Introduction

Regenerative medicine is a multidisciplinary field of science that aims for the functional and physical renewal of organs and tissues, as well as the enhancement of their functions via therapeutic methods or applications, for the ultimate improvement of life quality of patients1,2. The identification of stem cells and the recognition of their therapeutic potential have brought a new direction to tissue engineering and regenerative medicine3; where the most frequently used cell type is mesenchymal stem cells.

One of the commonly used methods to increase regenerative capacity is photobiomodulation. The term photobiomodulation refers to biological changes in live cells, such as cellular activity or metabolism, in response to irradiation with a specific wavelength of light under certain conditions. Photobiomodulation, also referred to as low-level laser therapy (LLLT), is executed with the use of red and near-infrared lasers with a wavelength of 600–1100 nm at a low energy density (fluence) of 0.04–50 J/cm2 in a continuous or pulsed mode4,5. This optical window enables a maximal tissue penetration with a minimum absorption of light by water6. Photobiomodulation was indicated to promote healing in injured cells and tissues, improve regenerative capacity7, modulate cell metabolism8, enhance cell migration during gingival wound healing9, reduce inflammation and accelerate wound healing in diabetic patients10. Moreover, it was also reported to increase cell viability and proliferation11, modulate inflammation12, reduce apoptosis and support cell migration, which may play a key role in tissue repair13 in mesenchymal stem cells. Although efforts are being made, the exact molecular mechanisms of photobiomodulation could not be explained in detail yet. Unlike pharmaceutical agents, the effects of low-level laser irradiation on cells depend on parameters such as the laser’s wavelength and applied energy density as well as the type of the irradiated cell11,14. Studies reported that photobiomodulation via usage of 600 to 700 nm wavelengths stimulate proliferation and increase the differentiation of mesenchymal stem cells11. Within this wavelength range, Moore et al.15 have demonstrated that the most efficient cell proliferation response of mouse vascular endothelial cells and fibroblasts were observed in 665 to 675 nm. The use of red and infrared light in photobiomodulation was reported to exert a biphasic dose response, where low energy density (fluence) results in weak responses at the cellular level, the increase in energy density results in beneficial effects and high fluence may disrupt cell proliferation and even induce apoptosis16,17. For example, low-level laser irradiation at low fluence increases tumor growth factor expression, while higher fluencies tend to suppress it18,19. Therefore, assessing dose-dependent effects of photobiomodulation is crucial. In addition, increased cell metabolism and proliferation were observed in both normal and cancer cell lines after photobiomodulation. Consequently, a better understanding of the cellular processes is pivotal for determining probable outcomes of photobiomodulation or for the design of efficient treatment protocols.

Recent scientific advancements highlight that cellular processes, stemming from molecular functioning, are deeply interconnected and largely reliant on the structural and dynamic properties of biomolecules20. Pathological conditions or therapeutic interventions can significantly alter the molecular structure, composition, and dynamics of tissues, cells, and membranes. Therefore, understanding the relationships between structure and function is essential for recognizing disease pathophysiology and the mechanisms of therapeutic actions. Fourier transform infrared (FTIR) spectroscopy is a fast, reliable, and increasingly valuable tool for examining the structural and dynamic properties of biomolecules in various samples. The recent increase in its application underscores its effectiveness in accurately interpreting cellular functions and metabolic processes21,22,23,24,25,26,27,28,29,30,31,32,33. FTIR spectroscopy was previously used to analyze the effect of Er-YAG laser on human dentin34 and on rat bone slices together with human enamel and dentin35. As far as is known, FTIR spectroscopy was previously used to characterize molecular and biochemical alterations following photobiomodulation on snake venom intoxicated muscles36 and MCF-7 breast cancer cells37. Magrini et al.37 have reported that light can influence cell metabolism of MCF-7 cancer cells depending on the laser fluence.

This study aims to evaluate the initial biophysical response of adipose-derived mesenchymal stem cells (ADSCs) when exposed to photobiomodulation, a 671 nm low-level laser irradiation at a fluence of 25 J/cm², by the use of FTIR spectroscopy. The use of 671 nm was selected as this wavelength falls within the effective photobiomodulation range for ADSCs, providing an optimal tissue penetration and mitochondrial activation. As far as is known, this current study provides the first detailed FTIR-based spectral characterization of biomolecular responses in ADSCs following photobiomodulation. The structural, compositional and dynamic response in ADSCs at 1-hour and 3-hour post low-level laser exposure was reported for the first time in this current study. Our findings indicated ongoing differences over time in lipid composition and structure, protein structure and phosphorylation, metabolic turnover, membrane order and fluidity in ADSCs exposed to 671 nm low-level laser. The herein reported short-term structural and compositional responses to photobiomodulation might provide new insights into the mechanism of low-level laser action at cellular level and have implications for the design of future therapeutic applications in regenerative medicine.

Results

Cell proliferation and Hsp70 protein expression levels

Figure 1a shows a summary of the applied experimental procedures in the study and Fig. 1b demonstrates the experimental setup for laser irradiation. After the isolation and characterization of ADSCs, the cells were exposed to 671 nm laser with energy densities of 0 (control), 1, 5, 10, 25 and 50 J/cm2 and cell viability of the irradiated cells were determined via MTT test (Fig. 1c). Cell viability was significantly improved following laser irradiation at fluencies of 25 and 50 J/cm2, with the most notable effect seen at 50 J/cm2. Although the dose of 50 J/cm2 resulted in a more pronounced increase in cell viability in initial tests, the exposure time required for this dose -due to the single-channel laser system and sequential well illumination - was twice than that of 25 J/cm2. Consequently, the usage of 50 J/cm2 could significantly prolong the total experiment time and thereby could introduce variability in incubation conditions across wells. In order to maintain consistency, minimize time-based artifacts, and ensure compatibility with live-cell analyses, we proceeded with 25 J/cm2 as the highest practical dose for detailed investigation. This decision effectively limited the irradiation time to around 3 min per well for subsequent experiments to reduce environmental stress and maintain cell viability.

Fig. 1
Fig. 1
Full size image

(a) The schematic diagram of the timeline of the applied procedures. (b) The experimental setup for laser irradiation to the ADSCs. (c) the results of the MTT test demonstrating the changes in cell viability upon laser exposure at different energy densities of 0 (control), 1, 5, 10, 25 and 50 J/cm2. (d) Western blot images of Hsp70 and GAPDH protein levels and (e) the fold change in Hsp70 protein expression levels compared to that of GAPDH in ADSCs of control, 1 H post-L and 3 H post-L groups. The full-length Western blots used for calculation are presented in Supplementary Figure S1. The data of MTT test and Hsp70 protein expression levels were compared by the use of one-way ANOVA with Dunn’s test as a post-hoc test. The degree of significance was denoted as *p < 0.05. The analysis of Hsp70 protein expression levels revealed non-significant alterations in between the groups.

Figure 1d demonstrates the Western blot images of Hsp70 and GAPDH proteins and Fig. 1e shows the fold change in Hsp70 protein expression levels compared to that of GAPDH in ADSCs after exposure to 25 J/cm² laser irradiation. The corresponding full-length blots are presented in the Supplementary Figure S1. The Hsp70 protein expression levels were decreased in 1 H post-L and increased in 3 H post-L groups, but the alterations were not statistically significant.

FTIR spectral characterization

Figure 2 demonstrates the representative FTIR spectra of control, 1 H and 3 H post-L groups in 4000–850 cm−1 (a), 3050–2800 cm−1 (b), 1800–800 cm−1 (c) and 1200–850 cm−1 (d) spectral regions. The assignments of the labelled spectral bands in Fig. 2 were given in Table 1. The alterations in the frequencies, band areas and bandwidths of some of the spectral bands were represented in Table 2. The shift in frequency of a spectral band give information on the structural alterations in the related function group, whereas the intensities and/or the area under spectral bands give relative concentrations of the related functional groups according to Beer-Lambert’s law33. In order to eliminate potential experimental misrepresentation, band area ratios were used for a more precise comparison of the effects of low-level laser irradiation on AD-MSCs. The calculated band area ratios and their interpretation were given in Table 3, and the results were represented in Fig. 3. FTIR spectra of 1 H and 3 H post-L groups displayed apparent differences in most of the spectral bands. In order for a more explicable representation, the results of spectral characterization were organized in three subsections as lipids, proteins and protein structure, nucleic acids and carbohydrates.

Fig. 2
Fig. 2
Full size image

(a) The representative FTIR spectra of control ADSCs in 4000–850 cm−1 region, the representative FTIR spectra of control, 1 H post-L and 3 H post-L groups (b) in 3050–2800 cm−1, (c) in 1800–800 cm−1, (d) in 1200–850 cm−1 region. Spectra were baseline corrected and normalized with respect to Amide A band located at 3282 cm−1 for visual demonstration.

Table 1 The assignments of the spectral bands labelled in Fig. 2.
Full size table
Table 2 Changes in the frequency, band area and bandwidth values of some absorption bands in the FTIR spectra of ADSCs in control, 1 H post-L and 3 H post-L groups.
Full size table
Table 3 The calculated area ratios of the IR bands, their biomolecular origins and indications.
Full size table
Fig. 3
Fig. 3
Full size image

The bar graphs demonstrating the alterations in band area ratios of (a) Unsaturated lipid content, (b) Unsaturated/saturated lipids, (c) Carbonyl amount in lipids, (d) Aliphatic chain length, (e) Lipid/protein, (f) Protein concentration, (g) Protein phosphorylation, (h) Collagen integrity, (i) RNA/PO2− antisym. + sym. stretching, (j) PO2− antisym. + sym. stretching/protein, (k) Glucose/protein, (l) Glucose/phosphate for control 1 H post-L and 3 H post-L groups. The data was analyzed via one-way ANOVA with Tukey’s test as a post-hoc test. The degree of significance was denoted as *p < 0.05, **p < 0.01, ***p < 0.001.

Lipids

The olefinic band located around 3010 cm−1 can be used to monitor unsaturated lipids, whereas the CH2 antisymmetric and symmetric stretching bands located at 2923 and 2852 cm−1, respectively, are used to monitor saturated lipids26,27,32,33,38. The area of olefinic band was observed to be slightly decreased in 1 H post-L group and significantly increased in 3 H post-L group in comparison to 1 H post-L group (p < 0.05, Table 2). In order to obtain a more precise result, unsaturated lipid content (A3010/A2957+2923+2852+1453) and unsaturated/saturated lipid content (A3010/A2923+2852) were calculated26,27,33 (Table 3). The unsaturated lipid content, which demonstrate the content of double bonds in lipids, and the unsaturated/saturated lipid ratio were both observed to be decreased in 1 H post-L group (p < 0.001) and to be increased in 3 H post-L group (p < 0.001, Fig. 3a, b). Moreover, the alterations in the carbonyl group amount in lipids (band area ratio A1740/A2923+2852)26,32, aliphatic chain length (band area ratio A2923/A2957)30 were monitored by the use of relevant band areas (Table 3). The carbonyl amount in lipids and lipid aliphatic chain length were both observed to be significantly decreased in 3 H post-L group (p < 0.05) in comparison to 1 H post-L group (Fig. 3c and d). These findings indicated that following laser irradiation, lipid composition and distribution were altered differently at 1-hour and 3-hour post-irradiation.

The frequency shifts and alterations in bandwidth of the CH2 antisymmetric stretching band give information on the lipid order and lipid dynamics of the system, respectively21,27,33. The wavenumber (band location) of the CH2 antisymmetric stretching band shifted to lower values (p < 0.05) in 1 H post-L group indicating an increase in the order of the system21. Whereas in the 3 H post-L group, the wavenumber of this band shifted to higher values (p < 0.001) compared to 1 H post-L group (Table 2), indicating a decrease in the order. The bandwidth of the CH2 antisymmetric stretching band was observed to be increased in both laser irradiated groups in comparison to control (p < 0.001, Table 2), indicating an increase in lipid dynamics, i.e. fluidity21,33.

Proteins and protein structure

Amide I and II are the main protein bands in an FTIR spectrum, and the band areas of amide I and amide II bands were both observed to be slightly increased in 1 H post-L group and significantly decreased in 3 H post-L group (p < 0.05, Table 2). Since band area ratios yield a more reliable comparison of the groups, the amide I/amide I + II (A1643/A1643+1542) ratio32,33 (Table 3) was calculated monitor the alteration in protein concentration of ADSCs. This ratio revealed that total protein content was not altered in control and laser irradiated groups (Fig. 3f). Lipid to protein ratio (A2923+2852/A1643+1542)32,33 was slightly decreased in 1 H post-L group, but was significantly increased in 3 H post-L group (p < 0.001, Fig. 3e). These data suggested that lipid content was significantly increased in 3 H post-L group, as protein concentration remained unchanged.

However, there were significant alterations in protein structure as revealed by FTIR spectral characterization. Firstly, the band located at 970 cm−1, which was attributed to the PO4–2 symmetric stretching mode of dianionic phosphate monoesters of phosphorylated proteins or nucleic acids23,38,39,40, was prominently notable in 1 H post-L group, when compared to 3 H post-L group (Fig. 2d) in the FTIR spectra. In order to obtain a more precise information, band area ratio of the PO2– antisymmetric stretching to the CH3 antisymmetric stretching (A1236/A2957) was calculated (Table 3) to comment on the level of protein phosphorylation in the ADSCs22,32,33. Protein phosphorylation was observed to be significantly increased in 1 H post-L group in comparison to control (p < 0.05), and significantly decreased in 3 H post-L group in comparison to both control (p < 0.01) and 1 H post-L group (p < 0.001) (Fig. 3g).

As laser irradiation was suggested to stimulate collagen production41, the influence of photobiomodulation on quality of collagen in ADSCs were monitored by the band area ratio A1542/A1340 (Table 3) calculated by dividing the area of amide II band to that of the band located at 1340 cm−1, which is assigned to CH2 side chain vibrations of collagen that can be used to monitor collagen integrity21,29. An increase in this ratio indicates the decrease in quality of collagen in the system together with a decrease in collagen integrity21. There was an increase in A1542/A1340 ratio in 1 H post-L group (p < 0.001), indicating the decrease in collagen quality, whereas the ratio was observed to be decreased in 3H post-L group (p < 0.001) in comparison to that of 1 H post-L group (Fig. 3h). This finding implicated that the quality of collagen gradually increases over time.

The secondary structure of proteins in experimental groups were analyzed via comparing peak intensities in the second derivative spectra in 1700–1600 cm−1 spectral region21 (Fig. 4). The results were presented in Table 4, and significant alterations in protein secondary structure were observed. The decreased intensities of peaks corresponding to antiparallel β-sheet (1681 cm−1) and α-helical (1659 cm−1)21,42 structures in 1 H post-L group were observed to be increased significantly in 3 H post-L group (p < 0.001) in comparison to 1 H post-L group. A similar alteration was observed in peak intensities of native β-sheet structures (1637 cm−1)21, but the alterations were not statistically significant. Moreover, the peak intensities of aggregated β-sheet structure (1628 cm−1)21 were slightly increased in 1 H post-L group and was significantly decreased in 3 H post-L group (p < 0.05) when compared to 1 H post-L group.

Fig. 4
Fig. 4
Full size image

The representative absorbance (a) and second derivative (b) spectra of control, 1 H post-L and 3 H post-L groups in 1700–1600 cm−1 region. The investigated protein secondary structural sub-bands under amide I band were labelled in the figure with numbers [1: antiparallel β-sheet structure; 2: α-helical structure; 3: native β-sheet structure; and 4: aggregated β-sheet structure].

Table 4 Alterations in the intensities of main functional groups in the second derivative FTIR data in 1700–1600 cm−1 region of FTIR spectra representing the alterations in protein secondary structure for ADSCs in control, 1 H post-L and 3 H post-L groups.
Full size table

Nucleic acids and carbohydrates

The most apparent differences in between 1 H and 3 H post-L groups were noted in the 1250–850 cm−1 region of the FTIR spectra (Fig. 2d). The bands located at 1236 and 1079 cm−1 are assigned to the PO2− antisymmetric and symmetric stretching vibrations, respectively, which both have contributions from nucleic acids and phospholipids21,24,28,32,33,43. As previously mentioned, the sub-band located at 970 cm−1 can be easily distinguished in the spectra of 1 H post-L group. Moreover, the sub-bands located at 1035 cm−1, which originates from the C-O stretching vibrations of carbohydrates and glucose26,28,33, and 990 cm−1, which is assigned to RNA backbone vibrations33,39,40,44, were clearly noticed in the spectra of 3 H post-L group (Fig. 2d). The band areas of these sub-bands were significantly increased in 3 H post-L group (Table 2).

The ratio of band areas of 990 cm−1 to the sum of the PO2− antisymmetric and symmetric stretching band areas (A990/A1236+1079) was used to monitor RNA content33 in the ADSCs. This ratio was observed to be significantly increased in 3 H post-L group (p < 0.001, Fig. 3i). This finding indicated that RNA and therefore protein synthesis was amplified at 3-hour post-irradiation, which will later on result in an increased protein content. Moreover, the PO2− antisymmetric + symmetric/protein ratio (A1236+1079/A1643+1542)33, which was calculated by the use of relevant band areas (Table 3), was observed to be significantly increased in 3 H post-L group (p < 0.001, Fig. 3j). This finding also indicated that the nucleic acid or phospholipid content was significantly amplified at 3-hour post-irradiation.

In order to more precisely comment on the utilization of carbohydrate resources in ADSCs, glucose/protein (A1035/A1643+1542) and glucose/phosphate (A1035/A1079) band area ratios were calculated26,33,45,46 (Table 3). These ratios were observed to be decreased at 1 H post-L group and significantly increased in 3 H post-L group, which revealed a significantly decreased utilization of carbohydrate sources in 3 H post-L group (p < 0.001, Fig. 3k and l). This was further supported by the increase in the glucose or carbohydrate band located at 1035 cm−1 in the spectra of 3 H post-L group (Fig. 2).

Discussion

The application of red or infrared spectrum in photobiomodulation may lead to inconstant outcomes in cellular processes or “hormesis” in cells, due to the biphasic dose response observed in this optical window16,17. Furthermore, photobiomodulation has been reported to give rise to differential outcomes such as stimulating physiological processes or inhibiting the progression of a pathological condition8,47, where the exact molecular mechanism is yet to be resolved. Accordingly, this spectral characterization study aimed to assess the short-term outcomes of photobiomodulation on composition, structure and function of biomolecules in ADSCs subjected to 671 nm low-level laser irradiation, utilizing FTIR spectroscopy. The results revealed apparent differences in the FTIR spectra at 1-hour and 3-hour post-irradiation. Upon low-level laser exposure at 671 nm, short-term alterations were observed over time in protein secondary structure, protein phosphorylation, lipid composition, metabolic turnover of carbohydrate sources, membrane structure and function with a temporary lateral phase separation in ADSCs.

In a cell, the content, composition and distribution of proteins and lipids, such as level of unsaturation, aliphatic chain length, membrane thickness etc., are crucial parameters for the structural and functional association of molecules, and are strongly connected to ion channel and membrane function20. Exposure to 671 nm laser irradiation at a fluency of 25 J/cm2 resulted in temporary alterations in lipid composition and structure in ADSCs. Mainly, the unsaturated lipid amount and the ratio of unsaturated/saturated lipids were significantly decreased at 1-hour post-irradiation, and were significantly increased after three hours. The content of carbonyl-ester group containing lipids remained unchanged at 1-hour post-irradiation, however, was significantly decreased after three hours. Moreover, the aliphatic chain length of lipids was slightly increased at 1-hour post-irradiation, and were significantly decreased at 3-hour post-irradiation. These findings indicate a change in lipid composition and distribution as a response to laser irradiation over time, which may lead to an irregularity in the distribution of lipids between the inner and outer leaflets of the bilayer, which refers to membrane asymmetry27. The variations in membrane asymmetry together with alterations in membrane thickness may in turn disturb membrane function, as structural organization of lipids and proteins may influence ion channel functioning and membrane potential20,48.

Lipid order is a structural parameter that reflects the organization and flexibility of individual acyl chains in lipids33. In contrast, membrane fluidity is a dynamic parameter that measures the mobility of lipid components, such as lateral diffusion. An increase in membrane order typically leads to enhanced membrane rigidity, while an increase in membrane fluidity is generally accompanied by a decrease in order, indicating greater acyl chain flexibility21. One hour after laser exposure, we observed an increase in order, demonstrated by a shift to lower wavenumber values of the CH2 antisymmetric stretching vibration, along with an increased fluidity, indicated by the increase in bandwidth of this peak. This inconsistency had been reported to designate the presence of lateral phase separation within the membrane structure21. Phase separation or domain formation is one of the physical phenomena resulting from the dynamic organization of the lipid bilayer. Not only lipids, but also proteins that are embedded or attached to membranes can be laterally arranged within a phase separated membrane according to the lipid-protein interactions49,50. One such example for the formation of immiscible domains are membrane rafts, which are liquid-ordered patches in membranes with an increased thickness compared to the surrounding and faster dynamics. Another important factor is that living cell membranes are in a thermodynamic non-equilibrium and constantly undergoes lipid turnover, membrane trafficking, etc51. Recently, Shen et al.51. have studied phase behavior of endoplasmic reticulum (ER) via stimulated Raman scattering microscopy in live cells, and reported that metabolism of saturated fatty acids, having high phase transition temperatures and longer fatty acyl chain lengths, led to solid-like domain formation in the ER. However, metabolism of unsaturated fatty acids, having lower phase transition temperatures, did not form membrane domains51. Therefore, lipid composition in membrane structure could be a determining factor for the observed lateral phase separation. Our results have also shown the altered lipid composition, a slightly increased aliphatic chain length at 1-hour post-irradiation together with a temporary lateral phase separation in membranes. However, at 3-hour post-irradiation, this separation was terminated with a decreased order accompanied by an increased membrane fluidity, along with a significant decrease in aliphatic chain length of lipids.

Membrane fluidity not only influences the permeability of the membrane, but the structure and function of membrane proteins are also sensitive to the composition, order and fluidity of lipid environments49,52. In addition to the observed increase in membrane order and fluidity one hour after irradiation, we noted changes in protein secondary structure, including a decrease in α-helical and antiparallel β-sheet structures along with a slight increase in aggregated β-sheet structures. Furthermore, at one-hour post-irradiation, the quality of collagen molecules appeared to decline while protein phosphorylation showed a significant increase (Fig. 3). Besides, a distinct peak located at 970 cm−1, which is attributed to phosphorylated proteins and nucleic acids23,38,39,40, was easily noticed in the FTIR spectra at one-hour post-irradiation. These alterations were observed to be reversed at 3-hour post-irradiation. Photobiomodulation was previously reported to lead to alterations in phosphorylation of some specific proteins that aid in cell survival16,53,54.

Heat shock proteins (HSPs) are responsible of modulation of cellular homeostasis and their expression levels increase in stress conditions. Although low-level laser irradiation in photobiomodulation delivers low energy and does not result in an increase in heat in the exposed samples, photobiomodulation induces changes in cellular metabolism and ROS production in a small scale acting as a mild stressor, which is suggested to result in an increased viability and survival55. Exposure to 660 nm light emitting diode at 50 W/cm2 for 15, 30 and 60 min was reported to enhance Hsp expression in human fibroblasts56. A pre-exposure to 810 nm diode laser at fluencies of 34.27 J/cm2 and 44.88 J/cm2 was reported to promote cytoprotection in rat retina by resulting in an elevated Hsp70 expression and a reduction in the expression level of cleaved caspase-357. Our results indicated a slightly decreased Hsp70 protein expression level at 1-hour post-irradiation, which was elevated at 3-hour post-irradiation (Fig. 1d and e). This non-significant increase in Hsp70 protein expression level after 3 h may be a result of the protein phosphorylation and structural alterations observed at 1-hour post-irradiation, as Hsp70 was reported to play a role in protein repair56. Furthermore, studies have reported that alterations in membrane lipid composition, microheterogeneity and especially fluidity can trigger stress signaling responses through altered lipid-protein interactions and the formation of microdomains can modulate Hsp protein expression level49,58. Our finding of slightly increased Hsp70 protein expression level may be a consequence of the altered membrane lipid composition, order and fluidity. Future work is needed to confirm the relation in between photobiomodulation-induced membrane stress and heat shock response.

The metabolic turnover of carbohydrate resources in a cell can be monitored by the use of glucose/phosphate (A1035/A1079) and glucose/protein (A1035/A1643+1542) band area ratios in an FTIR spectrum26,45,46. An increase in the ratios implies the accumulation of carbohydrate resources, therefore a decreased metabolic turnover or activity. The results revealed a decreased metabolic activity at 3-hour post-irradiation, whereas at 1-hour post-irradiation, the metabolic turnover of the carbohydrate sources were more pronounced. This decreased utilization of carbohydrate resources may indicate the decrease in oxidative phosphorylation at 3-hour post-irradiation. Photobiomodulation via the use of red or near infrared light was reported to activate cytochrome c oxidase, a member of the mitochondrial respiratory chain16,47. Supporting our findings, Wang et al.47 have reported that photobiomodulation with 660 nm light resulted in a more pronounced ATP production in human ADSCs at lower doses (3 J/cm2), while higher doses (30 J/cm2) led to a progressively lower ATP production 3 h after irradiation. In addition, the mitochondrial membrane potential of human ADSCs was reported to be significantly increased with respect to control 1-hour after photobiomodulation with 660 nm light at 3 J/cm247.

The data presented collectively reveals the short-term alterations in composition, structure and function of biomolecules in ADSCs exposed to 671 nm photobiomodulation. This study is first to report the spectral characterization of the initial response of ADSCs exposed to photobiomodulation. However, there are some limitations of the study. This current study could be regarded as a preliminary work, as the effects of continuous low-level laser irradiation only at 671 nm at a fixed dose was considered. In addition, the study only concentrated to alterations at 1-hour and 3-hour post-irradiation. Therefore, how varying doses or different time intervals might affect the stated alterations needs further exploration. Furthermore, the sample size is relatively limited. It should be noted that this current study has been mainly focused on the spectral characterization of structural, compositional and dynamic alterations in biomolecules induced by photobiomodulation, and the expression level of only one protein was assessed. Future exploration related to the modulation of molecular signaling pathways by low-level laser irradiation, such as pathways associated with cell survival, proliferation and phosphorylated protein response, is crucial in order to confirm and justify the presented results.

Conclusion

This study comprehensively examined the short-term compositional and structural alterations induced by photobiomodulation on ADSCs using FTIR spectroscopy. Photobiomodulation initially triggered an immediate stress response, and over time it led to significant changes in lipid composition, protein phosphorylation and protein secondary structure, membrane order and dynamics with a temporary lateral phase separation, and alterations in utilization of carbohydrate sources in ADSCs. These findings may further implicate the activation of cellular adaptation mechanisms in ADSCs upon exposure to low-level laser irradiation, leading to the increased viability or “hormesis” in cells. The outcomes of the current study are further valuable in providing new insights into the mechanism of action of photobiomodulation and enhancing the understanding of its therapeutic potential for future applications.

Methods

Cell culture

This study is performed in accordance with relevant guidelines and regulations. All methods are reported in accordance with ARRIVE guidelines. In this current study, we have used the ADSCs that were isolated from an adult male Wistar-albino rat originating from the same primary source as reported in our earlier work59. The applied procedures for ADSC isolation were approved by the Ethics Committee of Aydın Adnan Menderes University (approval number 64583101/2017/087), and the use of archived ADSCs previously isolated and characterized59 was also approved by the Ethics Committee of Aydın Adnan Menderes University (approval number 64583101/2024/59). The ADSCs were transferred into flasks containing cell culture media (α-MEM + 5% L-glutamine). The cells were cultured in media supplemented with 5% Penicillin-Streptomycin and 10% FBS (Sigma-Aldrich, ABD) and incubated at 37 °C with 5% CO2, with media changes occurring every three days. Once the cells reached 80% confluence, they were passaged into new flasks. Following the isolation and expansion at the third passage, the ADSCs were divided into three study groups: a control group, a group that underwent low-level laser irradiation and was harvested one-hour post-irradiation (1 H post-L), and a group that was irradiated and harvested three hours’ post-irradiation (3 H post-L).

Low-level laser irradiation

After characterization, third-passage cells were plated in a 24-well plate at a density of 1 × 105 cells per well. The next day, each well was exposed to a 671 nm wavelength laser (MRL-FN-671, Thorlabs, China), with the laser output recorded at 243 mW at the experimental setup, as shown in Fig. 1b.

In order to identify most suitable dose of laser irradiation to observe the cellular responses, cells were irradiated with laser fluences at 1, 5, 10, 25, and 50 J/cm². This effective fluence range of 1–50 J/cm2 was selected based on the literature that apply photobiomodulation on mesenchymal stem cells4,60,61. To achieve target fluences of 1, 5, 10, 25, and 50 J/cm2, irradiation times were calculated using the formula D = P × t/A, where D is the desired fluence (J/cm2), P is the laser output (W), t is the exposure time (seconds), and A is the irradiated area (cm2), which corresponds approximately to the well surface area (1.9 cm2 per well in a 24-well plate). Initially, cells were exposed to these different fluences to identify the most effective dose for promoting proliferation at 24 hours’ post-irradiation. In order to assess this, MTT test was applied. Cells were plated in 24-well plates at a density of 5 × 104 cells per well using DMEM. After 24 h of incubation, the medium was replaced with serum-free and phenol red-free DMEM. The wells were then irradiated with a 671 nm laser at doses of 1, 5, 10, 25, or 50 J/cm² for varying durations. Following irradiation, the complete cell medium was added, and the cells were incubated for an additional 24 h. The next day, the cells were treated with DMEM containing 0.5 mg/mL MTT for 2 hours. The resulting formazan crystals were dissolved in 500 µL of DMSO, and the absorbance was measured at 457 nm using a plate reader.

Based on these findings, the fluence that exhibited the most significant biological effect was chosen for further investigation. Subsequently, to evaluate the short-term cellular responses to photobiomodulation, cells were irradiated with the selected fluence, and analyses were conducted at 1 h and 3 h following irradiation. The experiments were repeated twice, in each repetition three different wells were used and scanned as separate samples. For each experimental replicate, all groups were seeded and grown in a single 24-well plate, with one row of three wells designated as control, the subsequent row as 1 H post-L, and the other row as 3 H post-L groups. Therefore, in each experimental group, there were a total of 6 biological replicates of ADSCs that were investigated via FTIR spectroscopy.

Determination of Hsp70 protein expression level

Determination of the protein expression levels were performed as described previously62. ADSCs were lysed in RIPA lysis buffer (Merck Millipore, Darmstadt, Germany) containing a 1:10 protease inhibitor cocktail (Sigma-Aldrich, USA) and then centrifuged at 15,000 g for 10 min at 4 °C. The supernatants were combined with a sample buffer that included 0.5 M Tris-HCl (pH 6.8), glycerol, 10% SDS, bromophenol blue, and β-mercaptoethanol, and then boiled at 95 °C for 5 min. The protein concentration in the lysates was measured using a bicinchoninic acid protein assay kit (ThermoFisher Scientific, USA). An equal amount of protein (20 µg) from each group in triplicates was loaded onto a polyacrylamide gel (with 4% stacking and 12% resolving gels), subjected to gel electrophoresis, and transferred to a polyvinylidene fluoride membrane (Millipore, Bedford, MA, USA). After blocking, the membrane was incubated overnight at 4 °C with primary antibodies against Hsp70 (1:1000, Bioss, USA) and GAPDH (1:5000, Bioss, USA), followed by washing. The secondary antibody (1:2000 anti-mouse IgG HRP, Abcam, UK) was then applied at room temperature for 2 h, and ECL substrate (BIO-RAD, California, USA) was added after washing. Finally, the membrane was visualized using the UVP system (LAB Denmark APS, Denmark) with a consistent exposure time of 5 min. Protein band intensities were quantified using ImageJ (MD, USA, 64-bit version) software.

ATR-FTIR spectroscopic data collection and data analysis

FTIR spectra of ADSCs subjected to photobiomodulation at different time intervals (control, 1 H and 3 H post- irradiation) was collected by Perkin Elmer Spectrum-Two FTIR spectrometer (Perkin Elmer, USA) coupled with an attenuated total reflection (ATR) unit containing a diamond crystal (Perkin Elmer, USA). ADSCs were trypsinized, collected and washed twice with HBSS to prevent environmental contamination, and finally were re-suspended in HBSS at 2 × 106 cells/ml. Then, the cells were centrifuged at 1200 rpm for 3 min to remove the supernatant. Finally, the resultant cell suspension was placed on the ATR diamond crystal surface and dried under mild nitrogen flow at room temperature for 10 min to remove excess water. The spectra were collected within the spectral range of 4000–450 cm−1; 100 scans were collected for each interferogram at a resolution of 2 cm−1. A background scan was taken prior to the experiment, in order to prevent the interference of atmospheric water vapor and carbon dioxide. The spectra of three replicates from each sample was collected, all of which yielded similar spectra, and their average was used for data analysis. The collection of spectral data and their analysis was performed via Perkin Elmer Spectrum software (version 10.5.2, Perkin Elmer, USA). Band areas were calculated, and band positions were determined based on the center of weight at 75% of peak height, confirmed through the second derivatives of the peaks. The bandwidth values of specific absorption peaks were measured as the width corresponding to 75% band intensity. Spectra were baseline corrected and normalized relative to the Amide A band located at 3282 cm−1 for visual demonstration.

In order to determine the intensities associated with different protein secondary structures, the band intensities in second derivative spectra in 1700–1600 cm−1 region was analyzed using OPUS 8.5 software (Bruker Optik GmbH, Germany). The second derivative spectra were generated using a 9- point Savitzky-Golay smoothing algorithm. Subsequently, the spectra were vector normalized in the 1700–1600 cm−1 region, and the corresponding peak intensities at the minima were calculated by OPUS 8.5 software.

Statistical analysis

Data were expressed as mean ± standard error of mean. Statistical analysis was performed using the GraphPad Prism 10.5.0 software (La Jolla, CA, USA). First, the data was analyzed via Anderson Darling normality test, which revealed that data follow a normal distribution. Therefore, statistical analysis was performed via one-way ANOVA together with Tukey’s test applied as a post-hoc test. The results of MTT and Hsp70 protein expression levels were also evaluated via one-way ANOVA with Dunn’s test as a post-hoc test. A p value of less than 0.05 was considered as statistically significant. The degree of significance of the post-hoc test was denoted by * for the comparison of control with the other groups, by ‡ for the comparison of 1 H post-L group with 3 H post-L group (*p < 0.05, **p < 0.01, ***p < 0.001).

Data Availability

The datasets generated in the current study are available from the corresponding author on request.