Electron Donors Rather Than Reactive Oxygen Species Needed for Therapeutic Photochemical Reaction of Near-Infrared Photoimmunotherapy

Abstract

Near-infrared photoimmunotherapy (NIR-PIT) employs molecularly targeted antibodies conjugated with a photoabsorbing silicon-phthalocyanine dye derivative which binds to cancer cells. Application of NIR light following binding of the antibody–photoabsorber conjugates (APCs) results in ligand release on the dye, dramatic changes in solubility of the APC–antigen complex, and rapid, irreversible cell membrane damage of cancer cells in a highly selective manner, resulting in a highly immunogenic cell death. Clinically, this process results in edema after treatment mediated by reactive oxygen species (ROS). Based on the chemical and biological mechanism of NIR-PIT cytotoxicity and edema formation, in order to minimize acute inflammatory edema without compromising therapeutic effects, l-sodium ascorbate (l-NaAA) was administered to quench harmful ROS and accelerate the ligand release reaction. l-NaAA suppressed acute edema by reducing ROS after NIR-PIT yet did not alter the therapeutic effects. NIR-PIT could be performed safely under existence of l-NaAA without side effects caused by unnecessary ROS production.

Near-infrared photoimmunotherapy (NIR-PIT) is a newly developed cancer therapy that employs molecularly targeting antibodies conjugated with a photoabsorbing silicon-phthalocyanine derivative called IRDye700DX (IR700).^1,2 Approximately 1 day after intravenous administration, antibody–photoabsorber conjugates (APCs) bind to cancer cell surface markers. Once the APCs bind to the cell membrane, exposure to NIR light selectively induces rapid cell-specific, necrotic, and highly immunogenic cell death (ICD) mediated by the photoinduced ligand release reaction of IR700.³ Dramatic morphological changes occur in the APC and the APC–antigen complex.⁴ This photoactivated hydrolysis reaction is driven by electron transfer and IR700 anion radical formation, resulting in a break in the silicon–oxygen bond and silanol formation, which converts the dye from very hydrophilic to very hydrophobic.⁵ This NIR-light-induced ligand release reaction is preferably induced under electron donor-rich conditions including hypoxia and rich sulfhydryl groups that are frequently found on APCs in the cancer tissue.⁴ Therefore, NIR-PIT selectively kills target cancer cells.⁶⁻⁸

NIR-PIT has been translated into clinical use. A global phase III clinical trial of NIR-PIT in inoperable head and neck cancers, which targets the epidermal growth factor receptor (EGFR), is currently underway (https://clinicaltrials.gov/ct2/show/NCT03769506). In September 2020, the first clinically approved APC, cetuximab–IR700 (ASP1929, Akalux), was conditionally approved for advanced head and neck cancers and was registered by the Pharmaceuticals and Medical Devices Agency (PMDA) in Japan.

However, some patients experience transient acute edema after the therapy. In most cases, this is not consequential; however, in cases where the treated tumor was located near a critical structure such as the airway in the neck or the mediastinum, edema might cause serious side effects such as airway obstruction. In general, such edema related to acute inflammation is caused by reactive oxygen species (ROS) that are generated by monocytes and macrophages in the tumor bed.^9,10 Additionally, when irradiating NIR light during NIR-PIT, ROS are produced by unbound APC under oxygen-rich conditions.¹¹ Therefore, transient severe edema may be caused by ROS production in the area around the treated tumor beds. We hypothesized that an appropriate reducing agent, such as l-NaAA, which quenches ROS but is an electron donor which promotes NIR-light-induced ligand release reaction of IR700, could suppress acute edema while maintaining the efficacy of NIR-PIT.

Results

l-NaAA Accelerated Release of Axial Ligands with NIR Irradiation Regardless of the Presence of Sodium Azide (NaN₃)

The compound Pc 3, the phthalocyanine moiety of IR700, was synthesized according to Figure S1. The photoinduced ligand release reaction of Pc 3 after NIR light exposure was accelerated in the presence of l-NaAA, an electron donor, in a dose-dependent manner. The 700 nm fluorescence of Pc 3 was decreased in a dose-dependent manner of NIR light and l-NaAA (Figure 1a,b). The loss of fluorescence was associated with precipitation of Pc 3 in an l-NaAA dose-dependent manner because Pc 3 dramatically changed its solubility after light exposure (Figure 1c). Fluorescence emission of IR700 conjugated with panitumumab (panitumumab-IR700 or pan-IR700) was decreased in a dose-dependent manner of NIR light and l-NaAA (Figure 1d,e). In the presence of l-NaAA, NaN₃ did not inhibit the decrease of panitumumab–IR700 fluorescence after NIR light irradiation (Figure 1f). To identify intermediate products of IR700 and l-NaAA radicals, electron spin resonance (ESR) spectroscopy was performed. A broad signal was observed in the central field (Figure 1g i); however, this signal was also observed in empty quartz ESR tubes, indicating that this signal was derived from the quartz ESR tube itself. Thus, this signal could be subtracted and did not interfere with the ESR measurements. Under argon-saturated conditions (Figure 1g, left), a small but clear doublet signal with a g-value of 2.0051 and a splitting width of 0.18 mT was observed just by mixing IR700 with l-NaAA for several minutes (Figure 1g ii). The ESR parameters of this signal are almost the same as those reported previously for the l-NaAA radical,^12,13 which is produced during the oxidation of l-NaAA. This observation suggests that l-NaAA radicals are produced by the redox interaction between IR700 and l-NaAA without NIR light irradiation. When the sample containing 10 mM l-NaAA and 0.5 mM IR700 was irradiated with NIR light, a small signal with a broad line width of 0.87 mT and a g-value of 2.0006 appeared as indicated by an asterisk in the left spectrum of Figure 1g iii. Since this signal was not observed in the absence of an electron donor such as l-NaAA and the intensity of the signal in air, where electrophilic molecule oxygen is present, was much smaller than that under the argon-saturated conditions, this ESR signal was assigned to the IR700 anion radical produced by the electron transfer from l-NaAA to the NIR-light-induced triplet state IR700. In this experimental procedure, the scanning of the magnetic field and sample irradiation started at the same time and continued for 8 min/10 mT. Therefore, the central region where the signal due to the IR700 anion radical was detected occurred about 4 min into the scan. This signal derived from the IR700 anion radical was observed for at least 16 min after NIR light irradiation. This signal was derived from the IR700 anion radical, indicating that NIR light irradiation produces IR700 anion radicals in the presence of l-NaAA. In the left spectra of Figure 1g iv,v, the effect of NaN₃ on the formation of l-NaAA radicals and NIR-light-induced IR700 anion radicals was examined. The results showed that addition of 100 mM NaN₃ attenuated the formation of l-NaAA radicals but did not affect the formation of IR700 anion radicals produced by NIR light irradiation. Next, a similar experiment was carried out in the presence of oxygen. As shown in the right spectra of Figure 1g ii,iii, the reaction of IR700 with l-NaAA in air showed a stronger radical signal derived from l-NaAA radicals than under argon-saturated conditions, and the l-NaAA radicals were further enhanced by NIR light irradiation. Moreover, it was observed that NIR light irradiation produced IR700 anion radicals in air, but it was greatly reduced compared to the argon-saturated conditions. The effect of NaN₃ on l-NaAA radicals and IR700 anion radicals is shown in the right spectra of Figure 1g iv,v. The addition of 100 mM NaN₃ to the sample decreased the ESR signal intensity derived from the l-NaAA radical but did not appear to affect the formation of the IR700 anion radical produced by NIR irradiation. The axial ligand cleavage was observed under l-NaAA-added conditions, and it was not enhanced by NaN₃ (Figure 1h).

Figure 1.

l-NaAA accelerates ligand release from IR700. (a) 25 μM Pc 3 with various concentrations of l-NaAA in PBS was irradiated with NIR light and imaged using 700 nm fluorescence imaging. (b) Mean 700 nm fluorescence intensity was decreased in both an NIR light- and l-NaAA dose-dependent manner (n = 4, mean ± SEM). (c) Aggregation after NIR light exposure of Pc 3 with l-NaAA. Along with the change from hydrophilic to hydrophobic solubility, the cyan color of IR700 decreased, and aggregation appeared, while 700 nm fluorescence was lost in an l-NaAA dose-dependent manner. (d) 3 μM pan–IR700 with various concentrations of l-NaAA in PBS was irradiated with NIR light and imaged using 700 nm fluorescence imaging. (e) When more than 1 mM l-NaAA in PBS was used, mean fluorescence intensity decreased in an NIR light dose-dependent manner (n = 4, mean ± SEM). (f) In 3 μM pan–IR700 in PBS with any concentration of NaN₃, mean fluorescence intensity after NIR light exposure was unchanged (n = 4, mean ± SEM). (g) ESR spectra of samples containing IR700 (0.5 mM) and l-NaAA (10 mM) in PBS irradiated with NIR light under argon saturation conditions (left side) or in the presence of air (right side) in five groups as follows: (i) control spectrum (no drug); (ii) control spectrum of a sample containing IR700 (0.5 mM) and l-NaAA (10 mM) in PBS without NIR exposure; (iii) ESR spectra with NIR light exposure under the conditions of (ii); (iv) control spectrum of a sample containing IR700 (0.5 mM), L-NaAA (10 mM), and NaN₃ (100 mM) in PBS without NIR exposure; and (v) ESR spectra with NIR light exposure under the conditions of (iv). ▲, an undesirable but negligible signal originated from the quartz ESR tube as described in the text. (h) Remaining percentage of IR700 (post/pre) in conditions (iii,v). The averaged values are represented (n = 2).

NaN₃ Inhibited Ligand Release of IR700 When l-Cysteine was Added as an Electron Donor

The photoinduced ligand release reaction of Pc 3 and panitumumab–IR700 was evaluated under conditions of added l-cysteine, which is an alternative electron donor to l-NaAA. A similar phenomenon was observed, but it occurred at lower NIR light irradiation levels than when l-NaAA was used (Figure 2a–c). With concentrations of NaN₃ of 10 mM or higher combined with 1 mM l-cysteine, loss of panitumumab–IR700 fluorescence seen with l-cysteine alone was compromised (Figure 2d). However, the addition of 1 mM l-NaAA to 1 mM l-cysteine counteracted the effects of NaN₃ (Figure 2e).

Figure 2.

l-Cysteine accelerates ligand release after NIR light exposure, but NaN₃ inhibits this reaction. (a) Mean 700 nm fluorescence intensity in 25 μM Pc 3 with various concentrations of l-cysteine in PBS was decreased in both an NIR light- and l-cysteine dose-dependent manner (n = 4, mean ± SEM). (b) Aggregation after NIR light exposure in Pc 3 with l-cysteine mixture. The cyan color of IR700 decreased and aggregation appeared, while 700 nm fluorescence was lost in an l-cysteine dose-dependent manner. (c) 3 μM pan–IR700 with various concentrations of l-cysteine in PBS was irradiated with NIR light and imaged using 700 nm fluorescence imaging. Under more than 1 mM l-cysteine condition in PBS, mean 700 nm fluorescence intensity was decreased in an NIR light dose-dependent manner (n = 4, mean ± SEM). (d) Under more than 1 mM NaN₃ with 3 μM pan–IR700 in PBS, mean 700 nm fluorescence intensity after NIR light exposure was significantly decreased (n = 4, mean ± SEM, vs 0 mM NaN₃; two-way ANOVA repeated measures, followed by Dunnett’s test; *p < 0.05; **p < 0.01; ****p < 0.0001). (e) In 3 μM pan–IR700 in PBS with both 1 mM l-NaAA and l-cysteine, NaN₃ did not affect mean 700 nm fluorescence intensity after NIR light exposure (n = 4, mean ± SEM). (f) ESR spectra recorded from samples containing IR700 (0.5 mM) and l-cysteine (10 mM) in PBS when irradiated with NIR light under argon saturation conditions. (i) Control spectrum of a sample containing IR700 (0.5 mM) and l-cysteine (10 mM) in PBS without NIR light exposure. (ii) ESR spectra recorded when the samples were irradiated with NIR light. (iii) ESR spectra shown in red were obtained by the addition of 100 mM NaN₃ to the samples containing IR700 (0.5 mM) and l-cysteine (10 mM) in PBS and irradiating them with NIR under argon-saturated conditions. The ESR spectrum overlaid in gray is identical to the ESR spectrum recorded in the absence of NaN₃ shown in (ii). ▲, an undesirable but negligible signal originated from the quartz ESR tube. Cys, l-cysteine. (g) Remaining IR700% (post/pre) under conditions (ii) and (iii). The averaged values are represented (n = 2). (h) HPLC chart of l-cystine and l-cysteine in the presence of IR700 under several conditions. When the solution of IR700 containing l-cysteine was irradiated under hypoxic conditions, the peaks of l-cysteine and l-cystine were observed, although there was no peak of l-cystine without irradiation. On the other hand, in the case of aerobic conditions, the peak of l-cysteine almost disappeared after irradiation.

The control ESR spectrum of 0.5 mM IR700 and 10 mM l-cysteine without irradiation is shown in Figure 2f i. When the sample of 0.5 mM IR700 and 10 mM l-cysteine was dissolved in 100 mM phosphate buffer solution (pH 7.0) and irradiated with NIR light under argon-saturated conditions, a symmetrical ESR spectrum with a line width of 0.87 mT and a g-value of 2.0006, which was assigned to the IR700 anion radical, was observed, as shown in Figure 2f ii. Furthermore, to evaluate the effect of NaN₃ on NIR-light-induced IR700 anion radicals, similar experiments were carried out in the presence of 100 mM NaN₃ in the sample. The ESR spectrum in the presence of NaN₃ (red color) was superimposed on the spectrum in the absence of NaN₃ (gray; the same spectrum as Figure 2f ii), as shown in Figure 2f iii. These results showed that NaN₃ decreased the signal of the IR700 anion radical. When NaN₃ was added, ligand cleavage decreased (Figure 2g). This result is consistent with the results of ESR, in which the formation of radical anions was inhibited. To determine whether l-cystine is produced from l-cysteine, the solution of IR700 was irradiated in the presence of l-cysteine and analyzed by high-performance liquid chromatography (HPLC). When IR700 was irradiated under hypoxic conditions, the peaks of l-cysteine and l-cystine were observed, although there was no peak of l-cystine without NIR light irradiation (Figure 2h). On the other hand, upon NIR light irradiation in the presence of oxygen, the peak of l-cysteine was greatly reduced, while the peak of l-cystine was larger than that under hypoxic conditions. These results indicated that l-cysteine acted as a reductant to IR700 in the excited state and was converted to l-cystine. In addition, l-cysteine was consumed more efficiently in the presence of oxygen than in the absence of oxygen. This may be because oxygen quenches IR700 in the triplet excited state or the anion radical in the ground state, resulting in a more efficient cycle of excitation and reduction.

l-NaAA Counteracts the Suppression of NIR-PIT Cytotoxicity Caused by NaN₃

To investigate the effect of reducing agents on the effectiveness of NIR-PIT, the cytotoxic effects of NIR-PIT targeting cancer cells were quantitatively assessed by cell viability assays. Under low or pharmacological concentrations, l-NaAA enhanced the cytotoxic damage of panitumumab–IR700 NIR-PIT (Pan-PIT). Under supraphysiologic concentrations of l-NaAA (e.g., over 1 mM), cytotoxic damage by Pan-PIT was slightly inhibited. In contrast, l-cysteine slightly enhanced the cytotoxic damage of Pan-PIT but only at very low concentrations like 0.001 mM, while it compromised the cytotoxicity at concentrations of 0.1 mM or higher (Figure 3a,b). When both l-cysteine and l-NaAA were present in equal concentrations, l-cysteine tended to counteract the effects of l-NaAA because of the higher redox potential of l-cysteine compared to l-NaAA (Figure 3c). Next, we evaluated the impact of each reducing agent on cytotoxicity of Pan-PIT in the presence of NaN₃. l-NaAA overcame the reduced cytotoxicity caused by NaN₃, but l-cysteine could not reverse the effects of NaN₃ (Figure 3d,e). In another cell line, MDAMB468, l-NaAA also restored cytotoxicity of Pan-PIT suppressed by NaN₃ (Figure S2). A similar tendency was observed by bioluminescence imaging (BLI) (Figure S3). These results suggested that the suppression of NIR-PIT cytotoxicity by NaN₃ was overcome by l-NaAA but not by l-cysteine, despite its higher redox potential, probably because added l-cysteine formed disulfide bonds with the sulfhydryl groups of l-cysteine which compromised its ability to accelerate the photoinduced ligand release reaction.

Figure 3.

Efficacy of panitumumab-IR700 NIR-PIT (Pan-PIT) in A431 GFP-luc cells with reducing agents with NaN₃ added in vitro. Membrane damage of A431 GFP-luc cells induced by Pan-PIT with l-NaAA (a) or l-cysteine (b) was measured using PI staining (n = 4, mean ± SEM, versus NIR-PIT without l-NaAA or l-cysteine; one-way ANOVA, followed by Dunnett’s test; **p < 0.01; N.S., not significant). (c) Membrane damage of A431 GFP-luc cells by NIR-PIT with both l-NaAA and l-cysteine was measured using PI staining (n = 4, mean ± SEM, versus NIR-PIT without reducing agents; one-way ANOVA, followed by Dunnett’s test; **p < 0.01; ***p < 0.001; N.S., not significant). Under 10 mM NaN₃ conditions, membrane damage of A431 GFP-luc cells induced by Pan-PIT with l-NaAA (d) or l-cysteine (e) was measured by PI staining (n = 4, mean ± SEM; one-way ANOVA, followed by Tukey’s test; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; N.S., not significant).

l-NaAA Accelerates Ligand Release of IR700 after NIR Light Exposure

To evaluate whether l-NaAA or l-cysteine better facilitates ligand release from IR700 in vivo, we utilized albumin–IR700 (alb–IR700) and measured loss of fluorescence (Figure S4a). After exposing alb–IR700 to NIR light, the signal of 700 nm fluorescence decreased at the irradiated site, whereas the signal within the liver increased in an NIR light dose-dependent manner (Figure S4b,d). Under l-NaAA administration, fluorescence was significantly decreased compared to the condition without l-NaAA; however, there were no obvious changes in fluorescence with l-cysteine (Figure S4c,e). Decreased 700 nm fluorescence in the liver means that superior ligand release occurred with alb–IR700 after NIR light exposure because ligand release results in loss of fluorescence. Thus, these results demonstrated that l-NaAA accelerated ligand release of IR700 conjugated to albumin in vivo, but l-cysteine did not.

l-NaAA Did Not Affect the Efficacy of Cancer Cell-Targeted NIR-PIT

We investigated if l-NaAA affected the antitumor effects of cancer cell-targeted NIR-PIT utilizing a CD44–IR700 in vivo allograft tumor model. First, l-NaAA was confirmed not to affect tumor growth (Figure S5). The treatment and imaging regimen is shown (Figure 4a). The accumulation of CD44–IR700 within MC38-luc tumors was confirmed by fluorescence imaging, and this decreased after NIR light exposure regardless of whether l-NaAA was present (Figure 4b,e). The therapeutic efficacy of CD44-targeted NIR-PIT was evaluated with BLI as a readout (Figure 4d). Regardless of the presence of l-NaAA, both NIR-PIT groups showed significantly lower intensity on BLI compared to the control group, indicating cancer cell death and that there was no significant difference between NIR-PIT with l-NaAA and without l-NaAA (Figure 4e). In the NIR-PIT group without l-NaAA on day 1, severe edema formation was observed in the skin over the tumor. This likely attenuated much of the bioluminescence signal. The MC38-luc, LL/2-luc, and MOC2-luc tumor models all showed that the antitumor effects of CD44-targeted NIR-PIT were not changed with l-NaAA (Figure S6a–f). In athymic mice, l-NaAA also did not affect the efficacy of Pan-PIT (Figure S7).

Figure 4.

Efficacy of l-NaAA for CD44-targeted NIR-PIT in an MC38-luc tumor model. (a) Treatment schedule. (b) Fluorescence imaging and (c) mean 700 nm fluorescence intensity before and after CD44-targeted NIR-PIT with or without l-NaAA in the MC38-luc tumor-bearing mouse (n = 10, mean ± SEM; paired t-test; N.S., not significant). (d) BLI before (day 6) and after (days 8–12) CD44-targeted NIR-PIT in MC38-luc tumor bearing mice. (e) Luciferase activity calculated from BLI (n = 10; repeated measure two-way ANOVA, followed by Tukey’s test; ****p < 0.0001; N.S., not significant). (f) Edema formation (red arrowhead) was shown 1 day after CD44-targeted NIR-PIT. (g) Body weight curve (n = 10, mean ± SEM; repeated measure two-way ANOVA, followed by Tukey’s test; N.S., not significant). (h) T2WI fat-sat MRI imaging 1 day after NIR-PIT. The red arrowhead shows the MC38-luc subcutaneous tumor. (i) Mean high intensity area calculated from coronal T2WI fat-sat MR images (n = 3, mean ± SEM; unpaired t-test; *p < 0.05).

l-NaAA Suppressed Edema Formation after Treatment with NIR-PIT

We assessed edema formation after CD44-targeted NIR-PIT with and without l-NaAA. In MC38-luc, LL/2-luc, and MOC2-luc tumor-bearing mice, edema was observed on the dorsal surface around the subcutaneous tumor 1 day after NIR-PIT in mice without l-NaAA (Figures 4f and S6g). However, no edema was detected in mice treated by NIR-PIT with l-NaAA injection. No significant difference in body weight was observed between the two groups (Figure 4g). In other allograft models, no significant difference in body weight was detected after CD44-targeted NIR-PIT with l-NaAA (Figure S6h,i).

To quantify the degree of edema, magnetic resonance imaging (MRI) was performed 1 day after CD44-targeted NIR-PIT. In the non-l-NaAA group, T2-weighted and fat saturation images (T2WI fat-sat) showed extensive edema depicted as high signal intensity areas extending from the peripheral tumor site to the lower edge of the thorax and to the proximal side of the tail in NIR-PIT. In contrast, in the l-NaAA group, minimal edema was observed just around the treated tumors (Figure 4h). The area of high signal intensity corresponding to post-treatment edema on the T2WI fat-sat images was quantified and showed that the high signal area was significantly larger in the group treated without l-NaAA versus the l-NaAA group (Figure 4i). This was also confirmed on the short-TI inversion recovery images (STIR) (Figure S8). To investigate the cause of edema formation, ROS was evaluated with chemiluminescence imaging (CLI) using L-012 in MC38-luc tumor-bearing mice before and after NIR-PIT (Figure S9a–c). The CD44-targeted NIR-PIT with l-NaAA showed significantly lower intensity (lower ROS) on CLI compared with NIR-PIT without l-NaAA. These data suggested that l-NaAA inhibited edema formation induced by ROS after NIR-PIT.

l-NaAA Suppressed Edema Formation after NIR-PIT Targeting Immuno-Suppressive Cells without Interfering with Efficacy

The efficacy of CTLA4-targeted NIR-PIT with and without l-NaAA was evaluated in MC38-luc tumors. In both groups, CTLA4–IR700 fluorescence was observed 1 day post injection prior to NIR-PIT but decreased after NIR light exposure (Figure S10a–c). The NIR-PIT group without l-NaAA had a lower BLI tumor signal than the NIR-PIT group with l-NaAA at day 1 to day 3 after NIR light exposure (Figure S10d). However, by day 6 post NIR-PIT, no difference was observed between the two groups, and both NIR-PIT groups showed significantly lower BLI intensity compared to the control group (Figure S10e). Since CTLA4-targeted NIR-PIT showed stronger edema formation than CD44-targeted NIR-PIT, a lower intensity on BLI was shown in the NIR-PIT without the l-NaAA group due to the skin thickening. Edema formation in mice treated by NIR-PIT with l-NaAA was also suppressed both macroscopically and by MRI (Figure S10f–h). The day after NIR-PIT, a significant lower weight was observed in mice in the NIR-PIT with the l-NaAA group (Figure S10i), perhaps due to the comparative lack of edema. In CTLA4-targeted NIR-PIT, higher ROS was also detected in the NIR-PIT without the l-NaAA group (Figure S9d,e). To confirm whether selective cytotoxicity of CTLA4-targeted NIR-PIT was preserved in the NIR-PIT with the l-NaAA group, selective depletion of CTLA4-expressing cells was assessed in MC38-luc tumors 3 h after CTLA4-targeted NIR-PIT (Figure S10j). CTLA4^hi cells were selectively depleted in both NIR-PIT with and without l-NaAA groups. Thus, l-NaAA had no effect on the efficacy of immune-suppressive cell-targeted NIR-PIT, but it did suppress edema formation by ROS.

Discussion

Selective cytotoxicity induced by NIR-PIT is caused by the photoinduced ligand release reaction which occurs under hypoxic electron donor-rich conditions.⁴ The ligand release causes dramatic changes in solubility of the APC–antigen complex, causing damage to the cell membrane and cell death. However, under normoxic or hyperoxic conditions, NIR light induces ROS production in the APC that likely contributes to non-selective cytotoxicity and local edema following NIR-PIT. Another possible cause of local edema formation may be lymphatic obstruction associated with direct lymphatic damage due to NIR-PIT. However, no obvious lymphatic occlusion or morphological changes by NIR-PIT were observed (Figure S11). Thus, ROS increases vascular permeability, resulting in edema formation. l-NaAA is an electron donor that should facilitate NIR-PIT ligand release reaction yet quenches most of ROS, thereby reducing edema, as shown in Figures 4h and S10g on MRI. Since ROS quenching could, in theory, interfere with cancer cell cytotoxicity, it might affect the therapeutic effects of NIR-PIT. However, NIR-PIT with l-NaAA injection still showed acceptable cytotoxicity but without edema formation. Therefore, l-NaAA administration on NIR-PIT could result in an effective therapy that is safer when edema could affect the airways or mediastinum.

Under hypoxic conditions, l-NaAA radical production was minimal, but under normoxic conditions, the amount of l-NaAA radical production was extremely large (Figure 1g ii,iii), indicating that oxygen molecules are involved in the production of the l-NaAA radical. As a working hypothesis, it is possible that the IR700 anion radical donates one electron to oxygen to produce the superoxide anion radical (O₂^–·) in the presence of oxygen, or the triplet state IR700 generated by photoexcitation gives energy to triplet oxygen (³O₂) to produce singlet oxygen (¹O₂). It has been reported that l-NaAA reacts with O₂^–·and ¹O₂ to form the l-NaAA radical and hydrogen peroxide.^14,15 Therefore, it is possible that in the presence of oxygen, l-NaAA may react with these ROS, resulting in the generation of many l-NaAA radicals. If this hypothesis is correct, large doses of l-NaAA could consume generated ROS, reducing the side effects of NIR-PIT in clinical practice.

Although the IR700 anion radical was clearly observed when l-cysteine was used as the reducing agent (Figure 2f ii), the signal was very low when l-NaAA was used as the reducing agent (Figure 1g iii). This may be due to the difference in the properties of l-cysteine and l-NaAA as reducing agents. l-Cysteine is a simple reductant with one-electron reducing ability, while l-NaAA is ionized in water to become the l-NaAA monoanion, and this l-NaAA monoanion acts as a reductant in the reaction process. The l-NaAA monoanion not only has one-electron reducing ability but also acts as a one-hydrogen donor, producing ESR-detectable l-NaAA radicals with a relatively long lifetime.¹⁶ In our previous study employing quantum chemical calculations,⁵ it was shown that in the photoreduction of IR700 in the presence of a one-electron reductant such as l-cysteine, axial ligand cleavage requires hydrolysis by a one-electron transfer from the reductant and a one-proton transfer from H₂O as occurs with l-NaAA. When l-cysteine was used as a reducing agent, one electron was donated to IR700, and long-lived IR700 anion radicals were detected by ESR because the proton transfer from H₂O to IR700 anion radicals was relatively slow. However, when l-NaAA was used as the reducing agent, the donation of one electron and one proton from the l-NaAA monoanion to IR700 molecules occurred simultaneously, so the IR700 anion radical proceeded quickly to the axial ligand cleavage reaction as soon as it was generated and thus could not be detected by ESR (Figures 1g and 5a).

Figure 5.

Schema indicates the proposed mechanism of ROS quenching and accelerating ligand release of IR700 by l-NaAA. (a) Energy diagram of the photoinduced ligand release reaction of IR700. IR700 receives electrons and becomes a radical anion. l-Cysteine and l-NaAA act as electron donors. The ligand release reaction from the radical anion is accelerated under acidic conditions. Thus, ligand release reaction is accelerated by l-NaAA. (b) Mechanism of l-NaAA for suppressing acute edema after NIR-PIT. During NIR-PIT, ROS is also generated and causes edema. l-NaAA quenches ROS, resulting in preventing acute edema, yet facilitates photoinduced ligand release. ISC, intersystem crossing.

Previous studies have suggested that NaN₃, a known singlet oxygen and ROS quencher, partially suppress the cytotoxicity of NIR-PIT. However, these experiments were performed in vitro because the necessary dose of NaN₃ is toxic for living animals in vivo. In this study, we demonstrated that NaN₃ directly suppressed the NIR-light-induced ligand release reaction of IR700 by quenching IR700 radical anion formation as shown on ESR (Figures 1g and 2f). Suppression of ligand release is measurable by IR700 fluorescence loss, which also equates to treatment effectiveness. NaN₃ partially suppresses cytotoxicity by suppressing NIR-light-induced ligand release reaction of IR700. Therefore, the role of ROS in the mechanism of action of NIR-PIT cytotoxicity has been overstated in the literature.^11,17

l-Cysteine is a potentially good alternative to l-NaAA as a reducing agent due to its higher redox potential than l-NaAA and possible clinical use. In an IR700 solution, l-cysteine accelerated photoinduced ligand release by electron donation, especially under hypoxic conditions. However, once IR700 is conjugated to an antibody, NIR-light-induced ligand release reaction predominates under hypoxic conditions and is not highly influenced by l-cysteine. We suggest that the sulfhydryl groups on the l-cysteine side chains of antibodies could donate electrons to IR700 when covalently conjugated and activated by the NIR light. l-Cysteine addition even protected cells in vitro from NIR-PIT, probably because l-cysteine forms disulfide bonds with free sulfhydryl groups on the antibody under normoxic cell culture conditions. l-Cystine formation hampers effective electron donation to light-activated IR700, suppressing the ligand release reaction. l-NaAA both is an electron donor and can assist in protonation of excited IR700, which promotes the ligand release reaction. Therefore, although l-cysteine is a good reducing agent and might accelerate ligand release reaction in vitro, it is difficult to use it for this purpose in vivo.

There are a few limitations to this study. First, we used subcutaneous allograft models to evaluate the edema formation by NIR-PIT. An orthotopic tumor model may be more clinically appropriate; however, it is technically difficult to evaluate the edema formation in this model.^18,19 Although we detected l-cystine formation during NIR-PIT, we could not determine if it originated from free l-cysteine or from the l-cysteine residues of antibodies because it is technically difficult to determine exactly where the reaction occurs. Since l-NaAA has a short circulation half-life, ROS produced by NIR light exposure to IR700 was mostly quenched by l-NaAA in this experimental setting. However, ROS produced by monocytes and macrophages were also quenched, and this could negatively or positively affect the immune reaction induced by NIR-PIT. There do not appear to be significant short-term effects of l-NaAA; however, it is more difficult to assess the impact on long-term anti-cancer immunity. Moreover, due to the great variability in every step of immune response and the relatively small number of mice in each group, it is difficult to prove that there are either no or any effects to long-term immunity.²⁰ The fact that NIR-PIT is a local and not systemic therapy is reassuring in this regard.

Additionally, it would be interesting to show head-to-head difference between monoclonal antibody (mAb)–IR700 APC and mAb–conventional photodynamic therapy (PDT) photosensitizer conjugates or even between IR700 and a conventional PDT photosensitizer, which dedicatedly produce ROS after light exposure. Due to great differences of water-soluble characteristics between IR700 and PDT photosensitizers, there were technical difficulties in synthesizing comparable antibody conjugates in chemistry or cellular micro-localization of each reagent in biology. Therefore, we could not find an appropriate way to perform this comparison.

In summary, l-NaAA, acting as both a direct electron donor to IR700 and a reducing agent for ROS during NIR-PIT, facilitates selective cytotoxicity of NIR-PIT while suppressing local edema. Based on the relatively low toxicity of l-NaAA and the safety clinical profile at the dose used in this study, it may serve as a useful adjuvant to NIR-PIT, improving its safety profile while not interfering with its efficacy (Figure 5b).

Experimental Section

Reagents

A water-soluble, silica-phthalocyanine derivative IRDye700DX NHS ester was obtained from LI-COR Biosciences (Lincoln, NE, USA). Panitumumab, a fully humanized IgG2 mAb directed against EGFR, was purchased from Amgen (Thousand Oaks, CA). The anti-mouse CTLA4 antibody (Clone; 9D9) was purchased from Bio X Cell (West Lebanon, NH, USA). l-NaAA and l-cysteine were purchased from MilliporeSigma (Burlington, MA, USA). Sodium azide (NaN₃) was purchased from MP biomedicals, LLC (Solon, OH, USA).

Synthesis

The compounds were prepared by a slight modification of a previously reported method (Figure S1).⁴ General chemicals were of the best grade available, supplied by FUJIFILM Wako Pure Chemical Corporation, Tokyo Chemical Industries Co., Ltd., Kanto Chemical Co., INC., and Sigma-Aldrich Japan K. K., and were used without further purification. ¹H NMR spectra were recorded on a JNM-ECX400P or JMN-ECS400 (JEOL Ltd., Tokyo, Japan) instrument at 400 MHz and are reported relative to deuterated solvent signals.

Silicon Phthalocyanine Dihydroxide (Pc 1)

Silicon tetrachloride (1.93 g, 11.4 mmol) and 1,3-diiminoisoindoline (1.10 g, 7.59 mmol) were dissolved in quinoline (13 mL), and the mixture was refluxed for 2 h under an argon atmosphere. After the mixture was cooled to room temperature (RT), 5 M NaOH aq (10 mL) was added, and the mixture was refluxed for 2 h. The product was recovered by filtration, washed with MeOH, and dried in vacuo (970 mg, 1.69 mmol). The product was used for the next reaction without further purification.

Bis(3-Aminopropyldimethylsilyl Oxide) Silicon Phthalocyanine (Pc 2)

Pc 1 (500 mg, 0.87 mmol) and 3-aminopropyldimethylethoxysilane (1.12 g, 6.96 mmol) were dissolved in pyridine (250 mL), and the mixture was refluxed overnight under an argon atmosphere, concentrated by rotary evaporation. The residue was diluted, filtered, washed with a H₂O–ethanol solution (2:1), and dried in vacuo [650 mg, 0.807 mmol, yield 82% (2 steps)]. ¹H NMR (400 MHz, CD₃OD): δ −2.86 (s, 12 H), −2.33 to −2.27 (m, 4 H), −1.28 to −1.19 (m, 4 H), 1.18 (t, J = 7.2 Hz, 4 H), 8.35 (dd, J = 5.7, 2.8 Hz, 8 H), 9.66 (dd, J = 5.7, 2.8 Hz, 8 H).

Bis{3-[Tris(3-sulfopropyl)]ammoniopropyldimethylsilyloxide} Silicon Phthalocyanine (Pc 3)

Pc 2 (300 mg, 0.373 mmol), 1,3-propansultone (2.27 g, 18.6 mmol), and N,N-diisopropylethylamine (DIEA, 4.82 g, 37.4 mmol) were dissolved in EtOH (15 mL), and the mixture was stirred at 50 °C for 120 h under an argon atmosphere. The product was purified using an HPLC system (Shimadzu Co., Kyoto, Japan) with a reverse-phase column Inertsil ODS-3 (10 mm × 250 mm) (GL Sciences Inc., Tokyo, Japan) using eluent A [H₂O, 0.1 M triethylammonium acetate (TEAA)] and eluent B (99% MeCN, 1% H₂O) (A/B = 80/20 to 50/50 in 15 min, 50/50 to 0/100 in 5 min). The product was desalted with a Sep-Pak C18 cartridge (Waters Corporation, Milford, MA) and cation-exchange resin, affording Pc 3 (216 mg, 0.132 mmol, yield 36% as a sodium salt). ¹H NMR (400 MHz, CD₃OD): δ −2.79 (s, 12 H), −2.12 to −2.18 (m, 4 H), −0.88 to −0.98 (m, 4 H), 1.75–1.66 (m, 12H), 2.05–1.98 (m, 4 H), 2.80–2.73 (m, 24 H) 8.52 (dd, J = 5.7, 3.0 Hz, 8 H), 9.79 (dd, J = 5.7, 3.0 Hz, 8 H).

Synthesis of IR700–Conjugated Bovine Albumin, Panitumumab, the Anti-CD44 Antibody, and the Anti-CTLA4 Antibody

Bovine albumin (1 mg, 15.0 nmol), panitumumab (1 mg, 6.8 nmol), and anti-CD44 and anti-CTLA4 antibodies (1 mg, 6.7 nmol) were incubated with 5-fold molar excess of IR700 NHS ester (albumin; 146.9 μg, 75.2 nmol, panitumumab; 66.9 μg, 34.2 nmol, anti-CD44 and anti-CTLA4 antibodies; 65.1 μg, 33.3 nmol 10 mmol L^–1 in DMSO) in 0.1 mol L^–1 Na₂HPO₄ (pH 8.5) at RT for 1 h. The mixture was purified with a Sephadex G25 column (PD-10; GE Healthcare, Piscataway, NJ, USA). The concentration of IR700 was determined with absorption at 689 nm using UV–vis (8453 Value System; Agilent Technologies, Santa Clara, CA, USA). The protein concentration was confirmed with a Coomassie Plus protein assay kit (Thermo Fisher Scientific, Rockford, IL, USA) by measuring the absorption at 595 nm. We performed SDS-PAGE as a quality control for each conjugate. We abbreviate panitumumab–IR700 as pan–IR700, albumin–IR700 as alb–IR700, anti-CD44–IR700 as CD44–IR700, and anti-CTLA4–IR700 as CTLA4–IR700.

700 nm Fluorescence Evaluation of Pc 3 or mAb–IR700 after NIR Light Exposure in vitro

25 μM Pc 3 or 3 μM pan–IR700 solution with 0, 0.1, 0.5, 1.0, and 10 mM l-NaAA or l-cysteine in PBS buffer (pH 7.5) was prepared. For the analysis mixed with NaN₃, 3 μM mAb–IR700 solution with 1 mM l-cysteine and/or 1 mM l-NaAA adding different concentrations of NaN₃ (0, 0.01, 0.1, 1, 10, or 100 mM) diluted by PBS were prepared. Pan–IR700 was exposed to NIR light (690 nm, 150 mW cm^–2) under each condition (0, 5, 10, 25, 50, 100, 150, and 200 J cm^–2) with an ML7710 laser system (Modulight, Tampere, Finland). Before and after each irradiation, the 700 nm fluorescence intensity was obtained using a fluorescence imager (Pearl Imager, LI-COR Bioscience, Lincoln, NE, USA). Pearl Cam Software (LI-COR Biosciences) was used for analyzing fluorescence. The same regions of interest (ROIs) were put on the solution in each tube, and then, mean 700 nm fluorescence intensity was measured. The appearance of the tube before and after NIR light irradiation was imaged. All the experiments were carried out at RT.

ESR Spectroscopy

Thirty microliters of 0.5 mM IR700 in 100 mM phosphate buffer (pH 7.0) without or with 10 mM l-cysteine or 10 mM l-NaAA was filled into a gas-permeable polymethylpentene (TPXTM) tube (0.76 mm I.D. × 1.0 mm O.D. × 60 mm long, Toho Kasei Sangyo Co., Ltd., Tokyo, Japan), which was then sealed at both ends, and this TPX tube was placed in a natural quartz ESR tube (5 mm O.D. × 250 mm long, S-5-EPR-250S, Norell Inc., NC). The ESR sample tube was set in a cylindrical TE011 mode cavity (JEOL), and argon gas was allowed to flow into the ESR sample tube through a long capillary glass tube to remove oxygen from the sample. After 5 min of ventilation with argon gas, the ESR sample tube was sealed and X-band CW-ESR measurements were performed at ambient temperature using a JEOL-RE1X spectrometer (JEOL, Tokyo, Japan). The ESR scan (scan time: 8 min/10 mT) was started from the low-magnetic-field side at the same time as irradiation of NIR light started. The ESR parameters were as follows: incident microwave, 10 mW power; microwave frequency, 9.150 GHz; modulation frequency, 100 kHz; field modulation amplitude, 0.1 mT; time constant, 0.3 s; scan range, 325.17 ± 5 mT; and receiver gain, 5.0 × 10². The line width, intensity, and g-value of the ESR signal were estimated using the signal of a co-mounted Mn²⁺ marker (ES-DM1, JEOL) in cavity and Win-Rad software (Radical Research, Tokyo, Japan).

HPLC Analysis of Cystine Formation from Cysteine

The solution of 0.25 mM IR700 in 10 mM sodium phosphate buffer (pH 7.0) containing 1 mM l-cysteine was prepared in a sealed cuvette, and argon was bubbled through the rubber septum cap of the cuvette for 20 min. The deoxygenated solution was then irradiated with a laser (MLL-III-690, Changchun New Industries Optoelectronics Technology Co., Ltd., Changchun, China) (690 nm, 1 W cm^–2) for 1 h. The solution without argon bubbling was also irradiated as above. The irradiated solutions were filtered with a Millex-LH filter (pore size: 0.45 μm, hydrophilic, PTFE, Merck Millipore, US). The solutions were analyzed using an HPLC system (Shimadzu Co., Kyoto, Japan) with a reverse-phase column Inertsil ODS-3 (4.6 mm × 250 mm) (GL Sciences Inc., Tokyo, Japan) using eluent A (H₂O, 0.1% trifluoroacetic acid) and eluent B (99% MeCN, 1% H₂O) (A/B = 99/1) at a flow rate of 1 mL min^–1. The detection wavelength was 202 nm. The solutions were also co-injected with authentic compounds to identify the l-cysteine and l-cystine peaks.

Cell Culture

Luciferase-expressing human or murine cancer cell lines, A431 GFP-luc (EGFR-expressing epidermoid carcinoma), MDAMB468 GFP-luc (EGFR-expressing breast cancer), MC38-luc (colon cancer), LL/2-luc (lung cancer), and MOC2-luc (oral cancer) were used for this study. A431 GFP-luc, MDAMB468 GFP-luc, MC38-luc, and LL/2-luc cells were cultured in the RPMI1640 medium (Thermo Fisher Scientific) supplemented with 10% fetal bovine serum and 100 I.U. mL^–1 penicillin/streptomycin (Thermo Fisher Scientific). MOC2-luc cells were cultured in the mixture of the IMDM medium and Ham’s Nutrient Mixture F12 Media (at a ratio of 2:1, GE Health Healthcare Life Sciences) supplemented with 5% fetal bovine serum, 100 I.U. mL^–1 penicillin/streptomycin, 5 ng mL^–1 insulin (MilliporeSigma), 40 ng mL^–1 hydrocortisone (MilliporeSigma), and 3.5 ng mL^–1 human recombinant EGF (MilliporeSigma). All cells were cultured in a humidified incubator at 37 °C in an atmosphere of 95% air and 5% CO₂.

In vitro NIR-PIT with Reducing Agents

A431 GFP-luc or MDAMB468 GFP-luc cells (4 × 10⁵) were seeded into 12-well plates, incubated for 24 h, then exposed to media containing pan–IR700 (10 μg mL^–1) for 1 h at 37 °C. After aspirating media with APC, phenol-red-free media containing various concentrations of l-NaAA, l-cysteine, and/or NaN₃ were added in each well. NIR light (690 nm, 150 mW cm^–2, 1 or 2 J cm^–2 for A431 GFP-luc, 20 J cm^–2 for MDAMB468 GFP-luc) was irradiated to cancer cells. For flow cytometric analyses, the cells were collected with trypsin 1 h after NIR light exposure. Cells were then stained with propidium iodide (PI, 1 μg mL^–1) at RT for 5 min. The fluorescence of the cells was then analyzed with a flow cytometer (FACSCalibur, BD Biosciences, San Jose, CA, USA) and FlowJo software (BD Biosciences).

Animals and Tumor Models

All in vivo procedures were conducted in compliance with the Guide for the Care and Use of Laboratory Animal Resources (1996), US National Research Council, and approved by the local Animal Care and Use Committee. Six- to eight-week-old female C57BL/6 mice and homozygote athymic nude mice were purchased from The Jackson Laboratory (Bar Harbor, ME, USA) and Charles River Laboratories (Wilmington, MA, USA), respectively. A431 GFP-luc (1 × 10⁶), MC38-luc (1 × 10⁶), LL/2-luc (0.5 × 10⁶), or MOC2-luc (1 × 10⁶) cells were inoculated into the right side of the dorsum. The hair overlying the tumor site was removed before NIR-PIT and imaging studies for C57BL/6 mice. The body weight was measured from the day before to 4–6 days after NIR light exposure. The mice were euthanized with CO₂ when all experiments involving BLI were finished.

In vivo NIR-PIT

C57BL/6 tumor-bearing mice were randomized into three groups as follows: (i) no treatment (control), (ii) intravenous (I.V.) administration of APCs followed by NIR light exposure (NIR-PIT) without l-NaAA, and (iii) NIR-PIT with l-NaAA. The APC was injected 6 days after inoculation of cancer cells into C57BL/6 mice. Similarly, tumor-bearing mice of athymic mice were randomized into two groups as follows: (i) NIR-PIT without l-NaAA and (ii) NIR-PIT with l-NaAA. The dose of APC was 100 μg for CD44–IR700 and 50 μg for CTLA4–IR700. NIR light (690 nm, 150 mW cm^–2, 50 J cm^–2) was irradiated on the next day. l-NaAA (80 mg per mouse) or PBS was injected intraperitoneally (I.P.) 15 min before NIR light exposure. Upon NIR light exposure, the normal tissue adjacent to the tumor was covered with aluminum foil to ensure that the NIR light exposure is limited in the tumor site. Dorsal fluorescence images of IR700 were obtained with the 700 nm fluorescence channel of the Pearl Imager (LI-COR Biosciences). The images were taken pre- and post-NIR-PIT. ROIs were placed on the tumor. The dorsal edema formation of treated mouse 1 day after NIR light irradiation was imaged using a camera. Acute treatment efficacy was evaluated with BLI analysis, in which d-luciferin (3 mg per mouse; Gold Biotechnology) was intraperitoneally injected into mice. Luciferase activity was analyzed with a BLI system (Biospace Lab) using relative light units. ROIs were placed over the entire tumor. The counts per minute of relative light units were calculated using M3 Vision Software (Biospace Lab) and converted to the percentage based on those before NIR-PIT using the following formula: [(relative light units after treatment)/(relative light units before treatment) × 100 (%)]. Bioluminescence images were continually recorded until the onset of depilation-induced skin pigmentation precluded accurate measurement.

Magnetic Resonance Imaging

Under pentobarbital anesthesia, MRI was performed on a 3-T scanner using an in-house 10-inch circle-shaped mouse receiver coil array (Elition 3T; Philips Medical Systems, Best, Netherlands) 1 day after NIR light exposure. Scout images were obtained to accurately locate the tumor. All mice underwent T2WI fat-sat and STIR. All images were obtained in the coronal plane, and T2WI fat-sat images were also obtained in the axial plane. All images were analyzed using Image J software (http://rsb.info.nih.gov/ij/), and 3D imaging was reconstructed. The high signal intensity area derived from each image in T2WI fat-sat and STIR was calculated by Image J.

Flow Cytometric Analysis

To confirm the depression of CTLA4-expressing cells, the tumors of MC38-luc were harvested 3 h after NIR light exposure. Single-cell suspensions from tumor samples were prepared using the following protocol. Whole tumors were incubated in the RPMI 1640 medium (Thermo Fisher Scientific) containing collagenase type IV (1 mg mL^–1; Thermo Fisher Scientific) and DNase I (20 μg mL^–1; Millipore Sigma, Burlington, MA, USA) in 37 °C for 30 min and then gently cut with scissors and mashed with the back of the plunger of a 3 mL syringe. The tissues were passed through a 70 μm cell strainer (Corning, Corning, NY, USA). A total of 3.0 × 10⁶ cells were stained, and data for 5.0 × 10⁵ cells were collected for each tumor. The cells were stained with the anti-CTLA4 antibody (Clone; UC10-4B9) purchased from Thermo Fisher Scientific. To distinguish live cells from dead cells, cells were also stained with LIVE/DEAD Fixable Dead Cell Stain (Thermo Fisher Scientific). The fluorescence of the cells was then analyzed with a flow cytometer (FACSLyric, BD Biosciences) and FlowJo software (FlowJo LLC). Dead cells were removed from analysis based on FSC, SSC, and staining with LIVE/DEAD Fixable Dead Cell Stain.

Statistical Analysis

Data are expressed as means ± the standard error of the mean (SEM). Statistical analysis was performed with GraphPad Prism (GraphPad Software, LaJolla, CA, USA). The sample size (n) for each experiment is described in each figure legend. For one-time measurement, a two-tailed unpaired t-test (two groups) or a one-way analysis of variance (ANOVA), followed by Tukey’s test or Dunnett’s test (three or more groups), was used. For comparison of luciferase activity, biodistribution, and body weights, a repeated measures two-way ANOVA, followed by Tukey’s test, was used. A p-value of less than 0.05 was considered significant.

Glossary

Abbreviations

NIR-PIT

near-infrared photoimmunotherapy

APC

antibody–photoabsorber conjugate

l-NaAA

l-sodium ascorbate

ROS

reactive oxygen species

IR700

IRDye700DX

ICD

immunogenic cell death

EGFR

epidermal growth factor receptor

ESR

electron spin resonance

HPLC

high-performance liquid chromatography

BLI

bioluminescence imaging

MRI

magnetic resonance imaging

CLI

chemiluminescence imaging

mAb

monoclonal antibody

PDT

photodynamic therapy

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsptsci.1c00184.

• Additional synthesis, NIR-PIT in vitro and in vivo, and MRI data (PDF)

• 3D magnetic resonanance images reconstructed from coronal T2WI fat-sat in the group of NIR-PIT without L-NaAA 1 day after CD44-targeted NIR-PIT (AVI)

• 3D magnetic resonance images reconstructed from coronal T2WI fat-sat in the group of NIR-PIT with L-NaAA 1 day after CD44-targeted NIR-PIT (AVI)

Author Contributions

All authors read and approved the final version of the manuscript. T.K. mainly designed and conducted experiments, performed analysis, verified data, and wrote the manuscript; R.O., Y.G., A.F., F.I., H.W., H.F., D.D., B.T., H.T., and O.I. performed chemical and biological experiments and analysis; P.L.C. wrote the manuscript and supervised the project; M.O. designed and conducted experiments and performed analysis; and H.K. planned and initiated the project, designed and conducted experiments, verified data, wrote the manuscript, and supervised the entire project.

The authors declare no competing financial interest.