1. Introduction

In nature, a significant proportion of microorganisms exist in a structured ecosystem, called biofilms, where cooperation is critical for survival under suboptimal conditions [1,2]. Bacterial biofilms consist of organized communities of bacteria that form an extracellular matrix, allowing them to adhere to surfaces and shield themselves from hostile surroundings. In response to environmental changes, some bacteria secrete extracellular polymeric substances (EPS), a sticky glue-like matrix, which allows them to bind to each other, attach to a surface, aggregate into clusters, and eventually form biofilm [2–6]. Bacterial biofilm allows bacteria to survive environmental challenges such as inadequate nutrition, antibiotics, heat, or other stress factors [2,5–7]. This extracellular matrix barrier often comprises macromolecules including carbohydrates, peptidoglycans, and proteins [8].

The ability to synthetically generate biofilms can be useful or detrimental to humanity. Biofilms may be found on both living and non-living surfaces in a wide range of environments including medical, and industrial settings. Bacterial biofilms are of significant importance because of their ability to form complex and highly organized communities that can withstand difficult environmental conditions. Bacterial biofilms can be the source of antibiotic resistant bacteria causing serious issues in healthcare settings [9–13] as well as in other industries, such as food processing equipment and industrial pipelines [14]. On the other hand, biofilms of harmless bacteria, can be utilized in bioreactors and for the development of novel biomaterials [15,16]. In recent years, there has been a rise in interest in investigating bacterial communities and assessing methods for controlling biofilm formation and development [17,18].

In previous studies, scientists focused predominantly on mechanical, chemical, and biological approaches to suppress and control biofilms [15, 19–22]. For instance, biofilm development has been modulated by chemical interception or stimulation of regulatory processes responsible for biofilm development and intercellular signaling [20]. Synthetic or chemical approaches for activating and controlling biofilms have proven to be effective for the production of biomaterials and the engineering of biofilms into specific spatial structures [23–26]. In these approaches, genetic circuits and cell signaling are altered, allowing for the regulation of biofilm formation and the production of “programmable” biomaterials [24]. Using a synthetic method, “biofilm lithography” has been demonstrated, in which biofilm is only formed on illuminated surface regions by inducing a light-dependent phenotype [26]. Nevertheless, little is known about the optical methods to control biofilm dynamics, which can be used to regulate bacterial biofilms without introducing genetic light-dependent modifications.

From Ashkin's early work on optical trapping [27–29], for which he was awarded the Nobel Prize in Physics in 2018, the ability to trap and manipulate particles and live cells has been widely employed in biological, medicinal, atmospheric, and other multidisciplinary research applications [28–37]. When a living cell or a group of cells are near the highly focused laser beam, they experience an optical force pulling them toward the beam focus. The optical trapping technique has been successfully used to study DNA [30–32], viruses and bacteria [28,33,34], single proteins [35], neurons [36], cancer cells [37], and many other living organisms.

In this study, we investigate the effect of optical trapping on the aggregation and biofilm formation of Bacillus subtilis in minimal biofilm-promoting media (MSgg media). Specifically, we determine the optimal stage in bacterial biofilm formation for optical manipulation and investigate the impact of optical trapping at various wavelengths on biofilm development. We find that the ideal bacterial clusters for optical manipulation consist of three to fifteen cells, at which point the bacteria have undergone changes and secreted glue-like substance, but the clusters are still able to move under the influence of optical forces. The use of a laser with a wavelength range (820 nm - 830 nm) enables the prolonged optical trapping of biofilm clusters while minimizing the possibility of significant photodamage. Conversely, the employment of a laser with high optical absorption (473 nm) results in cell rupture and the disintegration of biofilm clusters. The investigation of optically controlled biofilm formation using optical tweezers offers novel methods for manipulating and controlling biofilm growth using optical technologies.

2. Methods

2.1 Bacterial strain and MSgg media

Bacillus subtilis strain NCIB 3610 (from Bacillus Genetic Stock Center (BGSC)) was chosen for our research due to its natural biofilm-forming capabilities and non-pathogenic nature [5,38,39]. B. subtilis is a rod-shaped, unicellular, gram-positive, 4 - 10 µm long soil bacterium. In harsh environments, B. subtilis forms biofilm by clustering together. This undomesticated strain of B. subtilis is a well-known model organism for biofilm studies.

As our primary experimental nutrient medium, we selected minimal salts glycerol glutamate (MSgg) medium, which is low in nutrients for B. subtilis and promotes biofilm formation. The composition of MSgg, a minimal biofilm-promoting media, includes 86.6 ml sterile ddH₂O, 10 ml 1 M MOPS (pH 7.0), 1.0 ml 0.132 M potassium phosphate buffer (pH 7.0), 1.0 ml 50% glycerol, 1.0 ml 50% glutamate monosodium salt, 0.1 ml 0.7 M CaCl₂, 0.1 ml 2 M MgCl₂, 0.1 ml 0.1 M FeCl₂, 0.1 ml 50 mM MnCl₂, 0.004 ml 25 mM ZnCl₂, 0.002 ml 0.1 M thiamine HCl [5,17]. To conduct consistent experiments, we prepare large volumes of stock solutions in advance and then combine the components of MSgg media at the time of experimentation. All stock solutions were made with deionized autoclaved water and then filtered for sterilization purposes. The low-nutrient environment is hostile to B. subtilis, which prompts the bacteria to begin biofilm formation.

2.2 Biofilm preparation for optical manipulation

B. subtilis is grown from −80°C frozen stock on Luria-Bertani (LB) agar plates. Then a single colony from the LB plate is inoculated into 3 ml of LB broth in a 15 ml culture tube and incubated in a water bath shaker (Fisher Scientific; Isotemp 15 L) at 37°C and 150 rpm for 12-15 hours. Next, we transfer 300 µl of culture in LB media to 2700 µl of MSgg media in a 15 ml culture tube. We repeat this dilution step for a second time to ensure the dilution of LB in the experimental culture. After the series dilution, the final dilution of culture in LB media to MSgg media is 1:100. The culture in MSgg media is then incubated in a water-bath shaker at 37°C for 3-5 hours to promote biofilm formation. This culture contains cells in small clumps which have the potential to start the formation of biofilm. A longer time of incubation corresponds with the growth of B. subtilis in media, the resulting size of bacterial clusters, and the amount of biofilm formed. The bacterial growth curve of B. subtilis in MSgg media as a function of incubation time obtained using a spectrophotometer is shown in Fig. S1 in the Supplement 1.

2.3 Optical tweezers experimental setup

After obtaining a culture containing small biofilm clusters, we use our experimental setup to conduct optical trapping experiments. Figure 1 illustrates the experimental configuration. The optical tweezers system is constructed around an inverted microscope (Olympus IX83) which has an optical breadboard deck and allows the inclusion of optical components just before a microscope objective. Depending on the intended laser wavelength for our investigations, we use one of two continuous-wave lasers: a blue laser (Laser Quantum, gem473 DPSS, λ = 473 nm) or a NIR tunable Ti: Sapphire laser (Msquared, SolsTiS 4000 PSX XF, λ = 700 - 1000 nm). Incorporating reflected collimators (Thorlabs, RC08FC-01) and a customized single-mode fiber (Thorlabs), the laser beam is delivered to the microscope system. The dichroic mirror (Semrock FF495-Di03-25 × 36 or Semrock FF647-SDi01-25 × 36) reflects the laser beam into an objective for optical trapping while permitting illumination to pass through to the camera for imaging. The additional notch filter (LP03-532RE-25 or FF02-694/SP-25) is included before the camera (Hamamatsu, ORCA-FLASH4.0LT) to completely block the laser beam. The objective lens with a high numerical aperture (100x 1.4 NA) focuses the laser beam on the sample, creating strong optical gradient forces capable of trapping micron-sized objects. To permit prolonged observation under the microscope, a custom-built environmental chamber surrounding the microscope maintains a temperature of 37°C. The bacteria sample is placed in a petri dish with a cover glass at the bottom (Cellvis, D29-20-1.5-N) and a custom-built microfluidic insert. A syringe pump (Harvard Apparatus, PHD ULTRA) supplies new MSgg media at a flow rate of 500 µl per hour.

 

Fig. 1. Schematic illustration of the experimental setup to observe biofilm formation with optical tweezers. The laser beam is delivered through a fiber into the microscope and focused by the high numerical aperture objective into the sample. The environmental chamber and the syringe pump provide proper temperature and nutrition for prolonged bio-sample observation.

Download Full Size | PDF

3. Results and discussion

3.1 Characterization of biofilm growth and optimal size of bacterial clusters for optical manipulation

We use MSgg media to promote B. subtilis to begin the process of biofilm formation and observe bacterial aggregation by taking a small sample of cells from the culture tube. Our studies are focused on the initial phase of biofilm formation, since it is much simpler to optically manipulate small bacterial clusters into the desired location and construct bacterial patterns. We established that the optimal sizes of bacterial clusters for optical manipulation are comprised of around three to fifteen cells per cluster, a stage reached after three to four hours of growing in MSgg medium with minimal nutrition. Depending on the exact nutrition concentration and cell replication rate in the medium, the precise moment at which the optimal small clusters develop will vary. Figure 2 depicts the growth and aggregation of B. subtilis in MSgg media over time. Cells produce insufficient EPS after 2 to 3 hours to adhere cells together (Fig. 2(a)). These cells begin to cluster together after 4 hours (Fig. 2(b)). Cells have already begun to form larger clusters after 5 hours, and clusters of cells have begun to attach to surfaces (Fig. 2(c)). After 20 hours, a large biofilm is developed that is difficult to manipulate with optical forces (Fig. 2(d)).

 

Fig. 2. Biofilm growth of B. subtilis in MSgg media. (a-d) Clustering and biofilm development of B. subtilis over time after inoculation: after 2.5 hours, 4 hours, 5 hours, and 20 hours.

Download Full Size | PDF

The ideal growth stage and size range of bacterial clusters for optical manipulations have been determined based on the duration required for bacteria to undergo changes and through optical trapping experiments. The lower limit for an optimal cluster size is established by ensuring that the bacteria exhibit a response to a low-nutrient environment and initiate the production of an adhesive substance to form biofilm. If the culture in MSgg media consists only of a few clusters with 2-3 attached cells, it indicates that the bacteria may continue to undergo regular division, with daughter cells staying attached to one another. However, if clusters of 3-4 cells begin to form in the bacterial sample, it is nearly certain that the bacteria are held together by an adhesive substance. The upper limit for an optimal cluster size is determined by the ability to manipulate clusters using optical tweezers. For our experiments, we use NIR laser at 820 nm–830 nm wavelength with a 30-50 mW power for our optical manipulation. A cluster with a size exceeding approximately 15 cells is more likely to undergo a relatively rapid attachment to a glass surface. Once a cluster adheres to a glass surface, it becomes significantly hard to move it using optical forces, mainly because of the strong adhesive properties of the glue. If a cluster is connected to a glass surface by only a small portion of bacteria, it can be forcefully detached using a high-power laser [34]. Clusters of a bigger size become attached to the glass surface within a few minutes of sample preparation. Whereas the smaller size clusters can still be found freely moving in MSgg media 30- 60 minutes after the sample preparation in a petri dish. Occasionally, in freshly prepared samples, we successfully found and optically manipulated clusters of over 20 or 30 cells per cluster; however, these clusters have a higher tendency to stick to a surface. So, for future bacteria biofilm applications, selecting a stage with smaller bacterial clusters allows for more time to move and assemble bacterial clusters using optical tweezers.

After 4 hours of incubation in a water bath shaker, B. subtilis in MSgg media is transferred to a microfluidic petri dish for microscope observation. For prolonged biofilm formation experiments, 500 µl of culture is transferred into a petri dish, and fresh MSgg media is continuously supplied at a flow rate of 500 µl per hour. To decrease the bacterial concentration for optical trapping studies, the culture is diluted at a ratio of 1:4 with fresh MSgg medium. Without the bath shaker, which shakes the sample and prevents attachment to surfaces, the bacteria aggregate and stick to surfaces more easily. Figure 3 shows the aggregation of cells and the adhesion of biofilm clusters to each other over time. The sample is illuminated with a regular white light source from the microscope for viewing purposes. Within just 30 minutes, the bacteria aggregate to form substantial clusters (Fig. 3(c)). After 2 hours, they begin to establish a network on the surface of the cover glass (Fig. 3(g)). Visualization 1 shows bacteria aggregation and biofilm formation over 3 hours.

 

Fig. 3. B. subtilis aggregation and biofilm formation in MSgg media over time without laser exposure. The bacteria were cultured for 4 hours in a bath shaker and then transferred to a microfluidic petri dish device for microscope observation.

Download Full Size | PDF

3.2 Biofilm growth dynamics under focused laser light illumination from optical tweezers

In order to manipulate bacterial clusters optically for extended periods of time using a laser beam, low-absorption wavelengths must be selected. The absorbance measurement of B. subtilis in MSgg media after 4 hours of inoculation and MSgg media alone is shown in Fig. 4. The bacteria in the MSgg medium have a large absorption band below 400 nm and an absorption peak from water at 980 nm. As ultraviolet (UV) light damages cellular DNA, it is expected that lower visible wavelengths will influence the growth and composition of bacterial biofilm [40–42]. Prolonged exposure to shorter visible wavelengths has the potential to induce rapid cellular heating and permanent damage to cells. Therefore, the optimal spectrum ranges for optical manipulation and trapping are between 800 nm and 930 nm and 1040 nm and 1100 nm, respectively, in order to minimize damage.

 

Fig. 4. UV-Vis absorption spectra of MSgg media and B. subtilis in MSgg after 4 hours after inoculation.

Download Full Size | PDF

To determine the exact effect of laser damage on cells, a phenomenon often called as “phototoxicity”, we compare bacteria growth without laser and with different types of laser illumination. For this study, we cultured B. subtilis in MSgg media for 4 hours before transferring them to a microfluidic petri dish device for microscope observation and laser manipulation. Because bacteria move rapidly when first placed into a petri dish, we let the bacteria settle in for around half an hour before beginning any observations. After half an hour, the bacteria aggregate into large clusters, as shown in Fig. 3(b,c), allowing us to investigate the effect of laser exposure on biofilm formation and bacterium development. When the 473 nm laser, a highly absorbing wavelength of light, is turned on and an optical trap is present, the bacterial clusters are attracted to the beam and begin to disintegrate and break apart. Figure 5(a-d) shows the disintegration of a large bacterial cluster due to exposure to the optical trap using the 473 nm laser at 15 mW laser power (see Fig. S2 in Supplement 1 for additional images). The 473 nm laser generates considerable photodamage to trapped bacteria clusters and disrupts biofilm formation relatively quickly. The cells burst after 10 to 15 minutes of 473 nm laser exposure at 15 mW laser power at the focal plane or after just 1-2 minutes of exposure at 30 mW laser power. However, even after 1 hour of being subjected to optical trapping at 820 nm laser, a low absorbing wavelength of light, at 50 mW laser power, there is no notable photodamage (Fig. 5(e-h)). The bacteria were attracted to the optical trap, but the laser exposure had minimal negative effect on the development of biofilm. Furthermore, there is a minimal disparity in the size of bacterial clusters following a 1-hour exposure to an 820 nm laser (Fig. 5(h)), compared to the size of clusters without any laser exposure during a comparable time frame after being transferred to a petri dish (Fig. 3(e,f)). The optical trap exerts its effect just within close proximity of the laser focus spot, thereby resulting in localized phenomena such as cluster attraction and cell redistribution. Consequently, the NIR laser has minimal impact on the total formation of biofilm on a large scale.

 

Fig. 5. B. subtilis biofilm growth with laser exposure. (a-d) Prolonged exposure to a 473 nm laser (15 mW power) disrupts biofilm growth. (e-h) Prolonged exposure to 820 nm laser (50 mW power) no significant damage.

Download Full Size | PDF

During an hour of observing bacteria dynamics and biofilm formation with optical tweezers at 820 nm wavelength (Fig. 5(e-h)), we discovered that bacterial clusters aggregate near optically trapped clusters, adhere to the surface, and begin to form a microcolony. The region of the bacterial clusters exposed to the highly focused NIR laser causes no obvious disruption in biofilm formation. This implies that we can employ NIR wavelengths in the 800 nm - 850 nm range for extended periods of time, ranging from minutes to possibly hours, to achieve optical trapping, manipulation, and pattern formation of bacterial clusters. To avoid natural aggregation of cells during controlled pattern construction, the culture sample must be diluted with fresh MSgg media to reduce the concentration of cells.

When cells burst apart as a result of photodamage caused by prolonged exposure to a highly focused laser beam at strongly absorbing wavelengths (473 nm), the released broken pieces of cells cause significant scattering of the laser light. The glowing spots may be observed at the optical trap's site (Fig. 5(b,c)). Following the cell's rupture, optical tweezers capture highly scattering particles that are eventually destroyed and cease to scatter. However, if necrosis occurs as a result of low-power photodamage attributed to 473 nm laser, the glowing effect is rare. Cell death due to opticution manifests as a slowly fading contrast of the observed cell. Whereas, when a cell bursts due to a high-power laser, the pieces of the cell persist. Separate tests were performed using MSgg media and MSgg media with large quantities of iron to determine the source of the strong scattering. No laser scattering or absorption is seen in either MSgg media or MSgg media with high iron content. Consequently, we may infer that the MSgg medium does not contain metallic nanoparticles, which might accumulate inside cells. This indicates that the intense scattering is caused by other cell components released during the rapid bursting of bacterium cells. The specific cause for the intriguing strong shining and scattering of laser light when the cell is quickly ruptured by the laser is not clear from this study. It is possible that the quick burst of a cell does not provide enough time for enzymes to begin digesting the dying cell; therefore, the fragmented parts of the cell may have sharper edges, causing a significant scattering of light. Further investigation into how burst cells shine is needed. Future studies might include incorporating Raman spectroscopy into an optical tweezer system [43,44] or isolating the burst cells and carrying out SIM imaging.

3.3 Optical manipulation of bacterial clusters

We next test the optical manipulation ability of bacterial clusters at low absorbing wavelengths. For this study, we select clusters with a smaller size of 3-5 cells, which form 3 to 4 hours after inoculation, and clusters with a bigger size of 8-15 cells, which form 4 to 5 hours after inoculation. Estimating the precise number of cells in each cluster can be challenging due to the limitation of just being able to observe a two-dimensional image of the cluster. For these experiments, the bacteria already secrete EPS, but clusters are still mobile. To manipulate and position these bacterial clusters for attachment to neighboring clusters or a glass surface, we use a low-power (30 mW) NIR laser (820 nm - 830 nm) (Fig. 6). The optical trapping of B. subtilis cluster comprising around 3 cells and suspended in MSgg medium is shown in Fig. 6(a,b). Figure 6(c,d) demonstrates how optically trapped bacterial clusters may be moved throughout the sample to the appropriate position. Figure 6(e-h) illustrates of optical manipulation of a cluster of larger size, comprising around 10 cells.

 

Fig. 6. Optical manipulation of a B. subtilis biofilm bacterial cluster. (a-d) Optical trapping and movement of a small cluster comprising around 3 cells by using the 830 nm laser (30 mW power). (e-h) Optical trapping and manipulation of a large cluster comprising around 10 cells by using the 820 nm laser (30 mW power).

Download Full Size | PDF

It is critical to keep the time and laser wavelength in mind while arranging bacterial clusters into desirable places and patterns. Because cells prefer to attach to glass surfaces over time and begin the production of biofilms at surfaces, it is essential to manipulate bacterial clusters relatively immediately after preparing the sample. It is also vital to select low absorption wavelengths with low laser power for optical manipulation of bacterial clusters in order to cause the least amount of disruption to the cell's health. If the bacterial cluster is transported to the appropriate area quickly and with minimal exposure to the laser, it can continue its growth and cell division.

4. Conclusions

In response to suboptimal growth and environmental conditions, some bacteria engage in the production of biofilm as an inherent mechanism of self-protection. By regulating the nutrient content of the surrounding environment, we were able to regulate biofilm synthesis and determine the optimal time interval for the formation of bacterial clusters. After four hours in MSgg medium, B. subtilis aggregate into clusters of small size that are suitable for optical manipulation into the desired location and pattern. We determined that for optical trapping and manipulation the ideal bacterial cluster sizes range from three to fifteen cells per cluster. The utilization of a microfluidic petri dish enables the controlled delivery of new MSgg medium to the bacterial sample, enabling a prolonged investigation over several hours into the formation and dynamics of clusters under varying conditions, including exposure to laser light. By using a laser beam with a low absorption wavelength for the biofilm (820 nm - 830 nm), we successfully performed the manipulation and reorganization of bacterial clusters, while minimizing the possibility of significant damage. On the contrary, a laser with a high absorption wavelength for the biofilm exhibits strong heat effects, leading to the destruction of bacterial clusters. In addition, the utilization of high-power, high-absorption wavelengths causes the rapid bursting of optically trapped cells and the release of particles that exhibit strong scattering behaviors.

In conclusion, the utilization of optical trapping has provided new perspectives on the dynamics of the interactions between the biofilm of B. subtilis and light. Optical tweezers have the potential to exert optical control over the process of biofilm development, hence enabling the manipulation of bacterial growth patterns and structures in future applications. In order to do this, it is possible to utilize a near-infrared laser for the purpose of aggregating and manipulating bacterial clusters, directing them to specific locations as needed. Additionally, a blue laser may be employed to efficiently prevent the establishment of biofilms. The findings of this study have the potential to contribute to the advancement of innovative biomaterials composed of bacterial biofilms, as well as enhance our understanding of the complex interactions that occur inside biofilm communities. In future work, we will study how we can manipulate small clusters to construct desired structures made of bacterial biofilm and investigate how different types of laser beams can promote or repress biofilm growth.