Introduction

Nanotechnology is currently a well-recognized term that describes novel and significantly improved physical, chemical, and biological properties of materials [1, 2]. Specifically, it describes the science field in which nanometer-scale materials are studied. These nanomaterials have drastically different characteristics from the bulk materials from which they are derived, which strongly depend on the size and shape of the substance [3]. Metal nanoparticles (NPs) are of particular interest due to their unique physiochemical and biological properties and are widely used in various applications [4–7].

Different metals such as silver, gold, and selenium have been chemically or biologically produced as nanoscale materials in recent years, and various reports on the interaction of these NPs with prokaryotic and eukaryotic systems are published in the literature [6–10]. Currently, organic salts of bismuth (i.e., bismuth subsalicylate, bismuth subcitrate, and bismuth subnitrate) are used as antimicrobial agents in medicine [11–15]. Bismuth preparations are particularly effective in the treatment of different gastrointestinal disorders including gastritis, duodenal ulcers, and traveler’s diarrhea [14].

In the early 1980s, Helicobacter pylori, a small fastidious gram-negative bacilli, was isolated and an association between its infection and peptic ulcer disease was established.

Approximately 50 % of the world populations are known to be infected with H. pylori, mainly in the developing countries [15]. Eradication of this microorganism can result in curing of both infection and ulcer disease.

The management for the control of H. pylori infection involves the use of bismuth salts such as colloidal bismuth subcitrate, and proton pump inhibitors such as omeprazole, together with antibiotics like amoxicillin and metronidazole [16, 17]. Since the toxicity reported for elemental compounds at the nanoscale is lower than the toxicity of metal ion compounds, elemental NPs such as bismuth nanoparticles (Bi NPs) may be good candidates for replacement of other organic or inorganic salts of bismuth in future. The treatment failure rate for H. pylori infection is as high as 20 % and complete cure is not always achieved [16–18]. This failure is attributed to the selection for resistant strains of H. pylori due to repeated administration of anti-H. pylori medications, and the deep penetration of this bacterium into the gut epithelium, resulting in inaccessibility of antibiotics. These factors indicate a clear need for discovery and development of new anti-H. pylori agents [16–18].

In the present study, biogenic Bi NPs were intracellularly synthesized using a previously isolated bacterium [19]. The NPs were extracted from the bacteria, and their surface chemistry was studied by Fourier transform infrared spectroscopy (FTIR). Their antimicrobial effects were investigated by a serial agar dilution method against different clinical isolates of H. pylori as well as a standard test strain provided from American Type Culture Collection. Changes in intracellular metabolites in H. pylori following exposure to Bi NPs were also studied using nuclear magnetic resonance (NMR) metabolomic footprinting and the obtained NMR spectra were compared with control H. pylori cultures that were not treated with Bi NPs.

Materials and Methods

Biosynthesis of Bi NPs

Biosynthesis and purification of Bi NPs were conducted as described recently [19]. Briefly, nutrient broth medium containing 0.2 % bismuth subnitrate was prepared and inoculated with a pure-culture of Serratia marcescens isolated from the Caspian Sea (northern part of Iran). The medium was incubated on a rotary shaker at 150 rpm and 30 °C for 48 h. The bacterial cells were then removed by centrifugation and washed in 0.9 % NaCl to clean off any undesirable materials. The cells were disrupted by liquid nitrogen to release intracellular Bi NPs and an extracted using a two-phase system of 1:2 octanol and water. After 24 h, the released Bi NPs had settled in the bottom phase of the tubes. These settled materials were collected and re-extracted with chloroform. The upper phase was carefully transferred to a new test tube, centrifuged (10,000×g, 30 min), and sequentially washed with ethyl alcohol and distilled water. The sizes of the extracted Bi NPs before and after a long-term ultrasonic treatment (1 h) were determined by transmission electron microscopy (TEM). The composition of the prepared NPs was also studied by energy dispersive X-ray spectrometry (EDS).

Surface Chemistry of Biogenic Bismuth Nanoparticles

Functional groups present on the surface of Bi NPs prepared from S. marcescens were identified by drying the purified NPs and mixing them with potassium bromide. In the next step, a KBr tablet was made and subjected to FTIR. (Bruker Equinox55, Germany).

Antimicrobial Experiments

The H. pylori test strains used in this study were a standard strain (ATCC 26695) and four clinical isolates obtained from biopsies of the antrums of patients admitted to the endoscopy ward of Laleh Hospital in Tehran, Iran. The biopsies were transferred to the laboratory in thioglycolate broth medium at 4 °C within 5 h after removal from the patient. The samples were crushed and inoculated onto Columbia agar (Merck, Germany) plates supplemented with 10 % fetal bovine serum (FBS) (PAA, Austria) and 8 % defibrinated sheep blood, as well as an antibiotic supplement (vancomycin, polymyxin B, trimethoprim, cephalothin, amphotericin B) and campylobacter growth supplement purchased from Biolife, Italy. The plates were incubated at 37 °C under microaerophilic conditions with high humidity for 5 days [20, 21]. The strains were identified by colony morphology, gram staining, and oxidase, catalase, and urease tests.

The antimicrobial effects of biogenic Bi NPs against the selected test strains were determined by a modified serial solid agar dilution method adapted for H. pylori [21]. For this purpose, molten Columbia agar was subsequently supplemented with 10 % FBS, 8 % defibrinated sheep blood, and Bi NPs at concentrations of 40, 50, 60, 70, 80, 100, 120, and 140 μg/ml. Fresh cultures of each H. pylori test strain were harvested and suspended in normal saline to a final concentration of 108 CFU/ml (0.5 McFarland standard). The plates were inoculated with 5 μl of bacterial suspension, which provided an inoculum size of 105–106 CFU/spot [21]. Columbia agar media without any Bi NPs and Columbia agar supplemented Bi NPs without any inoculation were also used as positive and negative controls, respectively. All plates were incubated for 120 h in the microaerophilic condition described above. The MIC was determined as the lowest concentration of Bi NPs that inhibited bacterial growth. The antimicrobial tests were repeated twice.

Metabolomic Footprinting

1H NMR footprinting of metabolite changes in the culture medium of H. pylori treated with Bi NPs was investigated using two selected test strains, the standard strain (ATCC 26695) and a clinical isolate. Test tubes containing 2 ml of H. pylori bacterial suspension in phosphate-buffered saline were prepared at the inoculum density of 108 CFU/ml (0.5 McFarland standard). Bi NPs were added to each tube to reach the concentration equivalent to the measured MIC (100 μg/ml). The suspension samples were agitated using magnets in microaerophilic conditions with high humidity at 37 °C for 24 h. After incubation, 0.1 % sodium azide was added to prevent sample contamination with other bacteria. The treated suspension samples were then centrifuged at 10,000×g for 15 min and the supernatant was collected. The liquid supernatant was centrifuged again under the same conditions. The resulting upper phases were collected in glass vials and placed in an evaporator at temperatures not more than 40 °C to dry the samples. In the next step, aliquots (700 μl) of deuterium oxide containing 0.5 mM 4,4-dimethyl-4-silapentane-1-sulfonic acid were added to the dried materials and thoroughly shaken. The samples were centrifuged at 10,000×g for 15 min and the resulting clear liquids were subjected to NMR spectroscopy analysis.

The 1H NMR spectra of the samples were obtained using a presaturation technique at 500 MHz on a Bruker Avance spectrometer equipped with a 5-mm broadband probe and operating in the Fourier transform mode. All chemical shifts were referred to DSS as internal standard and reported in parts per million. Metabolic products were identified by Chenomx software by comparison of chemical shifts that are representative of the structure of a particular molecule.

Results and Discussion

Biosynthesis of Bi NPs and Their Surface Chemistry

In this investigation, S. marcescens was used to reduce Bi 3+ to elemental Bi NPs. These biogenic Bi NPs are accumulated inside the bacteria and require release and purification for characterization and biological experiments. A recently described aqueous–organic two-phase partitioning method [19] can be successfully used for purification of elemental Bi NPs deposited within S. marcescens. Figure 1 shows the different steps of this extraction. A transparent yellowish colloid resulted, which contained the Bi NPs. Other undesirable materials were removed during the two-step extraction of cell lysates with octanol and chloroform.

Fig. 1
figure 1

Partitioning of Bi NPs in a two-phase aqueous–organic system. From left to right: A extraction of cell lysate by octanol (upper phase), B settled Bi NPs after octanol extraction, C re-extraction of collected Bi NPs by chloroform (lower phase), and D collected yellowish color colloid which contains pure Bi NPs

Full size image

TEM micrograph (Fig. 2a) shows that the small NPs formed larger aggregated NPs around (100 nm). These aggregated irregular-shaped NPs were further subjected to a long-term ultrasonic agitation (1 h) and studied again by TEM. TEM images at higher magnification (Fig. 2b) clearly shows very small NPs (≤5 nm), and no aggregation in the particles were observed after long-term ultrasonic treatment (1 h). Figure 2c depicts the EDS spectrum of non-aggregated NPs revealing the presence of Bi element peaks. These EDS peaks confirm the existence of Bi NPs. It should be mentioned that additional peaks, belonging to copper and carbon element, are attributed to the grid used for TEM imaging.

Fig. 2
figure 2

TEM image of biogenic Bi NPs before (a) and after (b) long-term ultrasonic treatment. Lower illustration shows EDS of these Bi NPs which confirms the presence of Bi element. Additional peaks of copper and carbon elements are attributed to the grid used for TEM imaging

Full size image

The infrared spectrum of the purified Bi NPs is shown in Fig. 3. Two peaks appeared, with wave numbers of 1,056 and 1,642 cm−1, which are attributed to individual group vibrations of carbonyl and ─C─O─ groups. A broad peak at 2,500–3,300 cm−1 associated with a hydroxyl group (−OH) could also be detected in the FTIR spectrum of purified Bi NPs (Fig. 3). This indicated that the biogenic Bi NPs carried a carboxylic acid functional group on their surfaces.

Fig. 3
figure 3

Infrared spectrum of bismuth nanoparticles

Full size image

Antimicrobial Activity

Four clinical gram-negative H. pylori sp. with tiny transparent colonies were successfully isolated from biopsies of the antrums of patients and confirmed by positive oxidase, catalase, and strong urease activity. The antimicrobial activity of biogenic Bi NPs was evaluated in vitro against these four clinical isolates and a standard test strain of H. pylori (ATCC 26695). The Bi NPs showed good antimicrobial activity against all test strains of H. pylori (Table 1) The MIC of Bi NPs for the different H. pylori strains ranged from 60 to 100 μg/ml (Table 1). The most sensitive sample was one clinical isolate of H. pylori with a MIC value of 60 μg/ml, whereas growth of three other clinical isolates of H. pylori was inhibited at the higher concentrations of Bi NPs (70 or 80 μg/ml). The highest MIC was observed for H. pylori (ATCC 26695) against biogenic Bi NPs (100 μg/ml).

Table 1 MIC results for bismuth NPs against H. pylori strains
Full size table

In this investigation, the susceptibility of H. pylori (ATCC 26695) against bismuth subcitrate was also determined by a serial solid agar dilution method. No antibacterial effect was observed for bismuth in the form of bismuth subnitrate at concentrations lesser than 200 μg/ml (corresponding to 280 μg/ml of bismuth subnitrate). The antibacterial effect of bismuth in the form of elemental NPs against H. pylori was therefore higher than the anti-H. pylori effects of bismuth subnitrate, which is currently used as an antibacterial agent against this microorganism.

As mentioned in the introduction, a combination of bismuth organic salts and some antibiotics are currently used to manage H. pylori infections. The mechanism of action of bismuth compounds, despite their extensive therapeutic use, is not completely understood. It may cause cell wall degradation in H. pylori or may prevent bacterial adhesion on the gastric epithelium [22].

A recently published report in the literature indicates that Bi3+ ions inhibit the fumarase enzyme involved in the Krebs cycle. Fumarase is a key enzyme and modulates several oxidative reactions in the dicarboxylic acid arm of the Krebs cycle [23, 24]. The interaction of Bi3+ ions with this Krebs cycle enzyme, as well as its inhibitory effects on the other biological cycles involved in nucleotide pathways or urea metabolism, can lead to increases in free amino acids and other metabolites (i.e., uracil, formic acid, acetate) inside or outside of the bacterial cells [23, 24]. Therefore, analysis of small molecules or organic metabolites using footprinting techniques can provide important information about small changes in footprints of metabolic intermediates involved in different biochemical pathways.

The carboxyl-capped elemental Bi NPs biologically prepared during this investigation showed good antimicrobial activity against different test strains of H. pylori (Table 1). The effect of these Bi NPs on the footprints of various metabolites in culture media of H. pylori was studied by a 1H NMR metabolite footprinting technique. Figure 4 shows the different 1H NMR spectra recorded from the dried culture media inoculated with H. pylori (ATCC 26695) and incubated for 24 h in the absence (red lines) and presence of Bi NPs (blue lines). The 1H NMR spectrum of the biogenic Bi NPs was also recorded and used as a control (green line). The major differences in chemical shifts of the three samples are indicated in Fig. 4. The appearance of signals corresponding to uracil (58.75 ppm) (NMR spectrum B), valine at 361 ppm (NMR spectrum C), and glycine at 355 ppm (NMR spectrum C) are clearly observed in H. pylori culture medium supplemented with Bi NPs (100 μg/ml). These substances were not found in the other samples (untreated or control samples), and their presence may, therefore, be due to the effects of Bi NPs on metabolic pathways of H. pylori. Concentrations of other small molecules including formic acid (84 ppm) (NMR spectrum A), glutamate (24 ppm) (NMR spectrum D), and acetate (238 ppm) (NMR spectrum E) were also increased in the culture medium of H. pylori supplemented with Bi NPs. Illustration F in Fig. 4 shows a whole NMR spectrum recorded during this experiment. We obtained the same 1H NMR metabolite footprinting results for the clinical isolate of H. pylori treated with Bi NPs, and no differences were observed between the metabolite footprinting pattern of standard H. pylori (ATCC 26695) and the clinical isolate (data not shown).

Fig. 4
figure 4

1H NMR spectra recorded from dried culture media inoculated with H. pylori (ATCC 26695) and incubated for 24 h in the absence (red lines) and presence (blue lines) of Bi NPs. The green line indicates the 1H NMR spectrum of the extracted Bi NP colloid. Spectra (a–e) show corresponding signals related to formic acid (a), uracil (b), valine and glycine (c), glutamate (d), and acetate (e). Whole spectra obtained during current metabolic footprint analysis are also shown in image f

Full size image

Conclusion

The H. pylori bacterium is a causative agent in gastric ulcer disease and gastric adenocarcinoma. Appearance of resistant test strains and insufficient patient compliance toward current anti H. pylori drugs are the main drawbacks of chemotherapy for H. pylori. Therefore, new therapeutic agents for curing this infection are constantly needed. In this study, elemental Bi NPs (<100 nm) were intercellularly synthesized by S. marcescens, extracted, and used for further experiments. The surface chemistry of these biogenic Bi NPs was studied by FTIR spectroscopy, which revealed the existence of carboxylic groups on the surface of the biogenic NPs prepared by S. marcescens. The antimicrobial activity of these carboxyl-capped Bi NPs was evaluated by a serial agar dilution method against different test strains of H. pylori and MIC values were calculated. Purified Bi NPs showed good antimicrobial activity against all tested bacterial strains. The effects of Bi NPs on H. pylori were also investigated by a 1H NMR metabolite footprinting method. The concentrations of different small metabolites (some amino acids and organic acids) in the culture medium of H. pylori were increased in the presence of Bi NPs, indicating that these NPs interfere with vital biochemical pathways in H. pylori. Taken together, in vitro antimicrobial activity of Bi NPs against H. pylori was evident, so that these NPs may be useful for chemotherapy of H. pylori in the future.