In Vitro Activity of a Novel Antimicrobial Agent, TG44, for Treatment of Helicobacter pylori Infection

ABSTRACT Due to concerns about the current therapeutic modalities for Helicobacter pylori infection, e.g., the increased emergence of drug-resistant strains and the adverse reactions of drugs currently administered, there is a need to develop an anti-H. pylori agent with higher efficacy and less toxicity. The antibacterial activity of TG44, an anti-H. pylori agent with a novel structural formula, against 54 clinical isolates of H. pylori was examined and compared with those of amoxicillin (AMX), clarithromycin (CLR), and metronidazole (MNZ). Consequently, TG44 inhibited the growth of H. pylori in an MIC range of 0.0625 to 1 μg/ml. The MIC ranges of AMX, CLR, and MNZ were 0.0078 to 8 μg/ml, 0.0156 to 64 μg/ml, and 2 to 128 μg/ml, respectively. The antibacterial activity of TG44 against AMX-, CLR-, and MNZ-resistant strains was nearly comparable to that against drug-susceptible ones. In a pH range of 3 to 7, TG44 at 3.13 to 12.5 μg/ml exhibited potent bactericidal activity against H. pylori in the stationary phase of growth as early as 1 h after treatment began, in contrast to AMX, which showed no bactericidal activity at concentrations of up to 50 μg/ml at the same time point of treatment. TG44 at 25 μg/ml exhibited no antibacterial activity against 13 strains of aerobic bacteria, suggesting that its antibacterial activity against H. pylori is potent and highly specific. The present study indicated that TG44 possesses antibacterial activity which manifests quickly and is potentially useful for eradicating not only the antibiotic-susceptible but also the antibiotic-resistant strains of H. pylori by monotherapy.

In recent years, the association between Helicobacter pylori and gastritis, gastric ulceration, and duodenal ulceration (14,21), as well as gastric cancer, has been clarified (7,24,28). The eradication of H. pylori drastically reduces the recurrence of ulceration and is therefore considered essential to treat ulceration. Currently, proton pump inhibitor-based triple therapy using a proton pump inhibitor and two antibiotics is frequently conducted to eradicate H. pylori in patients with gastric and/or duodenal ulcer. Among the antibiotics frequently used are amoxicillin (AMX), clarithromycin (CLR), and metronidazole (MNZ). Many clinical studies have reported eradication rates of 80 to Ն90%, attained by the relevant triple therapy (1,2,3,8,9,15). However, many concerns remain to be addressed in the future, including the increased emergence of drug-resistant strains of H. pylori due to the overuse of antibiotics (4,5,6,11,12,16,17,19,23,25,26,27,29) and the indiscreet use of eradication therapy for the bacterium, as well as the adverse reactions (e.g., diarrhea, dysgeusia, and eruption) to the drugs administered. Therefore, there is a strong need to develop an anti-H. pylori agent which is suitable for the next generation of eradication therapy. Ideally, such an agent is expected to satisfy the following requisites: (i) potent antibacterial activity against H. pylori (administrable by mono-therapy); (ii) high specificity for H. pylori (without efficacy for other intestinal bacteria); (iii) bactericidal activity against AMX-, CLR-, or MNZ-resistant strains of H. pylori; (iv) stability in the stomach (10); (v) possible synergism with other drugs; and (vi) less likelihood of generating drugresistant strains of H. pylori.
In the present study, we used 56 strains of H. pylori, including two reference strains and 54 clinical isolates, to examine the antibacterial activity of TG44 in comparison with those of three antibiotics which are frequently used for the eradication of the bacterium, AMX, CLR, and MNZ.
Strains. The fifty-six strains of H. pylori used were supplied by the following sources: the American Type Culture Collection provided 2 reference strains (ATCC 43504 and ATCC 43629), S. Kamiya (Department of Microbiology, Kyorin University School of Medicine, Tokyo, Japan) provided 12 clinical isolates (KR 2098, TK 1003, TK 1025, TK 1027, TK 1030, TK 1042, TK 1047, TK  1126, TK 1147, TK 1307, TK 1308, and TK  In this assay system, ␤-CyD was used as a growth factor for H. pylori (20,22) as an alternative to bovine serum in an attempt to avoid the enzymatic inactivation of TG44 by plasma esterase.
Determination of the MICs for H. pylori. The MICs were determined by the agar dilution method using BHIA. Under microaerobic conditions (5% O 2 , 10% CO 2 , and 85% N 2 ) in an AnaeroPack Campylo jar (Mitsubishi Gas Chemical Co., Inc., Tokyo, Japan), the stock cultures of H. pylori were grown at 35°C for 24 h in BHIB supplemented with 0.1% ␤-CyD by shaking them on a shaker at 125 rpm. H. pylori cultures in the exponential phase of growth were diluted and adjusted to approximately 10 7 CFU/ml with BHIB. A standardized loop (diameter, 1 mm; streak, about 2 cm) was used to seed the bacterial suspension onto the BHIA plate, supplemented with 0.1% ␤-CyD and containing twofold serial dilutions of the compound. The agar plate was inverted and incubated at 35°C under microaerobic conditions for 3 days. The MIC was defined as the lowest concentration at which the compound inhibited visible bacterial growth.
Determination of the MICs for aerobic bacteria. The MICs were determined by the agar dilution method. The stock cultures of 13 reference strains of aerobic bacteria were grown at 37°C for 20 h in Mueller-Hinton S broth, and cultures of bacteria in the exponential phase of growth were diluted and adjusted to approximately 10 6 CFU/ml. A multipoint inoculator (1 l of sample) was used to seed the bacterial suspension onto Mueller-Hinton S agar containing twofold serial dilutions of the compound. The agar plate was inverted and incubated at 37°C for 20 h. The MIC was defined as the lowest concentration at which the test compound inhibited visible bacterial growth.
Bactericidal activity in a short-term assay. The bactericidal activities of TG44 and AMX were assessed at concentrations of 1/2ϫ the MIC, 1ϫ the MIC, 2ϫ the MIC, 4ϫ the MIC, and 8ϫ the MIC of each compound using H. pylori ATCC 43629. The bactericidal activity of each compound was also assessed at concentrations of 3.13 to 50 g/ml using H. pylori ATCC 43504. Under microaerobic conditions, the stock culture of H. pylori was grown at 37°C for 3 days in BHIB supplemented with 0.1% ␤-CyD. Cultures of bacteria in the exponential phase of growth were diluted and adjusted to approximately 2 ϫ 10 7 CFU/ml with BHIB. The bacterial suspension (20 l) was seeded into 4 ml of BHIB supplemented with 0.1% ␤-CyD and containing twofold serial dilutions of each of the compounds. The culture was incubated, and aliquots were collected at various time points. Each sample was serially diluted 10-fold with saline, and 10 l of the diluted sample was plated on BHIA supplemented with 7% horse blood (TG44treated sample) or on BHIA supplemented with 7% horse blood and 5% pen-  Effects of pH on bactericidal activity. At pH 7, 6, 5, and 3, H. pylori ATCC 43504 was used to examine the bactericidal activity of TG44 at 1ϫ, 2ϫ, 4ϫ, 8ϫ, and 16ϫ the MIC of the compound. The bacterial suspension (20 l at 2 ϫ 10 7 CFU/ml) was seeded into 4 ml of BHIB supplemented with 0.1% ␤-CyD and containing TG44 at each concentration. The culture medium suspensions were adjusted to pH 7, 6, and 5 with 1 N HCl and to pH 3 with 1 N HCl plus urea (1.4 mmol/liter). Urea was added to the medium because of H. pylori lethality under acidic pH conditions. With urea, a substrate of urease, H. pylori can produce ammonia to survive at pH 3. Cultures were incubated, and aliquots were collected at various time points. Each sample was serially diluted 10-fold with saline, and 10 l of the diluted sample was plated on BHIA supplemented with 7% horse blood (TG44-treated sample) or on BHIA supplemented with 7% horse blood and 5% penicillinase (AMX-treated sample). The plate was incubated at 37°C under microaerobic conditions for 3 days, and the number of colonies of H. pylori was counted.
Electron microscopy. Under microaerobic conditions (5% O 2 , 10% CO 2 , and 85% N 2 ) in an AnaeroPack Campylo jar (Mitsubishi Gas Chemical Co., Inc., Tokyo, Japan), the stock culture of H. pylori ATCC 43504 was grown at 37°C for 24 h in BHIB supplemented with 0.1% ␤-CyD by shaking the bacteria on a shaker at 125 rpm. The bacterial suspension (10 ml) in the exponential phase of growth was seeded into BHIB (990 ml) supplemented with 0.1% ␤-CyD and containing TG44 (at concentrations of 0.20, 0.39, 1.56, and 25 g/ml). After incubation at 37°C under microaerobic conditions for 3 and 6 h with shaking at 125 rpm, the sample was collected.
After prefixation with an aqueous solution of 1.5% glutaraldehyde in 0.05 M phosphate buffer (pH 7.2) for 30 min at 5°C, the sample was washed twice with 0.05 M phosphate buffer (pH 7.2) and then fixed with 1% OsO 4 in Veronal-acetate buffer (pH 6.1) for 16 h at room temperature by the method of Kellenberger et al. (13). After treatment with 0.5% uranyl acetate, the sample was subsequently dehydrated in alcohol solutions at serial concentrations.
For scanning electron microscopic observation, alcohol from serial concentrations was replaced with isoamyl acetate for further dehydration, and samples were then dried by the critical-point drying method and further evaporated with carbon and gold. The surface structures of bacterial cells were then observed with a scanning electron microscope (model JSM-35; Japan Electron Optics Laboratory, Tokyo, Japan).

MICs of TG44, AMX, CLR, and MNZ for drug-susceptible and drug-resistant
Antibacterial activity of TG44 against Helicobacter and related bacteria. TG44 showed no antibacterial activity against six strains among five species, despite showing weak antibacterial activity against Helicobacter mustelae (Table 3). It appears that TG44 possesses a very narrow antibacterial spectrum and high specificity against H. pylori species.  a MICs were determined by the agar dilution method on brain heart infusion agar supplemented with 0.1% ␤-cyclodextrin, which was seeded with a bacterial suspension of 10 7 CFU/ml. Antibacterial activities of TG44, AMX, CLR, and MNZ against common aerobic bacteria. The MICs of TG44, AMX, CLR, and MNZ for 13 reference strains of aerobic bacteria were determined by the agar dilution method. The MICs thereof are shown in Table 4. At a concentration of 25 g/ml, TG44 showed no antibacterial activity against gram-positive aerobic bacteria (i.e., S. aureus 209P JC, S. aureus Smith, S. epidermidis ATCC 12228, E. faecalis ATCC 29212, and B. subtilis PCI 219) or gram-negative aerobic bacteria (i.e., E. coli K-12, K. pneumoniae NCTC 9632, S. marcescens IFO 3736, P. vulgaris OX-19, P. mirabilis 1287, M. morganii KONO, P. rettgeri NIH 96, and P. aeruginosa PAO-1). The MICs of AMX and CLR for these aerobic bacteria ranged from 0.05 to Ͼ100 g/ml and from 0.10 to Ͼ100 g/ml, respectively.
Bactericidal activity of TG44 against H. pylori. The bactericidal activity of TG44 against reference strains of H. pylori was compared with that of AMX by using a short-term assay. Time-kill studies were conducted using the microbes at a concentration of approximately 10 5 CFU/ml for the initial seeding. Figure 2a shows the effects of TG44 on the viability of H. pylori ATCC 43629 at 1/2ϫ, 1ϫ, 2ϫ, 4ϫ, and 8ϫ the MIC for up to 24 h. TG44 showed potent bactericidal activity at a concentration of 3.13 g/ml for 1 to 24 h of treatment. Figure 2b shows the effects of AMX on the viability of H. pylori ATCC 43629 at concentrations of 1/2ϫ, 1ϫ, 2ϫ, 4ϫ, and 8ϫ the MIC for up to 24 h of treatment. AMX had almost no bactericidal activity at 8ϫ the MIC for up to 6 h. At 24 h of treatment, when the number of cells in the control group increased, AMX showed a slight inhibition of the viability of H. pylori. Figure 3a shows the effects of TG44 on the viability of H. pylori ATCC 43504 at concentrations of 3.13 to 50 g/ml for up to 1 h of treatment. TG44 showed potent bactericidal activity at 1 h of treatment. No visible microorganisms were detected after treatment with TG44 at 12.5 g/ml or higher concentrations for 1 h of treatment. Figure 3b shows the effects of AMX on the viability of H. pylori ATCC 43504 at concentrations of 3.13 to 50 g/ml for up to 1 h of treatment. AMX had no bactericidal activity at a concentration of 50 g/ml for 1 h of treatment. Figure 4 shows the bactericidal activity of TG44 against H. pylori ATCC 43504 under various pH conditions. TG44 at 3.13 g/ml showed bactericidal activity at pH 7 and 6 within 3 h of treatment. At pH 5, TG44 at 1.56 and 3.13 g/ml also reduced the viable numbers of H. pylori cells at 24 and 6 h of treatment, respectively. At an acidic pH of 3, H. pylori survived for 2 h in the presence of urea. TG44 exerted a predominant bactericidal effect at a concentration of 0.78 g/ml or higher for 2 h of treatment.

Electron microscopic observation of H. pylori.
Scanning electron micrographs of the cell surface of H. pylori ATCC 43504 are shown in Fig. 5. The morphology of H. pylori cells treated with TG44 at 0.20 g/ml was indistinguishable from that of nontreated cells (control). However, bleb-like structures were observed at 3 h of treatment with TG44 at concentrations of 0.39 and 1.56 g/ml. In addition, spherical cells with bleb-like structures were observed at 3 h of treatment with TG44 at 25 g/ml. Figure 6 shows the cell structures of H. pylori ATCC 43504 visualized by transmission electron microscopy, revealing the detachment of outer membranes, bleb-like structures which were observed by scanning electron microscopy (Fig. 5).
The coccoid form of H. pylori was not observed at any time point of treatment with TG44 or at any concentration examined.

DISCUSSION
H. pylori is well recognized as a major etiologic factor for gastritis and peptic ulceration (14,21) and has also been implicated as a risk factor for gastric lymphoma and carcinoma (7,24,28). Considerable progress in the therapeutic modalities for H. pylori infection has been achieved in recent years. However, the emergence of resistant bacteria has elicited a major clinical concern. The relevant resistance seems to be attributable to the types of antibiotics prescribed in eradication therapy, especially MNZ and CLR (16,17,23). Furthermore, the emergence of AMX-resistant strains has also been reported (4,5,6,11,19,25). Hence, there are great medical needs for the eradication of H. pylori by monotherapy if possible.
In the present study, in which the antibacterial activity of TG44 against 56 strains of H. pylori (including 2 reference strains and 54 clinical isolates) was examined in comparison with those of AMX, CLR, and MNZ, TG44 exhibited equivalent antibacterial activities against both susceptible bacterial strains and highly resistant clinical isolates.
TG44 was found to have a high specificity because it clearly exhibited antibacterial activity against H. pylori, a slight activity against H. mustelae, and no activity against other bacterial species examined.
Transmission electron microscopy showed the detachment of outer membranes of H. pylori, which might be the mechanism responsible for the rapid bactericidal activity of TG44, which is as short as 1 h of treatment. The mechanism of detachment is under investigation.
H. pylori is known to be transformed to the coccoid form after treatment with AMX or CLR. The coccoid form is hyposensitive to these antibiotics. In the present study, however, we observed no transformation of H. pylori to the coccoid form after treatment with TG44, implying its possible clinical relevance.
In conclusion, the present study revealed that (i) TG44 has equivalent antibacterial activities against both antibiotic-susceptible and -resistant strains of H. pylori and that (ii) TG44 exhibits bactericidal activity against H. pylori at a pH range of 3 to 7, confirming its high stability in the pH range as demonstrated in physicochemical studies (data not shown), in a short time of treatment. These facts suggest that TG44 is a promising chemotherapeutic agent which allows monotherapy against H. pylori infection, unlike conventional therapy which requires drug combinations and systemic circulation.
FIG. 6. Thin-section electron micrograph of H. pylori ATCC 43504 exposed to a 6-h treatment with TG44 at 25 g/ml. Arrowheads, detachment of outer membranes. The suspension of H. pylori was seeded into BHIB supplemented with 0.1% ␤-CyD containing TG44 at 25 g/ml. A sample was collected after 6 h of treatment at 37°C under microaerobic conditions. The bacterial cells were observed by transmission electron microscopy. Magnification, ϫ60,000.