In vitro antibacterial activity of Phx-3 against Helicobacter pylori

Article abstract of DOI:10.1248/bpb.33.188

Phx-3, one of the phenoxazine derivatives, is reported to have inhibitory effect on Mycobacterium species and Chlamydia pneumoniae but not Escherichia coli, Salmonella Typhimurium, Pseudomonas aeruginosa, Staphylococcus aureus, Listeria monocytogenes. The bactericidal activities of Phx-3 against Helicobacter pylori strains have not been assessed. Then, we measured minimum inhibitory concentration of Phx-3 for Helicobacter strains and assessed the morphological and biochemical effects of Phx-3 on H. pylori. In present study, it has shown that H. pylori strains including clarithromycin resistant strain and Helicobacter musterae were killed effectively by the treatment with Phx-3. Furthermore, severe morphological changes such as membrane blebbing and formation of hollows in H. pylori were detected. In addition, induction of heat shock protein 60 was observed. Taken together, Phx-3 has antibacterial activity against Helicobacter pylori.

Full text of DOI:10.1248/bpb.33.188

188
Biol. Pharm. Bull. 33(2) 188—191 (2010)
Vol. 33, No. 2
In Vitro Antibacterial Activity of Phx-3 against Helicobacter pylori
Tomoko HANAWA, ,a Takako OSAKI,a Taki MANZOKU,b Minoru FUKUDA,c Hayato KAWAKAMI,d
*
a
Akio TOMODA,e and Shigeru KAMIYA
a Department of Infectious Diseases, Division of Medical Microbiology, Kyorin University, School of Medicine;
c Laboratory for Electron Microscopy, Kyorin University, School of Medicine; d Second Department of Anatomy,
Department of Biochemistry, Kyorin University, School of Medicine; 6–20–2 Shinkawa, Mitaka, Tokyo 181–8611, Japan:
b Miyarisan Pharmaceutical Co., Ltd.; 1–10–3 Kamiyanazaro, Kita-ku, Tokyo 114–0016, Japan: e Tokyo Medical
University; 6–1–1 Shinjuku, Shinjuku-ku, Tokyo 160–8402, Japan.
Received June 1, 2009; accepted November 18, 2009; published online November 20, 2009
Phx-3, one of the phenoxazine derivatives, is reported to have inhibitory effect on Mycobacterium species
and Chlamydia pneumoniae but not Escherichia coli, Salmonella Typhimurium, Pseudomonas aeruginosa,
Staphylococcus aureus, Listeria monocytogenes. The bactericidal activities of Phx-3 against Helicobacter pylori
strains have not been assessed. Then, we measured minimum inhibitory concentration of Phx-3 for Helicobacter
strains and assessed the morphological and biochemical effects of Phx-3 on H. pylori. In present study, it has
shown that H. pylori strains including clarithromycin resistant strain and Helicobacter musterae were killed effec-
tively by the treatment with Phx-3. Furthermore, severe morphological changes such as membrane blebbing and
formation of hollows in H. pylori were detected. In addition, induction of heat shock protein 60 was observed.
Taken together, Phx-3 has antibacterial activity against Helicobacter pylori.
Key words Helicobacter pylori; phenoxazine; bactericidal agent
Phonoxyazine derivatives which are produced by microor- apy was approximately 20% of infected patients in Japan.10)
ganisms1—3) and found in insect4) possess various biological With the increase in the frequency of CAM-resistant H. py-
activities such as anti-tumor2) and anti-microbial activi- lori, there is rising concern about the potential decline in the
ties.1,5—8) Although biological activity of actinomycin was eradication rate of this infection.11) Therefore, it is important
elucidated, other phenoxyazine derivatives have not been ex- to develop new drugs to combat this bacterium. In the pres-
amined well. We have been studied the anti-bacterial effects ent study for further investigation of possibility of phenox-
of Phx-1 (2-amino-4,4alpha-dihydro-4alpha-7-dimethyl-3H- azine derivatives as bactericidal compound, we examined the
phenoxazine-3-one), Phx-2 (3-amino-1,4alpha-dihydro- effects of Phx-3 on H. pylori.
4alpha-8-dimethyl-2H-phenoxazine-2-one) and Phx-3 (2-
amino-phenoxazine-3-one), which are produced by the reac- MATERIALS AND METHODS
tion of o-aminophenol or its derivatives with bovine hemo-
globin.7) Additionally, the chemical structure of Phx-3 (Fig.
Bacteria and Cultivation We used three standard
1) is identical to questiomycin A, which was isolated from strains of H. pylori (American Type Culture Collection
the culture fluids of Streptomyces by Anzai et al.1) as a com- [ATCC] 43504, ATCC 43579, and The National Collection
pound with bactericidal activity against Mycobacterium of Type Cultures NCTC [NCTC] 11638 and three clinical
tuberculosis. Subsequent studies showed that Phx-3 also isolates (TK 1029, 1047, and 1402). Brucella broth (Difco
inhibits the growth of non-tuberculosis Mycobacterium Laboratories, Inc., Detroit, MI, U.S.A.) supplemented with
species.7) In contrast, some other Gram-negative and Gram- 7% horse serum was used for the liquid culture of H. pylori
positive bacteria were resistant to as much as 100 mg/ml of strains. For the plate culture, 1.5% of bacto agar (Difco Lab-
Phx-37) except for facultative intracellular bacteria Chlamy- oratories, Inc. Detroit, MI, U.S.A.) was supplemented to the
dia pneumoniae.8)
medium. Bacteria were incubated at 37 °C under microaero-
Helicobacter pylori, which infects approximately 50% of bic atmosphere produced by the AnaeroPack®-MicroAero
the population worldwide, is a causative agent for chronic (Mitsubishi Gas Chemical Co., Inc., Tokyo, Japan).
gastroduodenal ulcer, and possibly gastric cancer. Triple ther-
Chemicals Phx-1, Phx-2 and Phx-3 were synthesized
apy consisting of amoxicillin, clarithromycin (CAM) and a from 2-amino-5-methylphenol, 2-amino-4-methylphenol or
proton pump inhibitor has proven to be effective for the erad- o-aminophenol, respectively, with bovine hemoglobin ac-
ication of H. pylori so far.9) As in the case of other bacteria, cording to the method described in.12) Phx-3 was dissolved in
however, there has been an increase in the appearance of methanol as a stock solution at the concentration of 25 mM
drug-resistant strains of H. pylori. It has been reported that and was stored at ꢀ20 °C until use. The chemical structure
failure of eradication of H. pylori infection with triple ther- of Phx-3 is illustrated in Fig. 1.
Measurements of the Minimal Inhibitory Concentra-
tion (MIC) for H. pylori We evaluated the ability of Phx-3
to inhibit the growth of H. pylori by determining the MICs as
described by the guidelines of the Clinical and Laboratory
Standards Institute for Clinical Laboratory Standards.13)
Fig. 1. Chemical Structure of Phx-3 as Identified by Shimizu et al.7)
The structure, 2-amino-phenoxazine-3-one is identical to that of questiomycin A.2)
Briefly, the 72-h old subcultures diluted and adjusted to the
1ꢁ107 to 1ꢁ108 colony forming units per ml were replicated
© 2010 Pharmaceutical Society of Japan
∗ To whom correspondence should be addressed. e-mail: thanawa@ks.kyorin-u.ac.jp

February 2010
189
directly on the Phx-3-containing agar dilution plates. The
MICs were determined after 3 d-cultivation. The sensitivities
of these strains to CAM were measured by the E-test® ac-
cording to the manufacturer’s instructions (AB Biodisk,
Solna, Sweden).
Electro Microscopy The bacterial cells were collected
and prefixed in an aqueous solution of glutaraldehyde in
50 mM phosphate buffer (pH 7.4) at 4 °C. Following washing,
the bacterial cells were fixed with OsO4 in Veronal-acetate
buffer (pH 6.0) for 16 h at room temperature. After treatment
with 0.5% uranyl acetate, the samples were subsequently de-
hydrated in a graded series of alcohol. Micrographs with
scanning electron microscope were observed with model
JSM-5600LV (Japan Electron Optics Laboratory, Tokyo,
Japan).
Fig. 2. Bactericidal Effect of Phx-3 on H. pylori ATCC 43504
The cells were treated with Phx-3 at 0.5 mg/ml (solid circles), 1 mg/ml (solid trian-
gles), 2 mg/ml (solid diamonds), 4 mg/ml (solid squares) or 8 mg/ml (crosses). The cel-
lular viability was determined at 0, 3, 24, and 48 h by the viable cell counting on Bru-
cella agar supplemented with 7% horse serum. The number of bacteria in the culture in
the absence of Phx-3 was also measured (open circles). All experiments were per-
formed in duplicate.
Western Blotting H. pylori ATCC 43504 strain was cul-
tured for 8 h, and 12 mg/ml of Phx-3 was added. Following
the 6-h, bacteria cells were harvested by centrifugation at
8500ꢁg for 10 min and washed with phosphate buffered
saline (PBS). The bacterial cells were disrupted by ultra-
sonic waves (Sonifier, Bronson Ultrasonics Co., Danbury CT,
U.S.A.). Cell debris was removed by centrifugation at
8500ꢁg for 20 min. The cell lysates were developed with
12.5% acrylamide gel and transfer to the polyvinylidene di-
fluoride (PVDF) membrane. Following to the reaction with
primary and secondary antibodies, reacted proteins with pri-
mary antibodies were detected with DAB (3,3ꢂ-diaminoben-
zidine). The anti-Omp19 (Biodesign International, Saco,
ME, U.S.A.) and H2014) monoclonal antibodies were used for
the detection of Omp19 and HSP60, respectively.
RESULTS AND DISCUSSION
Fig. 3. Scanning Electron Microscopy of H. pylori Treated with (B and C)
or without (A) Treatment of 6 mg/ml Phx-3
Formation of spherical cells and membrane blebbing were observed (arrows) in Phx-
3-treated cells. Bars represent 1 mm.
We assessed the anti-microbial effect of Phx-3 on H. py-
lori ATCC 43504, ATCC 43579, NCTC 11638, TK 1029,
1047, and 1402. None of H. pylori strains were able to grow
in the presence of 1 mg/ml or more of Phx-3 (data not
shown). Among examined strains, H. pylori TK 1047 was graphs shown in Fig. 3 revealed that Phx-3 caused severe
considered CAM-resistant according to the CLSI guidelines morphological changes, including membrane blebbing, the
as the MIC of CAM against this strain was 1.5 mg/ml for formation of hollows and the formation of spherical cells. In-
CAM. Therefore, Phx-3 was able to prevent the growth of terestingly, no coccoid cells were observed until 8 h after
CAM resistant strain as well as sensitive strains.
treatment with Phx-3. Because the membrane-like structures
In addition, we found that the MIC of Phx-3 against Heli- were attached to the surface of the bacterial cells treated with
cobacter musterae ATCC 43772 was 0.5 mg/ml, which was Phx-3, the exfoliation of outer membrane was suspected.
similar to the MICs of Phx-3 for the H. pylori strains. There-
We next examined the localization of outer membrane
fore, it appears that Phx-3 inhibits the growth of Helicobac- protein in the H. pylori cells. The lysates were separated
ter species including H. pylori. on polyacrylamide gel electrophoresis and stained with
To further examine the antibacterial effect of Phx-3 on H. Coomassie brilliant blue (Fig. 4C). The Omp19 protein was
pylori, we counted the number of colonies of bacteria in the detected using monoclonal antibody against Omp19. We de-
culture at selected time points after the addition of various tected no differences in the amounts of Opm19 proteins be-
concentrations of Phx-3. As shown in Fig. 2, Phx-3 killed H. tween Phx-3 treated and un-treated H. pylori (Fig. 4A). The
pylori at the concentrations equal to or higher than the MIC. result showed that the outer membrane was not exfolitated by
After 48 h, Phx-3 reduced the numbers of bacteria to below the addition of Phx-3 contrary to our expectation. We also
the limit of detection (Fig. 2). Unlike other bacteria to which compared the amount of HSP60, general stress protein of H.
Phx-3 has weak antibacterial activity,7) our results suggest pylori after treatment with Phx-3. It was shown that Phx-3
that Helicobacter species are highly susceptible to Phx-3.
treated cells produced more heat shock protein 60 than un-
Anti-H. pylori agents have been shown to cause a variety treated cells (Fig. 4B). The damages of the bacterial cells
of morphological changes.15—17) We therefore examined the caused by Phx-3 may be generated in not only the surface of
morphological changes of H. pylori ATCC 43504 cells after the cells but also whole cells. As shown in Fig. 4B (lane 2),
a 6-h culture in the presence or absence of 6 mg/ml of Phx-3 H20 monoclonal antibody used in this study to detect HSP60
using scanning electron microscopy. The electron micro- of H. pylori reacted with unidentified proteins of H. pylori.

190
Vol. 33, No. 2
sion to three-dimensional change. It is possible that the pla-
nar structure of Phx-3 is important to its bactericidal activity.
Bendic and Volanschi evaluated the intermolecular interac-
tions between the drug-nucleic acid complexes by modeling
based on molecular mechanics optimization.19) Their molec-
ular modeling of questiomycin A (Phx-3) suggested that it
can bind in the minor groove after intercalation in double-
stranded DNA. The bactericidal activity of Phx-3 might be
therefore due to interaction with the genomic DNA of H. py-
lori.
It is not clear why Phx-3 shows the specificity of the bacte-
ricidal activity for Helicobacter. Specie specific intrinsic
drug resistance is thought to be determined by several factors
such as size of porin(s), efflux pump, modification of drug,
affinity of active point and surface charge. Exner et al.20)
identified four pore forming outer membrane proteins related
to porin, and the channel size of porin in H. pylori was rela-
tively small. This could contribute to the resistance of H. py-
lori to hydrophilic antibiotics. Among them, the efflux of the
toxic drugs is an important factor controlling the intrinsic
sensitivity. H. pylori is highly sensitive to many hydrophilic
and hydrophobic agents despite relatively low susceptibilities
to the polycation polymyxin B and cationic antimicrobial
peptides.21) Bina et al.22) proposed that reasonable uptake of
hydrophobic substances coupled with nonfunctioning efflux
systems could explain the relatively high susceptibility of H.
pylori to hydrophobic antibiotics. Phx-3 might therefore be-
Fig. 4. Whole Cell Lysates of H. pylori ATCC 43504 with or without
Treatment of Phx-3 Were Examined by Polyacrylamide Gel Electrophoresis,
followed by Western Blotting with Anti-Omp19 (A) and Anti-HSP-60 Anti-
bodies (B) or Staining with Coomassie Brilliant Blue (C)
Lane 1, H. pylori ATCC43504; lane 2 H. pylori ATCC43504 treated with Phx-3; lane
M: protein molecular weight markers. Numbers at the left of lane M indicate the molec-
ular weights of the markers.
Yamaguchi and coworkers14) purified H. pylori HSP60 using have like other hydrophobic antibiotics with respect to up-
affinity chromatography with H20 monoclonal antibody. take into and efflux from bacterial cells. After accumulation
After loading of the purified HSP60 fraction in Western blot of Phx-3, the molecules may intercalate into the double-
analysis, several proteins except for the HSP60 were still de- stranded DNA and inhibit transcription.
tected with H20 monoclonal antibody as minor bands. These
results suggest that they may be degraded HSP60 protein or tivity of Phx-3 against H. pylori in vitro. This specific sensi-
cross-reacting antigen induced by various stresses. tivity of H. pylori seems to be due to the genus specific struc-
In summary, we demonstrated the specific bactericidal ac-
Motility of H. pylori during colonization in the gastric mu- ture of cell-surface. And, possible mechanism of bactericidal
cosa is an important factor for the pathogenesis following H. activity of Phx-3 is interaction with bacterial DNA and inhi-
pylori infection.18) We therefore assessed the possibility that bition of its transcription. On the other hand, effects of Phx-3
Phx-3 affects the motility of H. pylori by measuring s arming on the eukaryote cells were previously assessed by Yama-
motility on the soft agar medium containing 0.4% agar. We guchi and colleagues.8) They described that concentrations of
found that pretreatment or simultaneous treatment with Phx- less than 10 mM of Phx-3 did not show cytotoxicity to HEp-2
3 at concentrations below the MIC did not alter the swarming and THP-1 cells.
of H. pylori (data not shown).
These results obtained provide important information for
The three-dimensional structure of chemical substances development of novel anti-H. pylori drugs.
sometimes provides clues to their mechanism of action. Phx-
3 is a planar molecule with 2-amino and 3-ketone residues.
These features provide structural rigidity, which could allow
Phx-3 to intercalate into the double-stranded DNA. Phx-3
could form hydrogen bonds with deoxyguanosine between
the GpC pairs. We therefore extracted the genomic DNA of
H. pylori and examined the DNA patterns by agarose gel
electrophoresis. By staining with ethidium bromide, we did
not detect any changes until 8 h after the treatment with Phx-
3 (data not shown).
Furthermore, we examined the bactericidal effect of Phx-1
and Phx-2 on H. pylori strains, and the MICs of these com-
pounds against tested nine H. pylori strains and H. mustelae
strain were between 50 mg/ml and 100 mg/ml, and 20 mg/ml
and 100 mg/ml, respectively. These results indicate no bacte-
ricidal activities of Phx-1 and Phx-2 against Helicobacter.
Difference of structures between Phx1 or Phx-2, and Phx-3
REFERENCES
1) Anzai K., Isono K., Okuma K., Suzuki S., J. Antibiot., 13, 125—132
(1960).
2) Brockmann H., Muxfeldt H., Chem. Ber., 91, 1242—1265 (1958).
3) Hollstein U., Chem. Rev., 74, 625—652 (1974).
4) Butenandt A., Biekert E., Kubler H., Linzen B., Physiol. Chem., 319,
238—256 (1960).
5) Iwata A., Yamaguchi T., Sato K., Izumi R., Tomoda A., Tohoku J. Exp.
Med., 200, 161—165 (2003).
6) Holmes S. C., Gait M. J., Nucleosides Nucleotides Nucleic Acids, 22,
1259—1262 (2003).
7) Shimizu S., Suzuki M., Tomoda A., Arai S., Taguchi H., Hanawa T.,
Kamiya S., Tohoku J. Exp. Med., 203, 47—52 (2004).
8) Uruma T., Yamaguchi H., Fukuda M., Kawakami H., Goto H., Kishi-
moto T., Yamamoto Y., Tomoda A., Kamiya S., J. Med. Microbiol., 54,
1143—1149 (2005).
9) Talley J. N., Lam S. K., Wong B. C., Xia H. H., Expert Opin. Pharma-
cother., 3, 1301—1311 (2002).
was the presence of methyl residues which give to the exten- 10) Asaka M., Sugiyama T., Kato M., Satoh K., Kuwayama H., Fukuda Y.,

February 2010
191
Fujioka T., Takemoto T., Kimura K., Shimoyama T., Shimizu K.,
Kobayashi S., Helicobacter, 6, 254—261 (2001).
T., Kamiya S., Antimicrob. Agents Chemother., 50, 3062—3069
(2006).
11) Adamek R. J., Suerbaum S., Pfaffenbach B., Opferkuch W., Am. J. 17) Nakao M., Malfertheiner P., Helicobacter, 3, 21—27 (1998).
Gastroenterol., 93, 386—389 (1998). 18) Suerbaum S., Trends Microbiol., 3, 168—171 (1995).
12) Iwata A., Yamaguchi T., Sato K., Yoshitake N., Tomoda A., Biol. 19) Bendic C., Volanschi E., Internet Electronic J. Mol. Design, 5, 320—
Pharm. Bull., 28, 905—907 (2005).
330 (2006).
13) Clinical and Laboratory Standards Institute, “Methods for Dilution
Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically,
Approved Standard M7-A5,” 6th ed., CLSI, Wayne, PA, 2000.
14) Yamaguchi H., Osaki T., Taguchi H., Hanawa T., Yamamoto T.,
Kamiya S., J. Med. Microbiol., 46, 819—824 (1997).
20) Exner M. M., Doig P., Trust T. J., Hancock R. E., Infect. Immun., 63,
1567—1572 (1995).
21) Wang G. E., Taylor D. E., “Helicobacter pylori Infection and Immu-
nity,” ed. by Yamamoto Y., Friedman H., Hoffman P., Kluwer Acade-
mic/Plenum Publishers, New York, 2002.
15) Akada J., Shirai M., Fujii K., Okita K., Nakazawa T., Antimicrob. 22) Bina J. E., Alm R. A., Uria-Nickelsen M., Thomas S. R., Trust T. J.,
Agents Chemother., 43, 1072—1076 (1999).
Hancock R. E. W., Antimicrob. Agents Chemother., 44, 248—254
16) Kamoda O., Anzai, K., Mizoguchi J., Shiojiri M., Yanagi T., Nishino
(2000).