Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jul 1.
Published in final edited form as: J Invest Dermatol. 2023 Jan 25;143(7):1279–1288.e9. doi: 10.1016/j.jid.2022.12.018

Topical photodynamic therapy generates bioactive microvesicle particles: evidence for a novel pathway involved in immunosuppressive effects

Oladayo A Oyebanji 1,#, Chad Brewer 1,#, Sharlo Bayless 1, Benjamin Schmeusser 1, Danielle A Corbin 1, Courtney E W Sulentic 1, Catherine MT Sherwin 1,3,4, Yanfang Chen 1, Christine M Rapp 1, Elizabeth E Cates 1, Yuhan Long 4, Jeffrey B Travers 1,2,5,*, Craig A Rohan 1,2,5
PMCID: PMC10293022  NIHMSID: NIHMS1870133  PMID: 36708950

Abstract

Though effective in treating actinic damage, topical photodynamic therapy (PDT) has been demonstrated to be immunosuppressive through unknown mechanisms, which could potentially limit its effectiveness. Multiple types of environmental stressors including PDT can produce the immunosuppressive lipid mediator Platelet-activating Factor (PAF). As PAF can produce subcellular microvesicle particles (MVP), the present studies tested if PDT can generate PAF and MVP release, and if these are involved in PDT-induced immunosuppression. Previously, topical PDT using blue light and 5-aminolevulinic acid was found to be a potent stimulus for PAF production in mice and human skin explants and human patients, and we show experimental PDT also generates high levels of MVP. PDT-generated MVP were independent of the PAF receptor (PAFR), but, were dependent upon the MVP-generating enzyme acid sphingomyelinase (aSMase). Patients undergoing topical PDT treatment to at least 10% of body surface area demonstrated local and systemic immunosuppression as measured by inhibition of delayed-type hypersensitivity reactions. Finally, using a murine model of contact hypersensitivity, PDT immunosuppression was blocked by genetic and pharmacologic inhibition of aSMase, and genetic inhibition of PAFR signaling. These studies describe a mechanism involving MVP by which PDT exerts immunomodulatory effects, providing a potential target to improve its effectiveness.

Introduction

Photodynamic therapy (PDT) involves the use of light at a specific wavelength acting on a photosensitizing chemical with the goal to cause cell death (Dougherty and Marcus, 1992). Topical PDT is FDA-approved for use as “field therapy” in treating relatively large areas of human skin to remove precancerous actinic keratoses (Babilas et al., 2006, Kwiatkowski et al., 2018). A commonly used protocol involves 5-aminolevulinic acid (5-ALA) pretreatment to skin to result in the generation of the highly reactive protoporphyrin IX (PpIX). Of importance, neoplastic cells tend to convert more 5-ALA into PpIX than their normal counterparts. Exposure to light at various wavelengths (e.g., blue at 410-420nm) then acts upon the PpIX resulting in high levels of reactive oxygen species (ROS) which then cause cell damage (Ochsner, 1997). Though effective, comparison studies of topical PDT with other modalities for treating large areas of actinic keratosis have revealed that topical 5-fluorouracil is approximately twice as efficacious as PDT (Gupta and Paquet, 2013, Jansen et al., 2019, Wehner, 2020).

Several lines of evidence suggest that topical PDT is immunosuppressive (Bayless et al., 2021, Mroz and Hamblin, 2011). Clinical evidence exists such as the well-known increased risk of herpes simplex viral outbreaks following PDT (Nobbe et al., 2011). There are also multiple reports, albeit some anecdotal in nature, of more aggressive skin cancers forming in previously PDT-treated skin (Fiechter et al., 2012, Wolf et al., 1997). Importantly, testing of delayed-type hypersensitivity (DTH) reactions to recall antigens in PDT-treated human skin has been shown to result in local immunosuppression (Frost et al., 2011, Matthews and Damian, 2010). Similarly, murine contact hypersensitivity (CHS) models of epicutaneously applied neoantigens have reported that topical PDT induces local (failure to sensitize at the site of PDT) as well as systemic (failure to sensitize at a non-PDT-exposed site) immunosuppression (Elmets and Bowen, 1986, Ferracini et al., 2015, Hayami et al., 2007). Yet the exact mechanism(s) by which a cutaneous insult can induce these immunosuppressive effects is not clear.

Microvesicle particles (MVP) are defined as submicron vesicles generated from cellular membranes via exocytosis. They are irregularly shaped and range from ~100-1000 nm in size (Bazzan et al., 2021, Chang et al., 2021). The generation of MVP occurs in response to numerous stimuli, most involving intracellular calcium mobilization responses. Once released into the circulation (blood and tissue), these subcellular particles are thought to be able to target other cell types through surface membrane interactions that then allow fusion of the MVP to the target cell membrane and subsequent delivery of the intravesicular contents. The pathways by which cells release MVP have been characterized, and are cellular- and stimulus-dependent (Norling and Dalli, 2013). Most MVP-generating stimuli involve activation and translocation of the enzyme acid sphingomyelinase (aSMase) to the plasma membrane (Bianco et al., 2009, Liu et al., 2021).

The functions of MVP in physiological and pathological processes are dependent upon their carried contents including proteins, lipids and nucleic acids. Metabolically labile compounds inside MVP can be somewhat protected from enzymatic degradation. Circulating MVP are increased in inflammatory skin diseases such as psoriasis, in which many cytokines are also elevated (Liu et al., 2016, Pelletier et al., 2011, Takeshita et al., 2014, Tamagawa-Mineoka et al., 2010). Elevated levels of MVP and smaller exosomes have been reported in vascular diseases and autoimmune disorders such as lupus erythematosus and dermatomyositis which correlated with the severity of disease, suggesting that these agents could serve as biomarkers (Beyer and Pisetsky, 2010, Herring et al., 2013, Liu et al., 2016, Li et al., 2021; Takeshita et al., 2014).

Platelet-activating Factor (PAF) is an activity which resides in sn-1 ether-linked glycerophosphocholines (GPC) with a short-chained fatty acid (usually acetate) at the sn-2 position (Travers, 2020). PAF is synthesized enzymatically using two distinct biochemical pathways (Nakanishi et al., 2006, Shimizu, 2009). The remodeling pathway associated with cellular stimulation from a variety of sources, consists of phospholipase A2 (often cytosolic PLA2) followed by an acetyl-CoA-dependent acetyltransferase (LPCAT) which attaches an acetate moiety to a lyso-PAF intermediate to generate PAF. PAF agonists are also produced non-enzymatically via free radical-mediated attack of sn-2 polyunsaturated fatty acids (often arachidonate) on the GPC resulting in oxidized GPC (Ox-GPC) (Travers, 2020). Once generated, PAF and Ox-GPC are rapidly metabolized to inactive lyso-GPC via acetylhydrolases (McIntyre et al., 2009). Exerting its effects by a single G-protein coupled receptor (the PAFR) that is widely expressed, PAF is involved in numerous processes ranging from acute inflammatory and allergic proinflammatory responses to serving as a potent mediator of delayed immunosuppression (Shimizu, 2009, Travers, 2020, Walterscheid et al., 2002). PAF and Ox-GPC have been demonstrated to be produced in skin in response to multiple environmental stressors including thermal injury, UVB, PDT, and topical chemotherapy (Ferracini et al., 2015, Marathe et al., 2005). As previously reported using mass spectrometric analysis of GPC in HaCaT keratinocytes, PDT generates multiple sn-2 GPC that are associated with enzymatic production, yet Ox-GPC species produced in response to reactive oxygen species-mediated attack of parent GPC were not identified (Ferracini et al., 2015). Of interest, PDT-induced systemic immunosuppressive effects, like those of UVB, are absent in PAFR-deficient mice (Ferracini et al., 2015).

Studies by our group have demonstrated that activation of the keratinocyte PAFR results in MVP release (Bihl et al., 2016, Liu et al., 2021, Liu et al., 2020). Additionally, UVB radiation, which generates PAF agonists (Marathe et al., 2005), also induces MVP in a PAFR-dependent manner (Liu et al., 2021). Use of aSMase-deficient mice and pharmacologic approaches has demonstrated that UVB-induced MVP mediate systemic immunosuppression (Liu et al., 2021). In contrast, thermal burn injury, which is a potent stimulus for PAF synthesis, generates MVP through both PAFR-dependent and -independent processes (Liu et al., 2020). Similarly, exposure to cells in vitro or topical application of protein kinase C activators including the phorbol ester tetradecanoyl-phorbol acetate (TPA) induces MVP release in a PAFR-independent manner (Liu et al., 2021, Thapa et al., 2022). The goal of the present studies is to assess if PDT can generate PAF and MVP in skin tissues, and ascertain whether these mediators are involved in the immunosuppressive effects of PDT.

Results

PDT generates PAF agonists.

Our first studies built upon our previous report that PDT generates PAF (Ferracini et al., 2015), to examine the ability of topical PDT to generate PAF in human skin explants, mice, and in human subjects undergoing clinical treatment for actinic damage. To measure PAF, we used a validated biochemical assay (see Figure 1a) which has the ability to quantify all PAFR agonistic activity from lipid extracts of treated cells/tissue through measuring IL-8 release from PAFR-positive KBP cells (Liu et al., 2021, Pei et al., 1998). As shown in Figure 1b, topical PDT resulted in increased PAFR agonistic activity in human skin explants by 5 minutes which was maximal by 10 minutes. Higher fluences of blue light combined with 5-ALA resulted in more PAF generation. In contrast, PDT to SKH-1 mice resulted in a much more immediate and long-lived (up to 30 min) PAF-generating responses. It should be noted that PDT is a much more potent PAF-producing stimulus than high UVB fluences (Figure 1c) in these mice. Using a 10 minute post-PDT time point, we next tested if can measure increased PAF following topical PDT under clinical conditions. Comparing the lipid extracts from 2 mm punch biopsies obtained on PDT-treated dorsal forearm skin vs untreated (yet covered) lateral upper arm skin 10 minutes following PDT, nine of the ten subjects tested displayed increased PAF activity in the PDT-treated skin (Fig 1d). These studies indicate that topical PDT generates PAF in both human and murine skin.

Figure 1. PDT generates PAF in HaCaT keratinocytes, human skin explants, mice and patients undergoing clinical treatment.

Figure 1.

Open in a new tab

a) Model of KBP-IL-8 assay where a complex mixture (lipid extracts) containing PAF agonists are incubated with PAF-R-positive KBP cells and IL-8 released used as surrogate of PAF-R activation, and compared to IL-8 release in response to CPAF treatment. b) Human skin explants were treated with 5-ALA for 4 hours, followed by exposure to 10 or 20J/cm2 blue light or blue light alone (without 5-ALA). At various times post-treatment, duplicate 6 mm punch biopsies were obtained, weighed, lipids extracted, and tested on KBP cells. The data are the mean ± SE % response compared to 10 nM CPAF from three experiments. c) SKH-1 mice were treated with either PDT using 10J/cm2 blue light, 2250 or 4500J/m2 UVB, or sham. At various times post treatment, groups of at least five mice were euthanized, and duplicate 6 mm punch biopsies obtained, weighed, and lipids extracted. The data are the mean ± SE % response compared to 10 nM CPAF. d) Human subjects undergoing PDT with 10J/cm2 BLU-U light source to forearms underwent a single 2 mm punch biopsy to PDT-treated forearm and a second 2 mm punch biopsy to untreated (covered) dorsal inner upper arm skin. The biopsies were weighed and lipids extracted. The data also include the mean ± SE from the 10 subjects. *Denotes statistically significant (P<0.05) changes in levels of PAF-R agonistic activity from sham control values.

PDT generates MVP independent of the PAFR.

As PAFR activation can generate MVP (Bihl et al., 2016, Liu et al., 2021), our next studies assessed whether experimental PDT can generate MVP release. For these studies, we employed the human keratinocyte-derived cell line HaCaT (Boukamp et al., 1988). As shown in Figure 2a,b, PDT resulted in the release of MVP in a dose- and time-dependent manner in the HaCaT keratinocyte cell line. These subcellular particles exhibited size greater than 100nm consistent with MVP (Supplemental Figure S1). Of note, maximal levels of MVP were noted by 4 hours, and did not increase further. It should be noted that PDT resulted in considerable toxicity, with only ~20% of the cells viable at 4h following 20 J/cm2 blue light (Supplementary Figure S2). PDT also generated MVP in telomerase-immortalized, diploid neonatal keratinocytes N/TERT keratinocytes (Supplementary Figure S3) and primary cultures of human fibroblasts (Figure 2c). Treatment of human skin explants also revealed that PDT can trigger MVP release (Figure 2d).

Figure 2. PDT generates MVP release in HaCaT keratinocytes, fibroblasts and human skin explants.

Figure 2.

Open in a new tab

a) HaCaT keratinocytes were treated with sham (no treatment), 0.1% ethanol vehicle control, 100 nM TPA, 100 nM CPAF, 1 mM 5-ALA for 4 hours, or 5-ALA followed by various fluences (2.5-20 J/cm2) blue light or blue light alone. Levels of MVPs released into supernatants were measured at 4 hours. b) HaCaT keratinocytes were treated with sham or 5-ALA + 20 J/cm2 blue light PDT and supernatants removed at various times and MVP measured. c) Human fibroblasts were treated with similar agents as in a) and levels of MVPs released at 4 hours measured. d) Human skin explants were treated with sham, 100 μL vehicle (90% ethanol + 10% DMSO), 1 μM TPA, 1 μM CPAF, 1 mM 5-ALA for 4 hours, blue light or 5-ALA + blue light at 20J/cm2. After four hours, duplicate 6 mm punch biopsies were obtained, weighed, and MVPs were measured. The data are the mean ± SE from 3-4 separate experiments conducted in duplicate. *Denotes statistically significant (P < 0.05) changes from sham values.

Our previous studies have determined that MVP generated by UVB or lower levels of thermal burn injury are dependent upon PAFR activation, yet high levels of thermal burn injury contain a PAFR-independent component (Liu et al., 2020). To ascertain the role of the PAFR in PDT-generated MVP, we used two separate approaches. First, we subjected PAFR-positive KBP and PAFR-negative KBM epidermoid cells (Pei et al., 1998) to experimental PDT. As shown in Figure 3a, KBP cells generated increased MVP in response to phorbol ester TPA and PAFR agonist carbamoyl PAF (CPAF). As expected, PAFR-negative KBM cells reacted to TPA but not CPAF. PDT at low (10J/cm2) and higher (20J/cm2) fluences appeared equally effective in generating MVP in KBM and KBP, suggesting that the PAFR was not necessary for MVP generation. Experimental PDT at high fluences was more effective in generating MVP than the PAFR agonist CPAF or the phorbol ester TPA. To similarly assess the role of the PAFR in topical PDT-generated MVP in vivo, we employed wild-type and PAFR-deficient (Ptafr−/−) mice. Depicted in Figure 3b, topical PDT generated approximately similar levels of increased MVP in both skin and plasma in both genotypes. These studies suggest that experimental PDT induces high levels of MVP in a process which appears independent of the PAFR.

Figure 3. The role of the PAFR in PDT-mediated MVP release.

Figure 3.

Open in a new tab

a) PAFR-positive KBP cells and PAFR-negative KBM cells were treated with sham, 0.1% ethanol vehicle, 100 nM TPA, 100 nM CPAF, 5-ALA alone for 4 hours, 20 J/cm2 blue light, or the combination (PDT-10J or 20J). Four hours later the supernatants were removed, and cells counted. The data are the mean ± SE MVP normalized to cell numbers from four separate experiments. b) Groups of 5-7 wild-type or PAFR-deficient (Ptafr−/−; PAFR KO) mice were treated with topical CPAF (250 ng in 100μL), or a four hour treatment with 100μL 5-ALA followed by 20 J/cm2 blue light (PDT). At 4 hours post-treatment, the mice were euthanized and MVP were measured from duplicate 6 mm punch biopsies and plasma. The data are mean ± SD MVP normalized to tissue weight/plasma volume. *Denotes statistically significant (P < 0.05) changes from Sham; NS, not statistically significant (P >0.05) changes from comparator denoted by line.

PDT-generated MVP are due to aSMase.

Stimuli-generated MVP often involve aSMase. For example, in keratinocytes, skin explants and mice, MVP produced in response to PAFR agonists, UVB, TPA, thermal burn injury but not C2 ceramide are dependent upon aSMase (Awoyemi et al., 2022, Liu et al., 2021, Liu et al., 2020). To define if PDT-induced MVP involved aSMase, two approaches were used. First, the pharmacologic aSMase inhibitor imipramine (Bianco et al., 2009, Liangpunsakul et al., 2012, Liu et al., 2021) was tested for its effects on PDT-induced MVP release. As shown in Figure 4a,b, imipramine treatment immediately post-PDT blocked MVP release in HaCaT keratinocytes as well as in human skin explants. Similarly, imipramine blocked PDT-induced MVP in N/TERT keratinocytes and human primary cultures of fibroblasts (Supplementary Figures S3, S4). Treatment of wild-type C57BL/6 mice with imipramine also decreased the MVP response from PDT in both skin and plasma (Figure 4c). The second approach using aSMase-deficient (Smpd1−/−) mice also demonstrated that this enzyme was necessary for MVP production in response to topical PDT (Figure 4d). These studies indicate that PDT-mediated MVP release is dependent upon the aSMase enzyme.

Figure 4. The role of aSMase in PDT-mediated MVP release.

Figure 4.

Open in a new tab

a) HaCaT keratinocytes were treated as in Figure 2, with the exception that in some experiments 50 μM of the aSMase inhibitor imipramine (IMIP) was added immediately after the PDT. Supernatants were collected at 4 hours and MVP measured. The data are the mean ± SE MVP fold change from sham-treated values. b) Human skin explants were treated as in Figure 2 with the exception that in some experiments 100 μL of 5mM IMIP was added post-PDT. MVP were measured in duplicate 6 mm punch biopsies obtained 4 hours post-treatment. The data are the mean ± SE MVP fold change from sham-treated values. c,d) murine studies. c) Wild-type mice were treated with either sham, 100 μL of 5mM IMIP, 1uM TPA, or PDT using 20 J/cm2 blue light. d) Wild-type or aSMase-deficient (Smpd1−/−) mice underwent sham or PDT using 20 J/cm2 blue light. At 4 hours post-treatment, the mice in c,d) were euthanized and MVP were measured from duplicate 6 mm punch biopsies and plasma. The data are mean ± SD fold increase in MVP normalized to tissue weight/plasma volume. *Denotes statistically significant (P < 0.05) changes from sham; # statistically significant (P < 0.05) changes from comparator denoted by line by two-way ANOVA; NS not statistically significant.

PDT results in both local and systemic immunosuppression.

Topical PDT has been demonstrated to be immunosuppressive in both mice and humans (Bayless et al., 2021). To test the immunosuppressive effects of standard 5-ALA-PDT used in our clinics, we enrolled volunteers with extensive actinic damage needing PDT to at least 10% body surface area (forearms + face/scalp) in a clinical study (see Consort Diagram in Figure S4). As outlined in Methods and Figure 5a, subjects underwent intradermal testing with candida and trichophyton antigens before and 7 days post-PDT to treated forearm vs untreated volar forearm. The induration of the skin reactions at 48 hours post-injection were measured and compared with pre-PDT treatment values. Two of the 11 subjects enrolled in this study did not initially react to either antigen (as compared to saline control) and thus did not undergo repeat DTH testing. As noted in Figure 5a, Supplementary Figure S6 and Supplementary Table S1, PDT resulted in decreased skin reactions to candida antigen testing. Of note, the local immunosuppression (at the site of PDT) was greater than the systemic immunosuppressive (non-PDT-treated volar forearm) responses. Though trichophyton responses were also measured in this population, the numbers of positive subjects were much smaller (three of the nine subjects), yet the trends appeared similar to the candida antigen testing (see Supplementary Table S2). These studies indicate that topical PDT as practiced in a clinical setting to at least 10% body surface area is immunosuppressive as denoted by its ability to diminish recall DTH reactions at both the treated as well as untreated skin.

Figure 5. Effect of PDT on human DTH skin reactions to Candida antigen.

Figure 5.

Open in a new tab

a) PDT testing protocol. Nine subjects undergoing PDT with 10 J/cm2 BLU-U to 10% body surface area (forearms + face/scalp) underwent skin testing with Candida or trichophyton antigens before and 7 days post-PDT, and 48 h skin reactions (area of induration) on PDT-treated dorsal forearms (local immunosuppression) vs untreated volar forearms (systemic immunosuppression) were compared with initial pre-treatment values. b) % Pre-treatment values for 9 subjects who underwent Candida antigen testing. The data are mean ± SD values. *Denotes statistically significant (*P < 0.05; **P < 0.01) changes from pre-treatment values using Wilcoxon Rank Sum Test.

To define the roles of the PAFR and aSMase in PDT-induced immunosuppressive effects, the next studies tested topical imipramine post-PDT in wild-type mice, and also employed PAFR- and aSMase-deficient mice. As outlined in Figure 6a, topical PDT resulted in both local and systemic immunosuppressive effects in wild-type mice using our CHS models employing the neoantigen dinitrofluorobenzene (DNFB). Topical application of imipramine immediately following PDT blocked both the local and systemic immunosuppressive effects of PDT. However, topical imipramine did not affect the inhibition of DNFB sensitization/elicitation CHS reactions in response to systemic CPAF treatment. Consistent with our findings using a topical aSMase inhibitor, PDT of aSMase-deficient mice did not result in decreased systemic and local DNFB responses (Figure 6b). Yet, treatment of mice lacking aSMase with intraperitoneal injection of CPAF resulted in systemic immunosuppressive CHS responses. As previously reported (Ferracini et al., 2015), PDT-induced systemic immunosuppression was totally eliminated in PAFR-deficient hosts. Surprisingly, in contrast to local immunosuppression from UVB (Sahu et al., 2012) we noted statistically significant diminished PDT local responses in the Ptafr−/− mice. Systemic histamine was used as a positive control for systemic immunosuppression in these mice (Liu et al., 2021, Zhang et al., 2008). These studies suggest that PAF and MVP are critical effectors for PDT-mediated immunomodulatory effects.

Figure 6. the roles of the PAFR and aSMase in PDT-mediated inhibition of murine CHS reactions.

Figure 6.

Open in a new tab

a) Wild-type mice were either injected with 250 ng CPAF i.p., or underwent PDT using 10 J/cm2 blue light, or no treatment (sham). Immediately following the above, the back skin which underwent PDT was treated with 10uL of 4% imipramine or DMSO:PEG vehicle. Five days later the mice underwent sensitization with DNFB in PDT-treated (local immunosuppression) or non-PDT-treated skin (systemic immunosuppression). Nine days later the mice underwent elicitation reactions to ears with ear thickness measured at 24 hours. b) Wild-type, aSMase-deficient (Smpd1−/−), or PAFR-deficient (Ptafr−/−) mice underwent treatment with CPAF, 200 μg histamine, or PDT followed by DNFB sensitization/elicitation reactions as outlined above. The data are mean ± SD values. *Denotes statistically significant (P < 0.05) changes from sham treatment values using two-way ANOVA.

Discussion

Though compelling evidence indicating that topical PDT is immunosuppressive exists, the exact mechanism(s) are as yet unclear (Bayless et al., 2021). The present studies examined the role of the PAF and MVP pathways in PDT-induced immunosuppression. Use of multiple approaches from in vitro cells to actual human subjects demonstrates that PDT is a potent stimulus for PAF synthesis. These studies confirm previous in vitro findings using HaCaT keratinocytes (Ferracini et al., 2015). The present work using cells and mice with/out PAFR also provides evidence that PDT generates MVP in a PAFR-independent manner. PDT is also effective in generating MVP in human dermal fibroblasts which lack PAFRs (Figure 2c), providing further evidence that the PAFR is not needed to generate MVP in response to this stimulus. This lack of involvement of the PAFR in PDT-induced MVP release is in contrast to the necessary role of PAFR signaling in UVB-mediated MVP release (Bihl et al., 2016, Liu et al., 2021).

PDT is a potent inducer of cellular ROS (Dougherty and Marcus, 1992, Ochsner, 1997). Our previous studies using HaCaT keratinocytes reported that antioxidants N-acetyl cysteine and ascorbic acid did not affect the ability of experimental PDT to generate PAF (Ferracini et al., 2015). These same antioxidants blocked PDT-induced MVP release in HaCaT cells (Supplementary Figure S7), indicating that PDT-induced PAF and MVP generation differ in that PAF is not dependent upon ROS, yet MVP production is due to the pro-oxidative effects of PDT.

Consistent with previously reported studies (Elmets and Bowen, 1986, Ferracini et al., 2015, Hayami et al., 2007), topical PDT resulted in an inhibition of CHS reactions to the allergen DNFB in wild-type mice. PDT blunted both local (allergen applied to PDT-treated site) as well as systemic (allergen applied to non-PDT-treated area) contact hypersensitivity reactions. Our use of PAFR-deficient mice (Figure 6b) confirms the involvement of PAFR signaling in systemic immunosuppression (Ferracini et al., 2015). Importantly, PAFR activation appears to play a potential role in local immunosuppressive responses to topical PDT (Figure 6b). Studies using the aSMase inhibitor imipramine as well as aSMase-deficient mice indicate the involvement of this enzyme on PDT-generated MVP release. The use of pharmacologic and genetic approaches to inhibit global PAFR and aSMase provides compelling evidence that both processes are involved in both the local and systemic immunosuppressive responses associated with PDT.

Topical PDT as practiced in a standard manner for patients undergoing this treatment as field therapy for actinic keratosis was found to both generate PAF (Figure 1d) as well as inhibit DTH reactions to recall antigens in human subjects (Figure 5). It should be noted that there are no reports we are aware of documenting PDT-induced systemic immunosuppression in humans. In fact, localized PDT has been reported to not result in systemic immunosuppression in humans (Frost et al., 2011, Matthews and Damian, 2010). We hypothesize that topical PDT would need to involve a relatively large body surface area for systemic immunosuppression to occur. Hence, for our studies, we only tested human volunteers undergoing PDT to at least 10% body surface area (see Supplementary Tables S1 and S2).

The consequences of PDT-induced immunosuppression are unclear, yet likely serve to diminish the effectiveness of this modality. The exact agent(s) associated with MVP which serve as the effectors for the immunosuppressive effects of PDT are at present unknown and is certainly an area for future investigation. Yet the current studies indicate that use of a topical aSMase inhibitor imipramine can block both MVP release as well as the immunosuppressive effects of PDT. Of note, imipramine treatment to HaCaT cells before or after the blue light treatment associated with experimental PDT had no effect upon decreased cellular viability (Supplemental Figure S8). These findings suggest that imipramine would not interfere with the effectiveness of topical PDT in which the cytotoxic effect is a necessary component.

In summary, the current studies demonstrate that PDT generates MVP, and these subcellular bodies appear to mediate the known immunosuppressive effects of this FDA-approved modality. Given that this bystander effect of immunosuppression from topical PDT could negatively impact the therapeutic efficacy of this modality commonly used for field therapy for actinic damage, agents that target aSMase could potentially improve PDT effectiveness.

Materials and Methods

Chemicals and in vitro cell studies

All chemicals were obtained from Sigma-Aldrich (St. Louis, MO) unless indicated otherwise. Cell lines were grown (Liu et al., 2021; Mahajan et al., 2021), and PDT of cells was carried out as previously reported (Ferracini et al., 2015). Please see Supplemental Materials and Methods for more details.

Mice

Female C57BL/6-wild type or SKH-1 mice (PAFR expressing; age 6-8 weeks) were purchased from The Charles River Laboratories (Wilmington, MA) or Envigo (Indianapolis, IN). PAFR-deficient (Ptafr−/−) and aSMase-deficient (Smpd1−/−) mice on C57BL/6 backgrounds are previously described (Liu et al., 2021). Murine experiments testing PDT and CHS reactions were carried out exactly as previously published (Liu et al., 2021; Ferracini et al., 2015) and are described further in Supplemental Materials and Methods.

Human skin explants

De-identified discarded skin was obtained from contouring surgeries (abdominoplasties and brachioplasties) (Liu et al., 2021). Skin was washed and fat trimmed, and placed in PBS warmed at 37 °C. After various times/treatments, duplicate 6 mm punch biopsies were obtained and weighed. The tissues were treated as detailed for murine skin.

Human studies

The studies outlined in this work consisted of two separate human studies involving volunteers who were undergoing topical PDT with 20% 5-ALA (LEVULAN KERASTICK, Dusa Pharmaceuticals, Inc., Billerica, MA) or 5-ALA 10% gel (Ameluz; Biofrontera, Inc., Woburn, MA) and the BLU-U light source (Dusa). Subjects were not on any topical/oral immunosuppressives, and not taking greater than minimum daily requirements of antioxidant vitamins. Please see Supplemental Materials and Methods for further details.

Isolation and measurement of MVPs

Microvesicle particles were collected from culture medium or tissue as previously described (Liu et al., 2021, Liu et al., 2020). See Supplemental Materials and Methods for more details.

Measurement of PAFR agonistic activity

The presence and quantitation of PAFR agonists in lipid extracts derived from HaCaT cells, murine or skin biopsies were assessed as previously described (Liu et al., 2021). See Supplemental Materials and Methods for more details.

Statistics

All statistical calculations were performed using Graph-Pad Prism, version 6.0 (GraphPad Software). Statistical significance was determined by 2-sided Student’s t test or 1-way ANOVA with the Holm-Šidák post hoc test (α = 5%). For the human studies involving examination of DTH reactions the Wilcoxon Rank Sum Test was conducted. A P value of less than 0.05 was considered significant.

Study approvals.

All studies involving mice were approved by the IACUC of Wright State University or Indiana University School of Medicine. All studies involving humans were approved by the IRB of Wright State University and followed Declaration of Helsinki Principles. Volunteers provided written informed consent before enrollment in the study.

Supplementary Material

1

Acknowledgements

This research was supported in part by grants from the National Institutes of Health grants R01 HL062996 (J.B.T.), R01 ES031087 (C.A.R., J.B.T., Y.C.), R21 AR070010 (C.E.W.S., J.B.T.) and Veteran’s Administration Merit Award 5I01BX000853 (J.B.T.). Author C.A.R. (craig.rohan@wright.edu) should be addressed regarding the human studies contained in this manuscript.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflicts of Interest

None of the authors have a conflict of interest.

Data Availability Statement

No large data sets were generated in these studies.

References Cited

  1. Awoyemi AA, Borchers C, Liu L, Chen Y, Rapp CM, Brewer CA, et al. Acute ethanol exposure stimulates microvesicle particle generation in keratinocytes. Toxicol Lett 2022;355:100–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Babilas P, Landthaler M, Szeimies RM. Photodynamic therapy in dermatology. Eur J Dermatol 2006;16(4):340–8. [PubMed] [Google Scholar]
  3. Bayless S, Travers JB, Sahu RP, Rohan CA. Inhibition of photodynamic therapy induced-immunosuppression with aminolevulinic acid leads to enhanced outcomes of tumors and pre-cancerous lesions. Oncol Lett 2021;22(3):664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bazzan E, Tine M, Casara A, Biondini D, Semenzato U, Cocconcelli E, et al. Critical Review of the Evolution of Extracellular Vesicles’ Knowledge: From 1946 to Today. Int J Mol Sci 2021;22(12). [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Beyer C, Pisetsky DS. The role of microparticles in the pathogenesis of rheumatic diseases. Nat Rev Rheumatol 2010;6(1):21–9. [DOI] [PubMed] [Google Scholar]
  6. Bianco F, Perrotta C, Novellino L, Francolini M, Riganti L, Menna E, et al. Acid sphingomyelinase activity triggers microparticle release from glial cells. EMBO J 2009;28(8):1043–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bihl JC, Rapp CM, Chen Y, Travers JB. UVB Generates Microvesicle Particle Release in Part Due to Platelet-activating Factor Signaling. Photochem Photobiol 2016;92(3):503–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Boukamp P, Petrussevska RT, Breitkreutz D, Hornung J, Markham A, Fusenig NE. Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. J Cell Biol 1988;106(3):761–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chang WH, Cerione RA, Antonyak MA. Extracellular Vesicles and Their Roles in Cancer Progression. Methods Mol Biol 2021;2174:143–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dougherty TJ, Marcus SL. Photodynamic therapy. Eur J Cancer 1992;28A(10):1734–42. [DOI] [PubMed] [Google Scholar]
  11. Elmets CA, Bowen KD. Immunological suppression in mice treated with hematoporphyrin derivative photoradiation. Cancer Res 1986;46(4 Pt 1):1608–11. [PubMed] [Google Scholar]
  12. Ferracini M, Sahu RP, Harrison KA, Waeiss RA, Murphy RC, Jancar S, et al. Topical photodynamic therapy induces systemic immunosuppression via generation of platelet-activating factor receptor ligands. J Invest Dermatol 2015;135(1):321–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fiechter S, Skaria A, Nievergelt H, Anex R, Borradori L, Parmentier L. Facial basal cell carcinomas recurring after photodynamic therapy: a retrospective analysis of histological subtypes. Dermatology 2012;224(4):346–51. [DOI] [PubMed] [Google Scholar]
  14. Frost GA, Halliday GM, Damian DL. Photodynamic therapy-induced immunosuppression in humans is prevented by reducing the rate of light delivery. J Invest Dermatol 2011;131(4):962–8. [DOI] [PubMed] [Google Scholar]
  15. Gupta AK, Paquet M. Network meta-analysis of the outcome ‘participant complete clearance’ in nonimmunosuppressed participants of eight interventions for actinic keratosis: a follow-up on a Cochrane review. Br J Dermatol 2013;169(2):250–9. [DOI] [PubMed] [Google Scholar]
  16. Hayami J, Okamoto H, Sugihara A, Horio T. Immunosuppressive effects of photodynamic therapy by topical aminolevulinic acid. J Dermatol 2007;34(5):320–7. [DOI] [PubMed] [Google Scholar]
  17. Herring JM, McMichael MA, Smith SA. Microparticles in health and disease. J Vet Intern Med 2013;27(5):1020–33. [DOI] [PubMed] [Google Scholar]
  18. Ishii S, Kuwaki T, Nagase T, Maki K, Tashiro F, Sunaga S, et al. Impaired anaphylactic responses with intact sensitivity to endotoxin in mice lacking a platelet-activating factor receptor. J Exp Med 1998;187(11):1779–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Jansen MHE, Kessels J, Nelemans PJ, Kouloubis N, Arits A, van Pelt HPA, et al. Randomized Trial of Four Treatment Approaches for Actinic Keratosis. N Engl J Med 2019;380(10):935–46. [DOI] [PubMed] [Google Scholar]
  20. Kwiatkowski S, Knap B, Przystupski D, Saczko J, Kedzierska E, Knap-Czop K, et al. Photodynamic therapy - mechanisms, photosensitizers and combinations. Biomed Pharmacother 2018;106:1098–107. [DOI] [PubMed] [Google Scholar]
  21. Li Y, Bax C, Patel J, Vazquez T, Ravishankar A, Bashir MM, et al. Plasma-derived DNA containing-extracellular vesicles induce STING-mediated proinflammatory responses in dermatomyositis. Theranostics. 2021. May 21;11(15):7144–7158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Liangpunsakul S, Rahmini Y, Ross RA, Zhao Z, Xu Y, Crabb DW. Imipramine blocks ethanol-induced ASMase activation, ceramide generation, and PP2A activation, and ameliorates hepatic steatosis in ethanol-fed mice. Am J Physiol Gastrointest Liver Physiol 2012;302(5):G515–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Liu L, Awoyemi AA, Fahy KE, Thapa P, Borchers C, Wu BY, et al. Keratinocyte-derived microvesicle particles mediate ultraviolet B radiation-induced systemic immunosuppression. J Clin Invest 2021;131(10). [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Liu L, Fahy KE, Awoyemi AA, Thapa P, Kelly LE, Chen J, et al. Thermal Burn Injury Generates Bioactive Microvesicles: Evidence for a Novel Transport Mechanism for the Lipid Mediator Platelet-Activating Factor (PAF) That Involves Subcellular Particles and the PAF Receptor. J Immunol 2020;205(1):193–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liu ML, Williams KJ, Werth VP. Microvesicles in Autoimmune Diseases. Adv Clin Chem 2016;77:125–75. [DOI] [PubMed] [Google Scholar]
  26. Mahajan AS, Arikatla VS, Thyagarajan A, Zhelay T, Sahu RP, Kemp MG, et al. Creatine and Nicotinamide Prevent Oxidant-Induced Senescence in Human Fibroblasts. Nutrients 2021;13(11). [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Marathe GK, Johnson C, Billings SD, Southall MD, Pei Y, Spandau D, et al. Ultraviolet B radiation generates platelet-activating factor-like phospholipids underlying cutaneous damage. J Biol Chem 2005;280(42):35448–57. [DOI] [PubMed] [Google Scholar]
  28. Matthews YJ, Damian DL. Topical photodynamic therapy is immunosuppressive in humans. Br J Dermatol 2010;162(3):637–41. [DOI] [PubMed] [Google Scholar]
  29. McIntyre TM, Prescott SM, Stafforini DM. The emerging roles of PAF acetylhydrolase. J Lipid Res 2009;50 Suppl:S255–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mroz P, Hamblin MR. The immunosuppressive side of PDT. Photochem Photobiol Sci 2011;10(5):751–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Nakanishi H, Shindou H, Hishikawa D, Harayama T, Ogasawara R, Suwabe A, et al. Cloning and characterization of mouse lung-type acyl-CoA:lysophosphatidylcholine acyltransferase 1 (LPCAT1). Expression in alveolar type II cells and possible involvement in surfactant production. J Biol Chem 2006;281(29):20140–7. [DOI] [PubMed] [Google Scholar]
  32. Nobbe S, Trueb RM, French LE, Hofbauer GF. Herpes simplex virus reactivation as a complication of photodynamic therapy. Photodermatol Photoimmunol Photomed 2011;27(1):51–2. [DOI] [PubMed] [Google Scholar]
  33. Norling LV, Dalli J. Microparticles are novel effectors of immunity. Curr Opin Pharmacol 2013;13(4):570–5. [DOI] [PubMed] [Google Scholar]
  34. Ochsner M. Photophysical and photobiological processes in the photodynamic therapy of tumours. J Photochem Photobiol B 1997;39(1):1–18. [DOI] [PubMed] [Google Scholar]
  35. Pei Y, Barber LA, Murphy RC, Johnson CA, Kelley SW, Dy LC, et al. Activation of the epidermal platelet-activating factor receptor results in cytokine and cyclooxygenase-2 biosynthesis. J Immunol 1998;161(4):1954–61. [PubMed] [Google Scholar]
  36. Pelletier F, Garnache-Ottou F, Angelot F, Biichle S, Vidal C, Humbert P, et al. Increased levels of circulating endothelial-derived microparticles and small-size platelet-derived microparticles in psoriasis. J Invest Dermatol 2011;131(7):1573–6. [DOI] [PubMed] [Google Scholar]
  37. Sahu RP, Yao Y, Konger RL, Travers JB. Platelet-activating factor does not mediate UVB-induced local immune suppression. Photochem Photobiol 2012;88(2):490–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Shimizu T Lipid mediators in health and disease: enzymes and receptors as therapeutic targets for the regulation of immunity and inflammation. Annu Rev Pharmacol Toxicol 2009;49:123–50. [DOI] [PubMed] [Google Scholar]
  39. Takeshita J, Mohler ER, Krishnamoorthy P, Moore J, Rogers WT, Zhang L, et al. Endothelial cell-, platelet-, and monocyte/macrophage-derived microparticles are elevated in psoriasis beyond cardiometabolic risk factors. J Am Heart Assoc 2014;3(1):e000507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Tamagawa-Mineoka R, Katoh N, Kishimoto S. Platelet activation in patients with psoriasis: increased plasma levels of platelet-derived microparticles and soluble P-selectin. J Am Acad Dermatol 2010;62(4):621–6. [DOI] [PubMed] [Google Scholar]
  41. Thapa P, Bhadri S, Borchers C, Liu L, Chen Y, Rapp CM, et al. Low UVB Fluences Augment Microvesicle Particle Generation in Keratinocytes. Photochem Photobiol 2022;98(1):248–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Travers JB. Platelet-Activating Factor as an Effector for Environmental Stressors. Handb Exp Pharmacol 2020;259:185–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Walterscheid JP, Ullrich SE, Nghiem DX. Platelet-activating factor, a molecular sensor for cellular damage, activates systemic immune suppression. J Exp Med 2002;195(2):171–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wehner MR. Comparing the efficacy of field treatments for actinic keratosis: a critical appraisal of a randomized trial in the New England Journal of Medicine. Br J Dermatol 2020;182(6):1343–4. [DOI] [PubMed] [Google Scholar]
  45. Wolf P, Fink-Puches R, Reimann-Weber A, Kerl H. Development of malignant melanoma after repeated topical photodynamic therapy with 5-aminolevulinic acid at the exposed site. Dermatology 1997;194(1):53–4. [DOI] [PubMed] [Google Scholar]
  46. Zhang Q, Yao Y, Konger RL, Sinn AL, Cai S, Pollok KE, et al. UVB radiation-mediated inhibition of contact hypersensitivity reactions is dependent on the platelet-activating factor system. J Invest Dermatol 2008;128(7):1780–7. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1870133-supplement-1.pdf (8.9MB, pdf)

Data Availability Statement

No large data sets were generated in these studies.

RESOURCES