as previously described (32). Briefly, Immulon 4 plates (Dynatech Laboratories, Chantilly, Va.) were coated with a capture antibody (Ab) for each cytokine using a rat anti-mouse IL-10 monoclonal Ab (MAb) (Genzyme Co., Cambridge, Mass.) or a mouse anti-human TGF- MAb (also specific to mouse TGF-) (Genzyme). A biotinylated rat anti-mouse IL-10 MAb (Pharmingen Co., San Diego, Calif.) or a chicken anti-human TGF- Ab (R & D Systems Inc., Minneapolis, Minn.) was used as the detecting Ab. After binding of alkaline phosphatase-conjugated streptavidin (Life Technologies Co., Gaithersburg, Md.) to biotinylated MAbs (IL-10 assay) or an alkaline phosphatase-conjugated rabbit anti-chicken-turkey immunoglobulin G Ab (Zymed Laboratories Inc., San Francisco, Calif.) to a chicken anti-human TGF- Ab (TGF- assay), color development was performed by using p-nitrophenyl phosphate tablets (Sigma Chemical Co., St. Louis, Mo.) as the substrate.

Expression of cytokine mRNAs. Reverse transcription (RT)-PCR analysis of cytokine mRNAs in lung tissues from mice infected with MAC was performed as described by Ashman et al. (3) with slight modifications. Total RNA was isolated from lung tissues harvested at weeks 4 and 8 after infection from MAC-infected mice with or without drug treatment by using the ISOGEN kit (Nippon Gene Co., Toyama, Japan). After DNase I (GIBCO-BRL, Rockville, Md.) treatment (1 U of DNase/µg of RNA sample) at room temperature for 15 min, the resultant RNA samples were reverse transcribed to the first chain of cDNA by using random hexamer primers (GIBCO) and 200 U of Superscript II reverse transcriptase (GIBCO) with a standard reaction mixture (20 µl) containing 1× RT buffer (pH 8.3); 1 mM each dATP, dCTP, dGTP, and dTTP (GIBCO); and 2.0 U of RNase inhibitor (GIBCO). After a 1-h reaction at 42°C and subsequent heating at 72°C for 15 min, 1-µl aliquots of the resultant cDNA were amplified specifically by PCR in a standard reaction mixture (50 µl) containing 1× PCR buffer (pH 8.3); 0.2 mM each dATP, dCTP, dGTP, and dTTP; 1 U of Taq polymerase (Takara Biomedicals Co., Tokyo, Japan); and 20 pmol of the sense and antisense primers for the test cytokines (synthesized by Greiner Labortechnik Co., Tokyo, Japan) as follows (sense/antisense): TNF-, AGCCCACGTCGTAGCAAACCACCAA/ACACCCATTCCCTTCACAGAGCAAT (14); IFN-, GAAAGCCTAGAAAGTCTGAATAACT/ATCAGCAGCGACTCCTTTTCCGCTT (14); IL-10, TGACTGGCATGAGGATCAGCAG/ATCCTGAGGGTCTTCAGCTT (34); TGF-₁, AGCCCTGGATACCAACTATTGCTTCAGCTCCACAG/AGGGGCGGGGCGGGGCGGGGCTTCAGCTGC (24). Reactions were carried out in a DNA Thermal Cycler (ASTEC Corp., Fukuoka, Japan) for 30 cycles including denaturing at 94°C for 1 min, annealing at 58°C for 2 min, and extension at 72°C for 2 min for each cycle. PCR products were analyzed by electrophoresis on ethidium bromide-stained 2% agarose gels.

Statistical analysis. Statistical analysis was performed by using Bonferroni's multiple t test or the Mann-Whitney test.

	RESULTS

Top Abstract Introduction Materials and methods Results Discussion References

Effects of MBST on the therapeutic activity of KRM against MAC infection. Tables 1 and 2 show the effects of MBST on the therapeutic efficacy of KRM against MAC infection in mice. The doses of KRM (20 mg/kg) and MBST (50 or 100 mg/kg) were nearly equivalent to their clinical dosages by weight. MBST at these doses caused no acute or subacute toxicity to mice and rats when given by gavage once daily for 13 weeks (data not shown). Table 1 indicates bacterial loads in the lungs and spleen at week 4 after infection. KRM displayed significant therapeutic efficacy in terms of reduction of bacterial growth in the lungs and spleens of KRM-treated mice compared to those of untreated control mice (P < 0.05). Notably, the therapeutic efficacy of KRM, in terms of MAC growth inhibition in the lungs, was moderately increased when it was given in combination with MBST (P < 0.05 in the Mann-Whitney test), although MBST alone did not exhibit a significant therapeutic effect against MAC infection.

Effects of MBST on M anti-MAC activity. It is of interest to study the immunological mechanisms of the increase in the therapeutic efficacy of KRM due to combined administration with MBST. In the next series of experiments, we investigated the effects of MBST on some M functions related to their antimicrobial activity, since Ms are known to play crucial roles in the expression of host resistance to mycobacterial infections (9).

View larger version (10K): [in this window] [in a new window]   FIG. 1.   Effects of MBST on M anti-MAC activity. Ms were preincubated in culture medium in the absence or presence of 1-, 10-, or 100-µg/ml MBST for 2 days, infected with MAC for 2 h, and further cultured in the same medium with or without the addition of MBST at corresponding doses (1, 10, or 100 µg/ml, respectively), in the absence (A) or the presence (B) of 1-µg/ml KRM, for up to 7 days. Symbols in panel A: , without MBST treatment; , , , treated with 1-, 10-, or 100-µg/ml MBST, respectively. Symbols in panel B: , no drug added; , KRM alone; , , treatment with KRM plus MBST at 10 and 100 µg/ml, respectively. Each symbol indicates the mean ± the standard error of the mean (n = 3). The asterisk indicates a value significantly smaller than that of Ms treated with KRM alone (P < 0.05 by Bonferroni's multiple t test).

View this table: [in this window] [in a new window]   TABLE 3.   Effect of MBST on RNI production by MAC-infected Msa

Effects of MBST on IL-10 and TGF- production by Ms. In the next experiment, we examined the effect of MBST on M production of IL-10 and TGF-, immunoregulatory cytokines which down-regulate M anti-MAC activity (6, 7, 10). As shown in Table 4, treatment of Ms with MBST at doses of 10 to 100 µg/ml caused partial reductions in the accumulation of IL-10 in culture fluids by MAC-infected Ms in the early phase of cultivation (day 2). The transiently increased level of IL-10 was thereafter decreased to undetectable levels by day 7 (data not shown). In the same experiment, the accumulation of TGF- by MAC-infected Ms was below the limit of detection in the early phase (day 2), followed by low-level accumulation in the middle phase (day 7) of M cultivation (data not shown). In this case, MBST treatment caused no significant change in M TGF- production as follows: MBST (100 µg/ml)-treated Ms, 0.29 ng/ml; untreated Ms, 0.33 ng/ml.

View this table: [in this window] [in a new window]   TABLE 4.   Effect of MBST on IL-10 production by MAC-infected Msa

Expression of RNA messages of cytokines in the lungs of MAC-infected mice with or without drug treatment. Profiles of the expression of the mRNAs of proinflammatory cytokines tumor necrosis factor alpha (TNF-) and gamma interferon (IFN-) and the immunosuppressive cytokines IL-10 and TGF- in the lung tissues of MAC-infected mice given KRM, MBST, or both were determined by using RT-PCR analysis. Figures 2 and 3 show representative results for the lungs of six mice separately subjected to RT-PCR assay.

View larger version (56K): [in this window] [in a new window]   FIG. 2.   Effects of MBST on the expression of TNF-, IFN-, IL-10, and TGF- mRNAs in the lungs of MAC-infected mice at week 4 after infection. Induction of MAC infection and subsequent drug treatment of infected mice were as described in Table 1. RT-PCR analysis of the lungs of mice was done at week 4. The relative intensities of the RT-PCR bands of individual cytokines were calculated by normalizing to the intensity of the -actin band. The values in parentheses are cytokine band/-actin band ratios (the mean of six mice). The standard error of the mean ranged from 5 to 28% (TNF-), from 14 to 29% (IFN-), from 11 to 28% (IL-10), and from 3 to 10% (TGF-).

View larger version (65K): [in this window] [in a new window]   FIG. 3.   Effects of MBST on the expression of TNF-, IFN-, IL-10, and TGF- mRNAs in the lungs of MAC-infected mice at week 8 after infection. Induction of MAC infection and subsequent drug treatment of infected mice were as described in Table 2. The other details are the same as those described in the legend to Fig. 2, except that mice were sacrificed at week 8.

	DISCUSSION

Top Abstract Introduction Materials and methods Results Discussion References

The present study has revealed that the Chinese medicine MBST is capable of moderately potentiating the therapeutic efficacy of KRM against MAC infection (Tables 1 and 2). This effect of MBST appears to be related to its ability to modulate M anti-MAC activity, since M treatment with MBST potentiated the microbicidal activity of KRM against intra-M MAC. As indicated in Table 3, MBST reduced the RNI-producing ability of MAC-infected Ms. In preliminary experiments, M production of reactive oxygen intermediates (ROI) was also down-regulated due to MBST treatment. Therefore, it may be thought that MBST-treated Ms preferentially use microbicidal effectors other than RNI and ROI. This concept is consistent with our previous findings that the degree of resistance of various MAC strains to RNI and ROI did not correlate with their virulence in mice (30), indicating that RNI and ROI alone are not decisive in the host defense against MAC. Collaboration among RNI, ROI, and other M antimicrobial effector molecules, such as free fatty acids and bactericidal peptides (1, 9, 18), may be required for expression of the anti-MAC activity of MBST-treated Ms.

As shown in Table 4, MBST treatment suppressed IL-10 production by MAC-infected Ms. Since IL-10 is an immunoregulatory cytokine with M-deactivating activity and is capable of down-regulating M anti-MAC activity (7, 10, 11), MBST-mediated potentiation of KRM activity against MAC residing in Ms was due in part to reduction of M IL-10 production by the action of MBST. On the other hand, MBST did not affect M TGF- production. This finding has some importance in relation to MBST administration in patients with MAC infection, since it excludes the possibility that MBST promotes the progression of MAC infection by enhancing M production of TGF-, a M-deactivating cytokine which down-regulates the anti-MAC activity of Ms (6).

RT-PCR analysis of cytokine mRNA expression in the lung tissues of MAC-infected mice with or without drug administration revealed the following. MAC infection elevated the levels of TNF-, IFN-, and IL-10 mRNAs at weeks 4 and 8, while such an effect on TGF- mRNA was obscure. Similar profiles have also been reported for TNF-, IFN-, and IL-10 by other investigators (2, 12, 15). KRM treatment tended to decrease the levels of all of the test cytokine mRNAs except TGF- mRNA at week 4. This is consistent with our previous finding that KRM administration reduced the expression of TNF-, IFN-, and IL-10 protein levels in the lungs and spleens of MAC-infected mice at weeks 4 and 8 (32). In this context, it has been recently reported that a potent anti-MAC drug, clarithromycin, reduced IFN- mRNA expression in spleen cells of MAC-infected mice at week 4 postinfection (13). In addition, the reduction of IL-10 mRNA levels in MBST-treated mice at this stage may be due to MBST-mediated down-regulation of the IL-10-producing ability of host Ms, as evidenced in Table 3.

While MBST treatment caused a reduction in the levels of all of the test cytokine mRNAs at week 4 after infection, such effects of MBST became obscure at week 8. Moreover, the profiles of cytokine mRNA expression in mice given the KRM-plus-MBST combination differed somewhat, depending on the stage of infection, i.e., between week 4 and week 8. These situations may be related more or less to the fact that cytokines involved in host protection are different from stage to stage of MAC infection (2). For instance, it has been reported that TNF- and IFN- are involved in early protection, while IFN-, but not TNF-, plays a role at later time points of infection (2).

Notably, at week 8 after infection, the combination of MBST with KRM caused a marked decrease in cytokine mRNA expression compared to that in mice given KRM alone. This is due in part to the reduction of bacterial loads in the lung tissues achieved by combination therapy of KRM with MBST. The remarkable decrease in the levels of the proinflammatory cytokine TNF- and IFN- mRNAs may be attributable to down-regulation of inflammatory reactions in host mice receiving treatment with KRM plus MBST. Such mice also showed lowered expression of IL-10 and TGF- mRNAs, both of which down-regulate M anti-MAC activity (6, 7, 10, 11, 33). This might be the reason for the potentiation of the chemotherapeutic efficacy of KRM due to combined administration with MBST, particularly at week 8 after infection.

In any case, the present finding that MBST increased the therapeutic activity of KRM against MAC infection appears to be of importance for clinical management of MAC infections using rifamycin derivatives, including KRM. This finding indicates that there may be no unfavorable drug-to-drug interaction between KRM and MBST, even if these drugs are concomitantly administered to MAC patients. That is, the therapeutic efficacies of rifamycins are not decreased when patients with MAC infection receiving KRM treatment are furthermore given MBST for other purposes, such as clinical control of the common cold and perennial nasal allergy. Further studies are currently under way in order to elucidate the effects of MBST on the T-cell functions of MAC-infected mice.