Photochemistry of tyrosine dimer: when an oxidative lesion of proteins is able to photoinduce further damage †

The tyrosine dimer (Tyr 2 ), a covalent bond between two tyrosines (Tyr), is one of the most important modi ﬁ cations of the oxidative damage of proteins. This compound is increasingly used as a marker of aging, stress and pathogenesis. At physiological pH, Tyr 2 is able to absorb radiation at wavelengths signi ﬁ - cantly present in the solar radiation and arti ﬁ cial sources of light. As a result, when Tyr 2 is formed in vivo , a new chromophore appears in the proteins. Despite the biomedical importance of Tyr 2 , the information of its photochemical properties is limited due to the drawbacks of its synthesis. Therefore, in this work we demonstrate that at physiological pH, Tyr 2 undergoes oxidation upon UV excitation yielding di ﬀ erent products which conserve the dimeric structure. During its photodegradation di ﬀ erent reactive oxygen species, like hydrogen peroxide, superoxide anion and singlet oxygen, are produced. Otherwise, we demonstrated that Tyr 2 is able to sensitize the photodegradation of tyrosine. The results presented in this work con ﬁ rm that Tyr 2 can act as a potential photosensitizer, contributing to the harmful e ﬀ ects of UV-A radiation on biological systems.


Introduction
One of the most important modifications in the oxidative damage of proteins is the covalent bond between two tyrosines (Tyr), yielding the dityrosine cross-link.The one-electron oxidation of Tyr leads to the long-lived tyrosyl radical (Tyr(-H) • ).The coupling of two Tyr(-H) • yields the dimer o,o′-dityrosine (Tyr 2 , Fig. 1) as the main product. 1 Tyr 2 can be formed photochemically 2,3 or in the dark through reactive oxygen species (ROS) mediated reactions. 4,5This linkage can occur between two Tyr residues in the same molecule (intramolecular), or between two molecules (intermolecular), 6 the latter leading to high molecular weight products. 7Tyr 2 is increasingly used as a marker of aging, stress and pathogenesis.It was identified in many pathological manifestations such as amyloid fibril formation, Parkinson's disease, cataract of the eye lens and epidermoid carcinoma.The phenol groups of Tyr 2 are much more acidic than that of Tyr and the corresponding pK a value was determined to be 7.25. 9The two acid-base forms have well differentiated spectral features, 10,11 with absorbance maxima at 283 and 315 nm for the acid and basic form, respectively (Fig. 1).Therefore, in contrast to Tyr, at physiological pH, more than 50% of Tyr 2 exists in the basic form and absorbs in the UV-B (280-320 nm) and UV-A (320-400 nm) spectral regions (Fig. 1).As a result, when Tyr 2 is formed in vivo a new chromophore appears in the proteins, which is able to absorb, unlike natural amino acids, at wavelengths significantly present in solar radiation and artificial sources of light.
UV radiation is the most harmful component of solar radiation; the UV-B radiation damages DNA and proteins through the direct excitation of the nucleobases and some amino acids, respectively.However, only a small proportion of the energy that reaches the Earth's surface is UV-B, radiation due to the absorption of the ozone layer in the stratosphere.In addition to direct absorption of light, damage to biomolecules can involve photosensitization reactions, which occur through the absorption of UV-A and/or visible light by endogenous or exogenous photosensitizers.The chemical changes in macromolecules ( proteins and DNA) and its components resulting from photosensitized reactions, can take place through different mechanisms: (i) energy transfer from the triplet excited state of the photosensitizer to the biomolecule, 12,13 (ii) photosensitized oxidations, which can take place, through the generation of radicals (Type I mechanism), e.g., via electron transfer or hydrogen abstraction, or/and the production of singlet oxygen ( 1 O 2 ) (Type II mechanism). 14,15Proteins, due to their relatively high abundance, their ability to bind chromophores and the reactivity of particular amino acid residues, are one of the preferential targets of the damaging effect of radiation on biological systems. 16][19][20] Even though Tyr 2 is one of most common oxidative lesions in proteins, absorbs above 300 nm and can be exposed to UV radiation in different situations, to the best of our knowledge; the photochemistry of Tyr 2 has not been investigated.Therefore, the aim of this work was to find out if Tyr 2 undergoes chemical changes upon UV excitation and if it can act as photosensitizer.
We previously developed a method to synthesize Tyr 2 by the photosensitization of Tyr with pterin (Ptr). 9Here, we used this method to prepare aqueous solutions of Tyr 2 , that afterwards were exposed to UV-B and UV-A radiation to study the photochemistry of the acid and basic forms (Fig. 1).In particular, we have determined the quantum yields of photodegradation, identified the resulting photoproducts and analyzed the kinetics, under different experimental conditions.In addition, we have studied the photosensitizing properties of Tyr 2 ; in particular, we have investigated the production of ROS and evaluated its capability to act as a photosensitizer using Tyr as target molecule.The results obtained are evaluated in connection with their biological implications.

General
Formic acid (HCOOH), tyrosine (Tyr), deuterium water (D 2 O), Cytochrome c (Cyt) from horse heart, and Superoxide Dismutase (SOD) from bovine erythrocytes were obtained from Sigma Aldrich (St Louis, MO).Singlet Oxygen Sensor Green (SOSG) was purchased by Life technologies.Aqueous solutions were prepared using ultrapure water from Milli-Q® purification system (Millipore Corporation, USA).
Preparation of Tyr 2 .Tyrosine dimer (Tyr 2 ) was obtained using the method described in ref. 9. Briefly, air equilibrated aqueous solutions containing Tyr (500 μM) and Ptr (100 μM) were irradiated for 10 min in quartz cells (1.0 cm optical path length) at room temperature using a Rayonet RPR lamp (Southern N. E. Ultraviolet Co.) with emission centered at 350 nm.A liquid chromatography equipment from Shimadzu (see below) and Synergi Polar-RP semi-preparative column (ether-linked phenyl phase with polar endcapping, 250 × 10 mm, 4 μm, Phenomenex) were used for isolation of Tyr 2 from HPLC runs ( preparative HPLC), by collecting the mobile phase after passing though the photodiode array detector (PDA).Aqueous solutions ( pH 3.0 ± 0.1) were used as mobile phase. 9The Tyr 2 concentration was determined using Lambert-Beer law (A λ = ε λ Tyr2 l[Tyr 2 ]), using ε 283 Tyr2 = (5680 ± 30) M −1 cm −1 for the molar extinction coefficient of the acid form. 11easurements of pH.The pH measurements were performed using a pH-meter sensION + pH31 GLP combined with a pH electrode 5010T (Hach) or microelectrode XC161 (Radiometer Analytical).The pH of the aqueous solutions was adjusted by adding drops of HCl and NaOH solutions from a micropipette.The concentration of the acid and the base used for this purpose ranged from 0.1 to 2 M.
UV/Vis analysis.UV/Vis absorption spectra were registered on a Shimadzu UV-1800 spectrophotometer, using quartz cells of 1 cm optical path length.

Steady-state irradiation
Irradiation setup.The continuous irradiation of aqueous solutions of Tyr 2 at different pH (5.5 ± 0.1 and 9.5 ± 0.1) was carried out at room temperature.The samples (1 mL) were photolyzed in quartz cell (4 mm optical path length) using an irradiation setup (Newport) composed by a Xenon Lamp (300 W), coupled, through a Mounting Kit (Model 74017), to a motorized monochromator UV-VIS (Oriel Cornerstone 130 1/ 8 m).The acid and basic forms of Tyr 2 were irradiated at (280 ± 20) nm and (320 ± 20) nm, respectively.The experiments were performed in the presence and in the absence of dissolved O 2 .Experiments with air-equilibrated solutions were carried out in open quartz cells without bubbling, whereas argon and oxygen-saturated solutions were obtained by bub-bling for 20 min with these gases, previously water saturated (Linde, purity >99.998%).
Actinometry.Aberchrome 540 (Aberchromics Ltd) was used as an actinometer for the measurement of the incident photon flux densitity (q 0;V n;p ) at the excitation wavelength, which is the amount of incident photons per time interval (q 0 n;p ) and divided by the volume (V) of the sample. 21Aberchrome 540 is the anhydride form of the (E)-α-(2,5-dimethyl-3-furylethylidene)(isopropylidene)succinic acid which, under irradiation in the spectral range 316-366 nm leads to a cyclized form.The method for the determination of q 0;V n;p has been described in detail. 22A value of 1.4 (±0.1) × 10 −6 Einstein per L per s was obtained for q 0;V n;p at 320 nm.The value of the absorbed photon flux density (q a;V n;p ) was calculated from q 0;V n;p according to the Lambert-Beer law: where A is the absorbance of the reactant at the excitation wavelength.
Taking into account that Aberchrome cannot be used to determine the q 0;V n;p at 280 nm, the q 0;V n;p value was calculated using the following equation: where q 0;V n;p (280) and q 0;V n;p (320) are the q 0;V n;p values at 280 nm and 320 nm, respectively and I(280) and I(320) are the emission intensities of the irradiation setup at 280 nm and at 320 nm, respectively.To determine I(280)/I(320) the corresponding emission spectra at the monochromator output were registered with an Ocean Optics SD 2000 spectrometer.
Mass spectrometry analysis.The liquid chromatography equipment coupled to mass spectrometry (LC-MS) system was equipped with an UPLC chromatograph (ACQUITY UPLC, Waters), and a UV/vis detector (Acquity TUV), coupled to a quadrupole time-of-flight mass spectrometer (Xevo G2-QTof-MS, Waters) (UPLC-QTof-MS), equipped with an electrospray ionization source (ESI).UPLC analyses were performed using the Acquity UPLC BEH Phenyl (1.7 μm, 2.1 × 50 mm) column (Waters).An isocratic elution with 99% of aqueous HCOOH (0.1% v/v) and 1% of ACN was used as mobile phase with a flow rate of 0.2 mL min −1 .Mass chromatograms, i.e. representations of mass spectrometry data as chromatograms (the x-axis representing time and the y-axis signal intensity), were registered using different scan ranges.The mass spectrometer was operated in both positive (ESI + ) and negative (ESI − ) ion modes.When solutions of pure Tyr 2 were analyzed, the signal corresponding to the intact molecular ion was observed in the sample stored in dark and after irradiation; new signals were detected and allowed the identification of different photoproducts.Given that the signals were obtained with better resolution in ESI − mode, the mass spectrometry analyses were carried out in this mode.

Reactive oxygen species
Hydrogen peroxide (H 2 O 2 ).For the determination of H 2 O 2 , a Cholesterol Kit (Wiener Laboratorios S.A.I.C.) was used.H 2 O 2 was quantified after reaction with 4-aminophenazone and phenol. 23,24Briefly, 500 μL of irradiated solution were added to 600 μL of reagent.The absorbance at 505 nm of the resulting mixture was measured after 30 min of incubation at room temperature, using the reagent as a blank.Aqueous H 2 O 2 solutions prepared from commercial standards were employed for calibration.

•−
).The O 2 •− was detected using the SOD-inhibitable Cyt reduction assay. 25Solutions of Tyr 2 were irradiated in the presence of 14 μM Cyt.The extent of Cyt reduction was determined by monitoring the increase in the absorbance at 550 nm.The absorption spectra of oxidized and reduced Cyt are very characteristic, and this spectral change is a sensitive assay for electron-transfer reactions.
Singlet oxygen ( 1 O 2 ) detection.For 1 O 2 detection the experiments were carried out at room temperature using deuterium water as a solvent since the lifetime of 1 O 2 (τ Δ ) is long enough to detect it.The 1 O 2 emission in the near-infrared (NIR) region was registered using a NIR PMT Module H10330-45 (Hamamatsu) coupled to a single-photon counting equipment FL3TCSPC-SP, as described elsewhere. 26Briefly, the sample solution (0.7 mL) in a quartz cell (1 cm × 0.4 cm) was irradiated with a CW 450 W Xenon source through an excitation monochromator (330 nm blaze grating).The luminescence in the NIR region, after passing through an emission monochromator (1000 nm blaze grating), was detected at 90°with respect to the incident beam.Corrected emission spectra obtained by excitation at 320 nm were recorded between 950 and 1400 nm.
Singlet oxygen quantum yields (Φ Δ ).The probe Singlet Oxygen Sensor Green (SOSG) was used to determine the Φ Δ .8][29] Air equilibrated aqueous solutions of Tyr 2 (A = 0.12 at 320 nm) were irradiated in the presence of 0.63 mg L −1 of SOSG.The SOSG-EP formed as a function of irradiation time, was detected by its fluorescence with maximum emission at 525 nm upon 470 nm excitation.To quantify 1 O 2 in these experiments, the 525 nm emission was determined using Pterin (Ptr) as a reference of 1 O 2 sensitizer (Φ Δ = 0.30). 30For determining Φ Δ of Tyr 2 , the initial rate of SOSG-EP production ((d[SOSG-EP]/dt ) 0 ) of air equilibrated aqueous solution of Tyr 2 and Ptr were alternatively measured, using matched absorbances at the wavelength(s) of excitation.The Φ Δ of Tyr 2 was calculated using the following equation: where (d[SOSG-EP]/dt ) 0,Tyr2 and (d[SOSG-EP]/dt ) 0,Ptr are the initial rate of SOSG-EP production of air equilibrated aqueous solution of Tyr 2 and Ptr, respectively and Φ Δ,Ptr is the quantum yield of 1 O 2 production by Ptr.

Quantum yield of Tyr 2 consumption
The quantum yield of Tyr 2 consumption (Φ −Tyr2 ) was determined for the acid and basic form of Tyr 2 at 280 and 320 nm, respectively.Values were obtained using eqn (4).
The initial rates were obtained from the slopes of the corresponding plots of Tyr 2 concentration vs. time at acid and alkaline pH.The evolution of Tyr 2 during irradiation was followed by HPLC-FL, using the characteristic emission properties of Tyr 2 at 410 nm when the sample is excited at 280 nm.

Photodegradation of Tyr 2 in aqueous solutions
The first aim of this work was to find out if Tyr 2 was able to suffer degradation in aqueous solutions upon UV irradiation.Therefore acidic ( pH = 5.5 ± 0.1) and alkaline ( pH = 9.5 ± 0.1) air-equilibrated aqueous solutions of Tyr 2 were irradiated at 280 and 320 nm, respectively.Significant changes in the absorption spectra of the alkaline solutions were registered upon irradiation (Fig. 2a).In particular, decrease of the intensity of the bands corresponding to the reactant and an increase of the absorbance at wavelengths longer than 350 nm was observed (Fig. 2a).In contrast, very small spectral changes were registered for the photolysis of the acid form of Tyr 2 (Fig. 2b).No further spectral changes were detected for irradiated solutions stored in the dark for several hours.
Chromatograms obtained by HPLC-FL analysis of irradiated oxygen-free, air-equilibrated and oxygen-saturated solutions (excitation at 280 nm and emission at 410 nm) showed that Tyr 2 suffers photodegradation in the presence of oxygen (Fig. 2).In the experiments lacking oxygen, the Tyr 2 concentration remained constant and no photoproducts were detected, which indicates that the photochemical degradation of Tyr 2 is a photooxidation process.On the other hand, no significant differences in the consumption of Tyr 2 were found in oxygen-saturated in comparison with air-equilibrated solution.
The quantum yield of Tyr 2 consumption (Φ −Tyr2 ) was determined in air-equilibrated acidic and alkaline solutions (eqn (4), Material and methods).For determining the initial rates of Tyr 2 consumption ((−d[Tyr 2 ]/dt ) 0 ), the plots of concentration vs. irradiation time must be linear.Generally, this condition is fulfilled only if the absorbed photon flux density (q a;V n;p ) by the reactant does not decrease significantly (<15%).Therefore, for each experiment it was considered a period of time, within this condition was satisfied.We took into account only the first 20 minutes for acidic media and 10 minutes for alkaline media.The Φ −Tyr2 values obtained for air-equilibrated aqueous solution of Tyr 2 were 0.023 ± 0.003 and 0.0053 ± 0.0005 for the alkaline and acid form, respectively.Considering that these values are higher than those reported for Tyr, 31 which, in addition, does not absorb UV-A radiation, it is clear that at physiological conditions, Tyr 2 is able to suffer photodegradation much easier than Tyr.These results demonstrated that Tyr 2 undergoes degradation upon UV-A irradiation.Since UV-A is the most important UV component of solar radiation and considering that the intensity of our light source is not higher than that of solar radiation, 32 the photochemistry of Tyr 2 becomes relevant and deserves to be studied.Moreover, the quantum yields measured are relatively high, which implicates that the photochemical reaction can be significant even upon exposure in short periods of time and with radiation of low energy.

Analysis of photoproducts
To investigate the products of the photodegradation of Tyr 2 , the irradiated solutions were analyzed by HPLC-PDA and HPLC-FL (Material and methods).The HPLC-PDA analysis showed that Tyr 2 photodegradation in alkaline media led to the formation of many products (Fig. 3).Although noisy, the spectra of three photoproducts, arbitrary named P1, P2 and P3, were recorded (Fig. 3).The photoproduct P3 showed spectral features similar to those reported for the typical oxidation products of Tyr in vivo, dopaminochrome or dopachrome. 33e absorption spectra of these compounds have a broad absorption band of low intensity centered on 475 nm.These compounds are formed by an irreversible intramolecular cyclization of 3,4-dihydroxy-L-phenylalanine (DOPA), which suggests the formation of DOPA as a photoproduct of Tyr 2 . 33n addition, Tyr was searched among the products using a standard solution and the characteristic fluorescence features of the amino acid. 34The chromatograms obtained by HPLC-FL analysis (λ exc = 275 nm, λ em = 300 nm) revealed that Tyr was not a product of Tyr 2 photodegradation, indicating that the dimerization process cannot be reverted photochemically by direct excitation of Tyr 2 .
To gain insight on the structural characteristics of the products formed by the photodegradation of Tyr 2 , a qualitative analysis of these compounds was carried out by liquid chromatography/mass spectrometry using a UPLC equipment coupled to a mass spectrometer (UPLC-QTof-MS, Material and methods).Before irradiation the chromatograms showed a main peak, and the mass spectrum recorded for this peak exhibited the signal corresponding to the intact molecular ion of Tyr 2 as [M − H] − species at m/z 359.1237 (M : Tyr 2 ) (Table 1).
After the irradiation, several new peaks were observed in the chromatograms.The mass spectra recorded at different retention times (t R ) revealed, at least, four products with m/z values higher than that of Tyr 2 (Table 1

) which corresponded to compounds that have incorporated one ([M + O], [M + O − 2H]) or two oxygen atoms ([M + 2O], [M + 2O − 2H]). Incorporation of one oxygen atom ([M + O]
) suggested the oxidation of one unit of Tyr to DOPA, a common oxidation product of Tyr. 35Incorporation of one oxygen atom and the loss of two hydrogen atoms ([M + O − 2H]) suggested the oxidation of one unit of Tyr to dopaquinone, another common oxidation product of Tyr. 33Therefore we propose that the four mentioned products are dimeric structures composed by a Tyr unit and an oxidized Tyr unit, or two oxidized Tyr units (Fig. 4).
Furthermore, UPLC-QTof-MS analysis also revealed the formation of products of molecular weights lower than that of the reactant.In particular, an intense signal was detected for m/z value of 327.0984 (Table 1).Taking into account that the intramolecular cyclization of quinone is a commun reaction, 33 the structure of this product could be a dimeric structure formed by a Tyr and a dopaminochrome (Fig. 4).This proposed structure justifies the absorbance in the visible region observed for  P3 in the HPLC-PDA chromatograms (Fig. 3).In agreement with the previous results of HPLC-FL analysis, Tyr was not found as a product of photodegradation of Tyr 2 .
The UPLC-QTof-MS equipment was used for tandem mass spectrometry (MS/MS) analysis.MS/MS spectra registered for Tyr 2 at different ionization energies showed the loss of different fragments of the side chains (Fig. 5).In contrast, the fragmentation did not involve the aromatic rings, which indicated that the bond between the two Tyr residues is strong enough to not be broken under our experimental conditions.This type of fragmentation pattern has already been observed for DOPA. 36ue to the low concentration, most of the MS/MS spectra of the products were noisy with weak signals.However, in all cases it was observed that the fragmentation took place in the side chains and not in the rings.Fig. S1 (ESI †) shows the MS/ MS spectrum registered for Tyr-DOPA, where a fragmentation pattern similar to that observed for Tyr 2 (Fig. 5) can be appreciated.The MS/MS spectrum of Tyr-DOPA confirmed that the incorporation of the oxygen atom takes place in one of the aromatic rings.In the case of Tyr-dopaminochrome, its MS/MS spectrum (Fig. S2, ESI †) showed the fragmentation of Tyr 2 in only one of the two side chains, supporting the hypothesis that the other chain is involved in the cyclization.
In overview, the results presented in this section showed that the photodegradation of Tyr 2 under UV-A radiation involves the oxidation of the aromatic ring of the chemical structure of Tyr, which leads to many dimers bearing the common oxidation products of Tyr such as DOPA and dopaquinone and colored cyclized compounds.In all the cases, the products conserve the dimeric structure, indicating that once Tyr 2 is formed, it undergoes photooxidation upon UV-A irradiation, but the chemical bond between the two Tyr moieties remains unaltered.

Production of reactive oxygen species
Aromatic amino acids, like Trp and Tyr, are able to produce ROS by direct light absorption 37 or by photosensitization reaction induced by endogenous or exogenous photosensitizers. 17,18We have demonstrated in the previous section that excitation of Tyr 2 with UV-A and UV-B radiation generates electronic excited states that initiate photoreactions.Therefore it is straightforward to think that Tyr 2 might also produce ROS photochemically, which would be relevant since Tyr 2 can be excited at longer wavelengths than Tyr and other aromatic amino acids.Consequently the production of H 2 O 2 , O 2 •− and 1 O 2 was evaluated in air-equilibrated aqueous solutions of Tyr 2 exposed to UV radiation.Hydrogen peroxide determination.In alkaline Tyr 2 solutions (14 μM, pH = 9.5), H 2 O 2 was formed and the concentration increased as a function of irradiation time (Fig. 6).The H 2 O 2 production was compared to the consumption of Tyr 2 (Δ[Tyr 2 ] = [Tyr 2 ] 0 − [Tyr 2 ] t ), simultaneously determined by HPLC.The results indicated that the amount of H 2 O 2 produced was equal, within experimental error, to the consumption of Tyr 2 .This fact suggests that H 2 O 2 is formed as a product of the photooxidation of Tyr 2 .

Photochemical & Photobiological Sciences Paper
In contrast, the concentration of H 2 O 2 detected in irradiated acidic solutions was negligible (Fig. 6), which is in agreement with the results shown for the photodegradation of the acid form of Tyr 2 .No consumption of Tyr 2 was detected when acid and alkaline aqueous solution of Tyr 2 were stored in the dark in the presence of H 2 O 2 , thus indicating that no thermal reaction took place between these compounds.On the other hand, control experiments were performed in the absence of O 2 , and as expected no H 2 O 2 production was observed.
Superoxide anion determination.The H 2 O 2 detected can be the product of the spontaneous disproportionation of O 2 •− , with its conjugate acid HO 2 • . 38Therefore, to investigate the formation of O 2 •− , air-equilibrated aqueous solutions containing 14 μM of Tyr 2 and 14 μM of Cyt at pH 5.5 and 9.5 were irradiated.In alkaline media the difference in the absorbance (ΔA 550 ) increased with irradiation time, reaching a plateau after ca.60 minutes (Fig. 7).The corresponding experimentaldifference (ED) spectra in the range 500-600 nm were in good agreement with those reported in the literature for the reduction of Cyt. 39Photoreduction of Cyt was inhibited in similar experiments performed in the presence of SOD (200 U mL −1 ), an enzyme that catalyzes the conversion of O 2 and H 2 O 2 (Fig. 7).Control experiments were performed to investigate thermal or photochemical processes that might interfere with the assay.No reduction of Cyt was observed in solutions containing 14 μM of Tyr 2 and 14 μM of Cyt at pH 9.5, when kept in the dark for more than 90 min (Fig. S3, ESI †).When solutions containing only Cyt were exposed to UV-A radiation, no changes were observed in the spectra (Fig. S3, ESI †).These results indicated that, under our experimental conditions, O 2 •− was generated photochemically by Tyr 2 and provided evidence for the existence of electron transfer reactions.When the same assay was carried out under acid pH conditions, photoreduction of Cyt was not observed (Fig. 7).This fact is compatible with the results previously shown on photodegradation of Tyr 2 in acidic media, e.g.low quantum yield of reactant consumption (Fig. 2) and negligible production of H 2 O 2 (Fig. 6).
Singlet oxygen.The production of 1 O 2 in Tyr 2 solutions was investigated by monitoring its near-infrared (NIR) luminescence at 1270 nm.Due to the weak emission of 1 O 2 in H 2 O, D 2 O was used as a solvent and the concentrations of Tyr 2 was higher (165 μM) than that used in the experiments presented up to now.Therefore alkaline solutions (pD = 10.5) were irradiated at 320 nm and NIR corrected emission spectra were recorded.As shown in Fig. 8, the typical phosphorescence spectrum of 1 O 2 centered at about 1270 nm was observed in air-equilibrated solutions and such emission disappeared when the oxygen was removed from the solution by bubbling with Argon.These results univocally demonstrated that the basic form of Tyr 2 is able to generate 1 O 2 upon UV-A irradiation.
Controls performed after recording the emission spectra of 1 O 2 revealed that a significant amount of Tyr 2 was consumed, indicating that the concentration and, consequently, the absorbance at 320 nm, did not remain constant during the NIR emission detection.The Tyr 2 consumption took place in these experiments since the Φ −Tyr2 value is high (vide supra) and, to improve the weak luminescence detection, the sample was irradiated with a high energy and for a relative long time.Under these conditions, the determination of the quantum yield of 1 O 2 production (Φ Δ ) by Tyr 2 , using emission measurements, would be very difficult and would lead to a very uncertain result.
Therefore the Φ Δ value in alkaline media was estimated using the probe Singlet Oxygen Sensor Green (SOSG) (Material and methods).Alkaline air-equilibrated aqueous solutions ( pH 9.5) of Tyr 2 (14 μM) were irradiated in the presence of SOSG (0.63 mg L −1 ) and the corresponding emission spectra of the SOSG-EP, upon excitation at 470 nm, were recorded (Material and methods).For each irradiation time, the spectrum before irradiation was subtracted and the resulting difference spectra showed the typical fluorescence band reported for the product of the reaction between SOSG and 1 O 2 (SOSG-EP) [27][28][29]40 and its intensity increased with irradiation time (Fig. 9). Thetime evolution of the fluorescence was registered for Tyr 2 and Ptr, used as a reference (Φ Δ = 0.30 (alkaline media)) 30 (Fig. 9), and Φ Δ value for Tyr 2 , calculated as explained in Material and methods section, was 0.15 ± 0.05.This value is similar to that obtained for free Tyr (Φ Δ = 0.138 ± 0.007) upon irradiation in the UV-C, and higher than those reported for others amino acids like Trp and Phe.37 Therefore, while Tyr is not a relevant 1 O 2 photosensitizer at physiological conditions because it does not absorb radiation coming from the sun or normal artificial sources, Tyr 2 produces 1 O 2 with the same quantum efficiency as Tyr, but upon excitation with the mentioned light sources, due to the shift in the absorption spectrum of the Tyr chromophores (Fig. 1).

Degradation of Tyr photoinduced by Tyr 2
Up to now we have shown that Tyr 2 , under UV-A irradiation, undergoes photodegradation and produces ROS.Considering this behavior, the next logical step was to evaluate if Tyr 2 is able to act as a photosensitizer of biomolecules.To investigate this issue Tyr was used as an oxidizable substrate.Therefore solutions containing Tyr 2 (15 μM) and Tyr (19.7 μM) ( pH = 9.5) were irradiated at 320 nm.Under these conditions only Tyr 2 was excited, whereas Tyr did not absorb.The concentration of reactants was determined by HPLC.As shown in Fig. 10, simultaneously with the expected consumption of Tyr 2 , the concen-tration of Tyr decreased as a function of irradiation time.In a control carried out in the same experimental conditions, but in the absence of Tyr 2 , no consumption of Tyr was observed.Similar results were obtained in experiments performed at different Tyr and Tyr 2 concentrations (Fig. S4, ESI †).
These results clearly demonstrated that Tyr is degraded by photosensitization with Tyr 2 , indicating that this compound is   able to photoinduced chemical modifications in oxidizable biomolecules.This fact is quite relevant because it means that a compound formed by photosensitization can subsequently act as an intrinsic photosensitizer, amplifying the harmful effects of UV radiation on biological systems.

Conclusions
We have studied the photodegradation of the acid and alkaline form of Tyr 2 in aqueous solution under UV-B and UV-A radiation, considering that both acid-base forms are present at physiological conditions.In the absence of oxygen Tyr 2 is photostable, whereas excitation in the presence of oxygen leads to the photodegradation of Tyr 2 .The quantum yield of Tyr 2 photodegradation values are 0.023 ± 0.003 and 0.0053 ± 0.0005 for the alkaline and acid form, respectively.Photolysis of Tyr 2 leads to oxidized products that conserve the dimeric structure, that is the chemical bond between the two Tyr moieties endures the UV exposure.
Reactive oxygen species such as H 2 O 2 , O 2 •− and 1 O 2 are produced upon excitation of the alkaline form of Tyr 2 .The quantum yield of 1 O 2 production is 0.15 ± 0.05, which is similar to that obtained for free Tyr.Therefore, while Tyr and other aromatic amino acids are not relevant 1 O 2 photosensitizers because do not absorb radiation coming from the sun or normal artificial sources, Tyr 2 produces 1 O 2 with the same quantum efficiency as Tyr, but upon excitation with the mentioned light sources, due to the shift in the absorption spectrum of the Tyr chromophore.In addition, we have proven that Tyr 2 is able to photoinduce the degradation of Tyr under UV-A radiation.
The results presented in this work indicate that when Tyr 2 is generated in a protein structure, an intrinsic potential photosensitizer is formed, extending the active fraction of light towards the UVA range.Therefore a product of a photosensitized process can act as a photosensitizer itself leading to further photosensitized damage, thus amplifying the harmful effects of UV radiation on biological systems.

Fig. 1
Fig. 1 Absorption spectra of air-equilibrated aqueous solutions of Tyr (dotted line) and, the acid (dashed line) and the alkaline (solid line) forms of Tyr 2 .Inset: Acid-base equilibrium in aqueous solutions of Tyr 2 .† Electronic supplementary information (ESI) available.See DOI: 10.1039/ c9pp00182d

Fig. 2
Fig. 2 Evolution of the absorption spectra of air equilibrated aqueous solution of Tyr 2 at (a) pH 9.5 and (b) pH 5.5, as a function of irradiation time.Inset: Time evolution of Tyr 2 concentration, calculated from HPLC analysis (FL detector, excitation at 280 nm, emission at 410 nm), in airequilibrated (•), oxygen-saturated (▼) and oxygen-free (▲) aqueous solutions at (a) pH 9.5 and (b) pH 5.5.

Fig. 4
Fig. 4 Proposed molecular structure for the oxidized products detected in LC-MS analysis of alkaline aqueous solution of Tyr 2 exposed to UV-A radiation.

Fig. 8
Fig. 8 NIR luminescence of 1 O 2 photosensitized by Tyr 2 in Argon (gray line) and air-equilibrated D 2 O solution (black line) at pD 10.5.Inset: Difference NIR emission spectrum obtained by subtraction of emission spectrum obtained in argon from that obtained in air-equilibrated solution.

Fig. 10
Fig. 10 Time evolution of the Tyr concentration in alkaline aqueous solutions under UV-A irradiation in the absence (•) and in the presence of Tyr 2 (▼).Inset: Time evolution of the Tyr 2 concentrations in alkaline aqueous solutions under UV irradiation in the presence of Tyr.λ irr = 320 nm, [Tyr] = 19.7 μM, [Tyr 2 ] = 15 μM, pH = 9.5.

Table 1
Molecular formula, observed and calculated mass and mass error of Tyr 2 and oxidized products detected in LC-MS spectra