Introduction

From 1928, antibiotic as one of the greatest discoveries of medicine has been led to save the lives of millions of people. Unfortunately, wrong prescription and misuse of antibiotic makes it almost affectless in many cases [1]. Therefore, several bacterial strains have become resistant to antibiotics because of their plasmid and transposons induced chromosomal changes and genetic mutations [2, 3]. Every year, the infections caused by these drug-resistant bacteria have been led to a considerable mortality worldwide [4]. Hence, a novel and well-developed method for solving this problem is crucial. In earlier decades, photodynamic inactivation (PDI) has been introduced as one of the promising methods in order to annihilate bacteria. In photodynamic inactivation, absorption of light by photosensitizing agents in the presence of molecular oxygen leads to production of highly toxic reactive oxygen species (ROS). The invariable reaction of the produced ROS with neighbor microorganism results in bacterial annihilation.

Recently, application of gold nanoparticles has been attracted by several researchers in order to modify the efficiency of antimicrobial photothermal and photodynamic inactivation [5,6,7,8,9,10,11]. This attention was originated from the special properties of gold nanoparticles such as their minimal toxicity and high potential to load several functions [12, 13]. Based on the results of previous researches about the application of gold nanoparticles for PDI, an elevated inactivation of microorganism during light illumination was observed [14,15,16]. For example, Simon-Deckers et al. (2008) obtained positive results in the destruction of Escherichia coli combining gold nanoparticles and x-ray radiations [10]. Also application of surface plasmon resonance (SPR) of gold nanoparticles in the range of 500–600 nm is recently under investigation in photothermal inactivation of bacteria. Zharovet al. (2006) studied photothermal treatment of infections caused by a variety of bacteria using high-power pulsed Nd:Yag lasers [5]. Despite their good results, some disadvantages were reported due to dangers of the high-power laser beams. In fact, high-power laser beams potentially causes more local cell damages in nearby healthy tissues.

According to our study, there are not many reports about the photodynamic inactivation (PDI) of bacteria using naked gold nanoparticles and low-power density lasers. Although there are some reports about metallic photosensitizers such as platinum and iridium as well as gadolinium, but the application of naked gold nanoparticles as a potent photosensitizer is not well studied yet [17,18,19]. Therefore, this paper attempts to fill this gap through utilizing naked gold nanoparticles and the irradiations of a low-power density laser. For this purpose, E. coli was used as one of the routine model microorganisms in order to demonstrate the possibility of PDI using collaboration of these factors. The results of this study could be considered as a prelude for future researches in order to destroy other drug-resistant pathogenic bacteria interfering this approach.

Experimental

In order to investigate the effects of light with or without the presence of gold nanoparticles on the viability of E. coli, several experiments were performed.

Gold nanoparticles and light source

Gold nanoparticles were purchased from the PNF (Payamavaran Nanofanavari Fardanegar) company. These gold nanoparticles are produced by the Electrical Explosion of Gold Wire (EEW) technique, a classified top-down approach in the field of nanomaterial production, in aqueous media. In this method, a thin gold wire was exploded by a high-voltage electric source to produce nano-sized gold particles dispersed in the liquid media [20]. Different concentrations of gold nanoparticles were achieved by diluting the initial concentration (1000 ppm) using deionized water. In order to study the effect of ionic media on size distribution of employed gold nanoparticles, the purchased gold nanoparticle with the concentration of 100 μg/ml was centrifuged at 14,000 rpm for 30 min and then washed with deionized water. After that, the remained nanoparticles were mixed with standard saline serum. Hence, the total concentration of gold nanoparticles reached to 100 μg/ml and the ionic solution was 0.154 M.

Instruments

TEM studies were performed with a Philips CM30 electron microscope. Dynamic light scattering measurements were performed by Cordouan Technology, France, using a diode laser at 658 nm with 75 mW output power. Absorption spectra for the samples were obtained using an UV-visible spectrophotometer (Spekol 2000, Analyticjena CO). Also, a CW second harmonic Nd:Yag laser (CNI Laser, model MGL-III-532) at 532 nm was used as the light source. The power and diameter of output beam were respectively 20 mW and 1.2 mm. Using a beam expander, the diameter of the beam was increased to 8 mm (in accordance with wells of a 96-well plate). Therefore, a uniform power density of ~ 40 mW/cm2 was finally obtained. For temperature control experiments, an infrared thermometer with the accuracy of 1.5 °C and the resolution of 0.1 °C was used to measure the solution temperature during light illumination.

Photo-induced singlet oxygen generation

In this section, the ability of naked gold nanoparticles to generate singlet oxygen through absorption of the radiations of the CW laser (40 mW/cm2, 532 nm) was studied based on the bleaching of 1, 3-diphenylisobenzofuran (DPBF) [21]. For this, DPBF as a strong singlet oxygen scavenger was purchased from Sigma-Aldrich. A 2 × 10−4 M solution of DPBF was prepared by diluting 5.4 mg of DPBF in 100 ml of 50/50 (v/v) mixture of water and ethanol and the solution was stirred in the dark. After that, 10 ml of this solution was mixed by the same volume of 50/50 (v/v) mixture of water and ethanol. Then, 4 ml of the prepared solution was injected into a quartz cuvette. This sample was exposed by the laser light source for 30 min, and for every 5 min, the absorbance of DPBF at 415 nm was obtained using scanning the absorption spectrum by the UV-visible spectrophotometer in the range of 350–700 nm. This experiment was repeated for a filled cuvette containing a 50/50 (v/v) mixture of 100 μg/ml gold nanoparticle and 2 × 10−4 M solution of DPBF.

Culture of bacteria

E. coli ATCC 25922 was purchased from the Iranian Research Organization for Science and Technology (IROST). The bacteria were cultured in 20 ml nutrient broth liquid medium and were kept for 18 h in a shaker incubator (FF-81, ParsAzma Co.) at 37 °C and rotating speed at 200 rpm. Then, the cultured bacteria were centrifuged at 3500 rpm for 15 min and the supernatant was removed by a pipette. The remaining bacteria were suspended in physiological saline solution and centrifuged at 3500 rpm for 10 min. The supernatant was removed by pipette and the remaining particles were mixed in saline solution. The obtained suspension was diluted with saline solution to achieve the McFarland dilution and considered as the reference suspension. To investigate the effects of light and/or gold nanoparticle, we introduce four groups.

The main control group

E. coli ATCC 25922 bacteria were mixed with an equal volume of saline solution and kept in the dark incubator at 37 °C for an hour. After that, the resulting solution was serially diluted and was cultured in a BHI medium and then kept at 37 °C for 24 h. At the end, the viability of the control group was obtained through counting the colonies.

The drug control group

The solutions of gold nanoparticles in concentrations of 1, 2, 10, 20, and 100 μg/ml were prepared and added to an equal volume of the reference bacteria. We hold these obtained solutions in the dark shaker incubator at 37 °C in 200 rpm for an hour. Then, the number of colonies was counted based on serial dilution protocol.

The light control group

The reference bacteria were mixed with an equal volume of saline solution in order to study effects of laser light at 532 nm on the bacterial survival. Then, we inject it in wells of a sterile 96-well plate. After that, samples were exposed with the Nd:Yag laser beam with an intensity of 40 mW/cm2 at various time intervals (15, 30, and 60 min). The number of bacterial colonies was counted after serial dilution of the result solutions through the protocol.

The photodynamic inactivation group

In order to study the synergistic effects of nontoxic gold nanoparticle and irradiations of the laser beams, reference bacteria were mixed with an equal volume of nontoxic concentration of gold nanoparticles. Then, the obtained solution was held in the dark shaker incubator for an hour and was injected in the wells of a sterile 96-well plate. Afterwards, the wells were illuminated with Nd:Yag laser beam at the wavelength of 532 nm and intensity of 40 mW/cm2. In addition, the illumination was performed at various exposure times (15, 30, and 60 min). The temperature of the media was immediately recorded before and after light illumination using an infrared thermometer with a resolution of 0.1 °C. In the final step, serial dilution of illuminated samples was performed and the number of colonies was counted after 24 h.

Statistical analyses

In this study, all experiments were repeated three times and results were statistically analyzed. One-way analysis of variance (ANOVA) and Tukey post hoc test were used for comparison of our PDI results. Analysis was done using SigmaPlot 12.0Ink. Results with P ˂ 0.05 were considered statistically significant.

Results and discussion

Characterization of gold nanoparticles

Figure 1 shows TEM images and size distributions of the purchased gold nanoparticles and gold nanoparticles diluted in saline serum. According to this figure, spherical gold nanoparticles diluted in dionized water had an average size ∼ 10 nm and a maximum size ∼ 50 nm. However, dilution in saline media led to aggregation of gold nanoparticles and a significant increase in the size distribution happened. The DLS result showed the size distribution broadened from 100 to 1000 nm. According to the experimental results, the obtained broad size distribution showed a peak at 220 nm. This result is in accordance with previous reports about the effect of ionic solvents on metallic nanoparticles [22,23,24,25]. In fact, the biological environments are considered as ionic media and therefore the resulting aggregation may probably be considered as an important factor in comparison with the non-ionic solvent.

Fig. 1
figure 1

TEM images of the naked gold nanoparticles dissolved in a deionized water and b 0.154 M saline serum and (c) their size distribution

Full size image

UV-visible absorption spectrum of the gold nanoparticles

Figure 2 shows UV-visible absorption spectrum of the gold nanoparticles in the presence or absence of ionic solvent in the range of 400–800 nm. As seen in this figure, purchased nanoparticles dissolved in deionized water revealed an absorption peak at 530 nm while the peak related to gold nanoparticles solved in saline serum shifted to 585 nm. These absorption bands originate from the surface plasmon resonance (SPR) of electrons on the surface of gold nanoparticles [26]. The red shift that appeared in the absorption spectrum is justified based on the effect of size on the radial plasmon resonance of electrons. Therefore, based on our experimental results, ionic solutions cause the aggregation of gold nanoparticles and as another result a considerable red shift in their absorption spectrum. This result is in good agreement with previous reports [22,23,24,25]. As shown in Fig. 2, the emission wavelength of the second harmonic of the Nd:Yag laser at 532 nm was also represented. According to this figure, the employed laser had good potential to excite both gold nanoparticles. Therefore, it is predictable that this energy transfer from laser to the gold nanoparticles will cause photochemical interactions in the targeted biological structures.

Fig. 2
figure 2

UV-visible absorption spectrum of the employed gold nanoparticles dissolved in deionized water (blue line) and gold nanoparticles dissolved in saline solution (black line) and the emission wavelength of the second harmonic Nd:Yag laser (green dashed line)

Full size image

The results of singlet oxygen detection

Figure 3a shows temporal variation of absorption spectrum of DPBF in the presence of gold nanoparticles. As seen in this figure, the peak at 415 nm got gradually decreased from 1.52 to 1.42 during 30 min illumination by 40 mW/cm2 laser light at 532 nm. Figure 3b represents a comparison between the natural logarithm of DPBF absorbance at 415 nm peak for the solutions with and without gold nanoparticles. Supposing a first-order reaction between the generated singlet oxygen and DPBF, the presence of gold nanoparticle causes a growth in the rate of DPBF bleaching. From our results, the rates of DPBF bleaching were obtained, − 0.0046 and − 0.0075, for the samples without and with gold nanoparticles, respectively. In our opinion, this result is a proof to establish the ability of gold nanoparticles in singlet oxygen photogeneration. The results are in accordance with previous studies about singlet oxygen generation by gold nanoparticles [27, 28].

Fig. 3
figure 3

Temporal variation of a absorption spectrum of gold nanoparticle and DPBF solution and b natural logarithm of DPBF absorbance at 415 nm during the laser light illumination

Full size image

Toxicity of gold nanoparticles

Figure 4 depicts antibacterial properties of utilized gold nanoparticles in concentrations of 0.5, 1, 5, 10, and 50 μg/ml on E. coli ATCC 25922. According to this figure, bacterial viability was decreased when concentration of gold nanoparticles was increased. So that, for the minimum concentration (0.5 μg/ml) and maximum concentration (50 μg/ml), the inhibition reached to 2.5 and 26%, respectively. That means the incubation of gold nanoparticles at concentrations of 0.5 and 5 μg/ml led accordingly to 0.09log and 0.36log reduction of viability, respectively. Therefore, although the use of these gold nanoparticles was accompanied with toxic effects on the bacteria, but, this effect was not significant in the concentration below 0.5 μg/ml. In order to use these gold nanoparticles for the inactivation process, we used gold nanoparticles at 0.5 μg/ml that revealed least toxic effects on E. coli ATCC 25922.

Fig. 4
figure 4

Effect of different concentrations of gold nanoparticles on the viability of E. coli ATCC 25922

Full size image

Bacterial inhibition by the Nd:Yag laser

Figure 5 shows the effect of the employed Nd:Yag laser with the wavelength of 532 nm and the power density of 40 mW/cm2 on the viability of E. coli ATCC 25922. According to this figure, viability reduction of the bacteria during 15 min illumination reached to a value of 10.5% and this reduction continues with an increasing trend over time until 60 min. According to our results, after 60 min illumination, a 30% reduction was observed on the population of the bacteria. In other words, light exposure by the LED system for 60 min led to 0.15log reduction of viability. However, this 30% reduction has no significant effect on inactivation of the bacteria. So, based on our experimental results, in the absence of gold nanoparticles, the light did not show a significant impact on the process of bacterial destruction. In order to justify these reductions, we invoke the report of Huang et al. (2007). They showed that light at wavelengths less than 600 nm is absorbed by macromolecules and proteins of the cell, and it also causes several changes in enzyme balance of bacteria during photochemical reactions [29]. Finally, these chemical imbalances may cause the death of the bacteria.

Fig. 5
figure 5

The impact of the Nd:Yag laser beams on E. coli ATCC 25922 at different illumination times

Full size image

Synergistic effects of Nd:Yag laser and gold nanoparticles

Figure 6 shows temporal variations of the viability when the bacterial solutions containing 0.5 μg/ml gold nanoparticles were illuminated by the laser for 15, 30, and 60 min exposure times. According to the results of our experiments, 15 min light illumination of the solution led to a considerable reduction of viability equal to 2.052log. In addition, based on our observations, this logarithmic reduction increased for 30 and 60 min irradiation times to 2.24log and 2.43log, respectively. Therefore, in the presence of nontoxic gold nanoparticles, the illumination of the bacterial solution by the laser source led to 99.11, 99.42, and 99.63% death of E. coli ATCC 25922 after 15, 30, and 60 min light exposure, respectively. Figure 7 depicts a comparison between the logarithmic viability reduction of the bacteria wherein the effect of the presence of gold nanoparticles was studied on the efficiency of light inactivation of the bacteria. As a result, based on Figs. 4–7, unlike individual effects of the laser light and gold nanoparticles, their synergistic effects revealed a considerable reduction at the same condition.

Fig. 6
figure 6

The synergistic effects of the gold nanoparticles and irradiation of the CW second harmonic Nd:Yag laser on photodynamic inactivation on E. coli ATCC 25922

Full size image
Fig. 7
figure 7

Viability reduction of E. coli ATCC 25922 due to light illumination in the presence or absence of gold nanoparticles

Full size image

In these studies, we compared temperature variations of a 0.5 μg/ml AuNP solution and the photodynamic inactivation group before and after 60 min illumination by the laser source. For the gold nanoparticles dissolved in deionized water, temperature rise was just recorded, ~ 0.1 °C. On the other side, 60 min illumination of the photodynamic inactivation group also led to an increase in bulk temperature of the solution ∼ 1 °C. Since the initial temperature of these solutions was recorded at about 24 °C, therefore, this change on temperature of the photodynamic group could not lead directly to death of the target microorganism through the hyperthermia mechanism. This claim is justified based on the paper previously reported by Zhenpeng Qin and John C. Bischof [30]. They obtained the necessary intensity for 10 °C temperature rise for single 30-nm gold nanoparticles and a 109 cell/mm3 system containing 1.83 × 1010 AuNPs/ml near to 104 and 2 W/cm2, respectively. These values are both much greater than the intensity of our employed laser source.

Based on the laser-tissue interaction diagram, the dominant mechanism for cell death in the power density of 40 mW/cm2 and exposure time more than second order is photochemical interactions [30]. We believe that the dominant mechanism for death of the bacteria in our experiments was, therefore, photochemical processes. In a simple initial model, gold nanoparticles (AuNP) were excited through absorption of the laser beam. After that, the excited gold nanoparticles (AuNP∗) produced highly toxic reactive oxygen species (ROS) (particularly singlet oxygen (1O2)) during their collisions with the nearby molecules or molecular oxygen (3O2) [31,32,33].

In our experiments, no photosensitizer was fed to the bacterial solution. Therefore, as first reason, we think the macromolecules and proteins inside the microorganism may be excited by light and produce ROS through their non-radiative relaxation pathways. This claim was already approved by some reports [34,35,36]. Not only has this fact played a direct role in justifying the toxic effects of the green laser light (for the light control group) but also it has been contributed in the results of the photodynamic inactivation group.

It must be mentioned that there are some studies about photosensitizing potentials of platinum, iridium, and gadolinium under the irradiation of visible light sources [17,18,19]. Similar to our results, these studies also emphasized on the fact that energy of the employed light is much lower than their photoelectric threshold. In the case of gold, the work function is ~ 5 eV that means the photoelectric effect occurs only when it is irradiated by UV rays by λ ≤ 240 nm. Whereas the energy of the employed visible laser light (λ = 532 nm, E = 2.33 eV) was not sufficient to produce photoelectrons, the increase of singlet oxygen generation is likely justified based on the energy transfer from the excited metal (in our study gold nanoparticles) to1O2 through the 1Δg → 3Σg transition [37]. As a result, the excited gold nanoparticles lose a part of their obtained energy through collisions with macromolecules, proteins, and dissolved molecular oxygen to generate more toxic ROS. We think these generated ROS had a great impact on the rate of microorganism inactivation through irreversible oxidation of bacterial organs. Supposing singlet oxygen (1O2) as the main produced ROS during light excitation of AuNPs, this simple model explains the bacterial death through the following equations:

$$ {\displaystyle \begin{array}{c}\mathrm{AuNP}+\mathrm{h}\upnu \overset{\mathrm{Absorption}\ \mathrm{of}\ 532\ nm}{\to }{\mathrm{AuNP}}^{\ast}\\ {}{\mathrm{AuNP}}^{\ast }+{}^3{O}_2\overset{\mathrm{Nonradiative}\ \mathrm{energy}\ \mathrm{transfer}}{\to }\ \mathrm{AuNP}+{}^1{O}_2\\ {}{}^1{O}_2+\mathrm{Bacteria}\to {\mathrm{Bacteria}}_{\mathrm{ox}}\end{array}} $$

This claim was invoked based on a number of reports [38,39,40]. For example, Pasparakis observed photooxidation of 1,3-dyphenylisobenzofuran (DPBF) in the presence of gold nanoparticles during light illumination of the solution [41]. In these experiments, Pasparakis compared the UV absorption quenching of DPBF induced by a CW 50 mW/cm2 green Nd:Yag and a pulsed nanosecond green Nd:Yag laser (7 ns, 532 nm). With regard to the previous report, singlet oxygen generation by light illumination of gold nanoparticles was observed for both CW and pulsed lasers with a preference of the pulsed laser. In addition, two photon-induced singlet oxygen generation was reported by Cuifeng Jiang et al. (2013) when gold nanoparticles were irradiated by an unfocused femtosecond laser beams at 800 nm with pulse duration of 60 fs and energy power density of 3 W/cm2 [42]. In accordance with our observations, the application of naked gold nanostructures for in vitro photodynamic treatment of cancerous cells was also reported [43,44,45,46,47,48].

Conclusion

In this study, the ability of a low-power density Nd:Yag laser to destroy E. coli ATCC 25922 was evaluated. Our results point to the second harmonic Nd:Yag laser beams with a power density of 40 mW/cm2 did not cause salient effects on the population of this bacteria alone. But laser exposure in the presence of nontoxic gold nanoparticles led to considerable impact on the viability of the bacteria. This cooperation has been able to purposefully destroy microorganisms without causing macroscopic effects of temperature on the solution. Therefore, this article gives us promising results about photodynamic inactivation of microorganisms in low intensities of laser light. This result is considered as a valuable merit for our method especially considering damages on neighbor tissues generated by high-power density lasers (Fig. 8).

Fig. 8
figure 8

Mechanism of photodynamic inactivation of bacteria through light excitation of the gold nanoparticles

Full size image

On the other hand, in order to enhance the efficiency of this PDI method, introduction of new strategies to increase generated singlet oxygen rate and consequently reduction of annihilation time will be demanded. For this, the effects of laser power density, size, and shape of gold nanostructures could be considered in the future. Furthermore, using suitable biomarkers for targeted photodynamic inactivation of pathogenic bacteria is proposed for future researchers. Finally, we propose modification of the photodynamic inactivation of microorganisms through simultaneous excitation of photosensitizer and gold nanostructures using two suitable light sources.