Molecular pathway of near-infrared laser phototoxicity involves ATF-4 orchestrated ER stress

Abstract

High power lasers are used extensively in medicine while lower power applications are popular for optical imaging, optogenetics, skin rejuvenation and a therapeutic modality termed photobiomodulation (PBM). This study addresses the therapeutic dose limits, biological safety and molecular pathway of near-infrared (NIR) laser phototoxicity. Increased erythema and tissue damage were noted in mice skin and cytotoxicity in cell cultures at phototoxic laser doses involving generation of reactive oxygen species (ROS) coupled with a rise in surface temperature (>45 °C). NIR laser phototoxicity results from Activating Transcription Factor-4 (ATF-4) mediated endoplasmic reticulum stress and autophagy. Neutralizations of heat or ROS and overexpressing ATF-4 were noted to rescue NIR laser phototoxicity. Further, NIR laser mediated phototoxicity was noted to be non-genotoxic and non-mutagenic. This study outlines the mechanism of NIR laser phototoxicity and the utility of monitoring surface temperature and ATF4 expression as potential biomarkers to develop safe and effective clinical applications.

Figure 1. Laser induces phototoxicity in vivo and in vitro.

Dorsal skin of 5-week-old C57BL/6NCr male mice were shaved and naired and were dynamically treated with 3.2 W laser to maintain (a) 45 °C for 10 s, (b) 45 °C for 30 s and (c) 55 °C for 30 s. The inset panels show (upper right) H&E staining and (lower right) skin surface temperature as monitored by the thermal camera and (lower left) higher magnification of clinical image. Scale bars = 70 μm (d) Surface temperature profile of mice treated with laser at 45 °C for 30 s as measured by the thermal camera (n = 13). (e) Damaged skin area was measured a day after laser treatment (45 °C for 30 s) and metal probe (n = 13 each). (f) Erythema scores measured immediately after treatment and 24 hours after phototoxic laser dose treatments. Significance based on one way ANOVA with the respective controls (n = 13). Laser treatments were performed in clear (g) and black (h) well plates on HaCaT and NOKSI cells and cellular viability was assessed 24 hrs after treatment with AlamarBlue as quantitated with a plate reader. Significance was determined using two-way analysis of variance (ANOVA) among different treatments using the Bonferroni’s multiple comparison test (n = 3). Statistical significance is denoted as P values <0.001 (**) and <0.00001 (****).

Figure 2. Laser induced phototoxicity is mediated by heat and ROS.

Surface temperature of cells treated with laser as assessed by IR camera (n = 3) (a) is shown. (b) ROS generation was assessed with DCFDA by FACS analyses following laser treatments in clear and black well plates (n = 3). In vitro activity of GR (c) and Catalase (d) were assessed in HaCaT and NOKSI cells treated with phototoxic laser doses (n = 3). (e) HaCaT and NOKSI cells were pre incubated with NAC (1mM), Catalase (1000 units/ml) or Helium (bubbled for 5 minute) and treated with phototoxic laser dose followed by cell viability assays (n = 3). Significance was determined using two-way ANOVA and Bonferroni’s multiple comparison tests. Phototoxic laser doses are highlighted in gray font. Statistical significance are indicated as P < 0.05 (*), <0.001 (**), <0.0001 (***) and <0.00001(****).

Figure 3. Laser does not cause DNA damage (Non-genotoxic).

Plasmid cleavage assay was performed using PUC19 and following laser treatments with varying doses (a) and wavelengths (b) plasmid was analyzed by gel electrophoresis on 1% agarose gel that was quantitated by densitometry (n = 3). HaCaT (c) cells were treated with different doses of laser and genomic DNA was assayed for number of abasic sites. UV treatment was used as a positive control, Significance was based on one-way ANOVA with the respective controls (n = 3). (d and e) Ames test was performed on TA100 and TA98 strains of the Salmonella typhimurium using different doses of laser and revertants were quantitated on 5^th day after treatment. Significance was noted as per the manufacturer’s manual (n = 3). HaCaT cells (f) (Scale bars = 200 μm) or mice (g) were treated with phototoxic laser dose and γ-H2AX immunostaining was performed to assess DNA damage. γ-radiation (10 Gy) was used as positive control. Scale bars = 70 μm. (h) HaCaT cells were treated with radiation (10 Gy) and sub-phototoxic laser doses and PCR arrays were performed. Representative scatter plot of differentially regulated genes (fold change ≥ 2) are shown. Detailed list of genes are available in supporting materials (n = 2). Statistical significance are indicated as P < 0.0005 (***) and <0.00001(****).

Figure 4. Laser induces ER stress at sub-phototoxic doses.

(a) ER stress markers after varying doses of laser treatment were assessed at 24 hrs using immunoblotting. (b) Localization of ATF 4 in HaCaT cells were assessed after 6 hrs following laser treatment. Scale bars = 200 μm. (c) Level of autophagy (LC3A/B-ii/i), ATF-4 and HSP70 were assessed by immunoblotting after 24 hrs of laser treatment. HaCaT cells were pretreated with HSP inhibitors (d) autophagy inhibitors (e) or pre-treated with LPS (2 ng/ml), PMA (1X) and TMFC (Tamoxifen Citrate 40 μM) (f) for 2 hrs and were then challenged with laser treatment and cell viability was assessed at 24 hrs. Significance based on one-way ANOVA with the respective controls (n = 3). HaCaT and NOKSI cells were pretreated with varying concentrations of Rapamycin and challenged with phototoxic dose followed by the assessment of cellular viability (g) and immunoblotting (h) Images in a gel are cropped (horizontally) and placed together for better clarity of results and were run in the same experimental conditions. Significance was determined using two-way ANOVA among different treatments using the Bonferroni’s multiple comparison test. Statistical significance are indicated as P < 0.0005 (***), <0.00001(****) and not significant (n.s.).

Figure 5. ATF-4 protects cells from cellular stress.

HaCaT cells were transfected with ATF-4 siRNA and were treated with different laser doses. Cellular viability (a) (n = 3) was assessed at 24 hrs. (b) Conversely, over expressing ATF-4 stable HaCaT cells was treated with increasing doses of laser and cellular viability was assessed. Statistical significance was determined using two-way ANOVA among different treatments using the Bonferroni’s multiple comparison test (n = 3). (c) Outline shows HaCaT cells treated with phototoxic dose of laser that generates two populations, surviving cells that remain adherent (A) and dying or dead cells that float (F) in media. These were collected separately, lysed and immunoblotting (d) was performed to assess ER stress pathway. Histological assessment of mice skin showing damaged area in laser-treated skin was performed by immunohistochemistry for ATF-4 (e,f), HSP70 (g,h) and (i,j) TUNEL positivity. (Scale bar = 70 μm). Quantitation of high power images from mice sections are plotted (n = 5). Insets show high power magnifications. Images in a gel are cropped (horizontally) and placed together for better clarity of results and were run in the same experimental conditions. Significance was assessed with paired Student’s t-test. Statistical significance are indicated as P < 0.05 (*), <0.001 (**), <0.0001 (***) and <0.00001(****).

Figure 6. Laser mediated damage is via heat and ROS in vivo.

(a) Images of the dorsal portion of 5-week-old C57BL/6NCr male mice treated with laser (55 °C) or following skin cooling for 30 s. Quantitation of damaged area (b) and erythema score (c) are shown (n = 5). (d and e) Representative image of Reactive Oxygen Species (ROS) induction as measured with ROSstar probe using IVIS in vivo imaging is shown (Radiation is used as a positive control). Treatment sites were pretreated with NAC or NAA and ROS level was quantitated after laser treatment (f). Dorsal area was photographed (g) and damaged area (h) and erythema (i) are shown (n = 5). (J-m) Laser treated mice skin tissues from these experiments were assessed with immunostaining for ATF-4 and HSP70 expression and their quantitation was performed with ImageJ (n = 5). Significance was assessed with paired Student’s t-test and denoted as P < 0.05 (*), <0.001 (**), <0.0001 (***), <0.00001(****) and not significant (n.s.).