Improved Natural Killer cell activity and retained anti-tumor CD8(+) T cell responses contribute to the induction of a pathological complete response in HER2-positive breast cancer patients undergoing neoadjuvant chemotherapy.

BACKGROUND: Locally advanced HER2-overexpressing breast cancer (BC) patients achieve a high rate of pathological complete responses (pCR) after neoadjuvant chemotherapy (NC). The apparently unaltered immune proficiency of these patients together with the immune-modulating activities of NC drugs suggest a potential contribution of host immunity in mediating clinical responses. We thus performed an extensive immunomonitoring in locally advanced BC patients undergoing NC to identify immunological correlates of pCR induction. METHODS: The immune profile of 40 HER2-positive and 38 HER2-negative BC patients was characterized at diagnosis and throughout NC (Paclitaxel and Trastuzumab, or Docetaxel and Epirubicin, respectively). The percentages of circulating immune cell subsets including T and B lymphocytes, Natural Killer (NK) cells, regulatory T cells, T helper 17 lymphocytes, were quantified by multiparametric flow cytometry. NK cells functional activity was evaluated through the analysis of NF-kB nuclear translocation by Multispectral flow cytometry, and with the in vitro monitoring of Trastuzumab-mediated antibody-dependent cell cytotoxicity (ADCC). CD8(+) T cell responses against six different tumor-associated antigens (TAA) were characterized by IFN-γ ELISPOT and IFN-γ/IL-2 DualSpot assays. RESULTS: After NC, HER2-positive patients showed a significant increase in the number of NK cells and regulatory T cells irrespective of the pathological response, whereas patients undergoing a pCR disclosed higher percentages of T helper 17 cells. Notably, a significant increase in the number of activated NK cells was observed only in HER2-positive patients achieving a pCR. Characterization of anti-tumor T cell responses highlighted sustained levels of CD8(+) T cells specific for survivin and mammaglobin-A throughout NC in patients undergoing a pCR in both arms. Moreover, HER2-positive patients achieving a pCR were characterized by a multi-epitopic and polyfunctional anti-tumor T cell response, markedly reduced in case of partial response. CONCLUSIONS: These results indicate that maintenance of functional T cell responses against selected antigens and improvement of NK cell proficiency during NC are probably critical requirements for pCR induction, especially in HER2-positive BC patients. Trail registration: TRIAL REGISTRATION NUMBER: NCT02307227, registered on ClinicalTrials.gov ( http://www.clinicaltrials.gov , November 26, 2014).

Cisplatin-gemcitabine as palliative chemotherapy in advanced squamous vulvar carcinoma: report of two cases.

Vulvar cancer (VC) is a rare disease, usually diagnosed in a stage still amenable to potentially curative treatments, including surgery and/or radiation therapy with or without chemotherapy. Several patients however present at diagnosis with metastatic disease and another 30-50% will relapse. Prognosis of metastatic or recurrent disease not amenable to salvage surgery or radiotherapy is very poor. Evidence about the efficacy of chemotherapy in this setting is limited and its role still remains unclear. At present there is no standard treatment for advanced VC and patients are usually treated with schedules adopted for chemoradiation or extrapolated from cervical cancer. We report our experience using a cisplatin-gemcitabine regimen in two cases of metastatic squamous cell VC. No response was obtained with this schedule. No other data are available in the literature about the choice of a cisplatin-gemcitabine regimen in this patient subset. The paucity of evidence about the role of palliative chemotherapy in metastatic VC justifies any effort to implement knowledge. For this reason we think it is notable to also report a negative experience. It is not possible for us to conclude that this chemotherapy would be unable to provide any benefit in a larger sample of patients; nonetheless we think that new agents, rather than combinations of older drugs, could hopefully provide more benefit.

4,6,4'-trimethylangelicin shows high anti-proliferative activity on DU145 cells under both UVA and blue light.

OBJECTIVES: Furocoumarins (psoralens and angelicins) have been already used under ultraviolet A light (UVA) for the treatment of skin diseases and cutaneous T-cell lymphoma. Besides their high anti-proliferative activity, some severe long-term side effects have been observed, for example genotoxicity and mutagenicity, likely strictly related to the formation of crosslinks. It has been demonstrated that blue light (BL) activation of 8-methoxypsoralen, an FDA-approved drug, leads to less mutagenic monoadducts in the DNA. So far, in this work the less toxic and more penetrating BL is proposed to activate 4,6,4'-trimethylangelicin (TMA), an already known UVA photoactivatable compound. MATERIALS AND METHODS: Photocleavage, crosslink formation and oxidative damage were detected in pBR322 plasmid DNA treated with 300.0 µmol/L TMA activated with various exposures of BL. Anti-proliferative activity, reactive oxygen species (ROS) formation and activation status of some signalling pathways involved in cell growth and apoptosis were verified on DU145 cells treated with 5.0 µmol/L TMA plus 2.0 J/cm2 of BL. RESULTS: Under BL-TMA, no mutagenic crosslinks, no photocleavage and neither photooxidative lesions were detected on isolated plasmid DNA. TMA showed high anti-proliferative activity on DU145 cells through induction of apoptosis. Besides ROS generation, the proapoptotic effect seemed to be related to activation of p38 and inhibition of p44/42 phosphorylation. Interestingly, the decrease in nuclear ß-catenin was coupled with a significant dropping of CD44-positive cells. CONCLUSION: Overall, our results indicate that TMA can be activated by BL and may be considered for targeted phototherapy of prostate cancer lesions.

4',5'-substituted methylangelicins: photocycloadducts with pyrimidine bases of DNA.

The isolation and characterization of photocycloadducts with pyrimidine bases from DNA samples irradiated (365 nm) in the presence of four 4',5'-substituted methylangelicins was performed. All these furocoumarins yielded mainly the cis-syn furan-side cycloadduct with thymine. For 4',5'-dimethyl-, 5,4',5'-trimethyl- and 6,4',5'-trimethylangelicin this adduct was accompanied by two pyrone-side adducts (cis-syn and cis-anti), whereas the 4,4',5'-trimethyl derivative gave the furan-side adduct with cytosine. The characterization of the regio- and stereochemistry of the adducts was accomplished by 1H NOE (nuclear Overhauser effect) and 1H-13C HMBC (heteronuclear multiple-bond connectivity) spectroscopies. The formation of different cycloadducts in DNA by the various derivatives highlights the role of the methyl groups in determining the regio- and stereochemistry of the cycloaddition.

Monofunctional angular furocoumarins: sequence specificity in DNA photobinding of 6,4,4'-trimethylangelicin and other angelicins.

The sequence specificity in the photoreaction (365 nm) of 6,4,4'-trimethylangelicin (TMA) with DNA fragments of the lac I gene of Escherichia coli was studied by using DNA sequencing methodology. In order to map the sites of TMA photoaddition, we took advantage of the (3'-5') exonuclease activity associated with T4 DNA polymerase, which is blocked by bulky adducts, such as furocoumarin photoadducts. A quantitative analysis of the sites of photoaddition is reported. TMA was demonstrated to photoreact with thymine and, to a lower extent, to cytosine. AT-rich sequences and TTT sites in a GC context are the most reactive sites towards TMA whereas TA, AT, CA, AC sites are weaker sites with similar reactivity. Cytosines in alternated CG sequences are also targets of TMA photobinding. We observed a less pronounced sequence specificity of TMA than that of other psoralen derivatives already studied (Sage and Moustacchi, 1987; Boyer et al., 1988). A comparison with other furocoumarins 4,4'-dimethylangelicin (4,4'-DMA), 4'-methylangelicin (4'-MA), angelicin, 4,5',8-trimethylpsoralen (TMP) and 8-methoxypsoralen (8-MOP) is also reported. The role of flanking sequence and consequently of the local conformation at the various sites of photoaddition is discussed. A preferential orientation of the TMA molecule during the intercalation in the dark is suggested. Hot alkali treatment of TMA-modified DNA did not reveal any DNA strand breakage due to photooxidized bases.

Photoaddition of 4,6-dimethyltetrahydrobenzoangelicin to thymine in DNA: X-ray studies and experiments with model oligonucleotides.

The crystal structures of 4,6-dimethyltetrahydrobenzoangelicin (THBA), a furocoumarin analog, and of its furan-side cis-syn cycloadduct with thymine formed in the photoreaction with DNA, have been determined. The crystal structure of the latter compound contained only one enantiomeric form corresponding to the addition to a 5'-XpT site. Contrary to most psoralen derivatives studied, THBA showed higher photoreactivity toward synthetic oligonucleotides containing that sequence than toward those with the 5'-TpX sequence.

Dark and photochemical interactions of dimethyltetrahydrobenzoangelicin with DNA.

A study of dark interaction and photoreaction between 4,6-dimethyltetrahydrobenzoangelicin (THBA) and DNA is described. 4,6-Dimethyltetrahydrobenzoangelicin is a furocoumarin derivative in which 4' and 5' carbons are linked by a four-methylene bridge. In spite of the bulky aliphatic ring, THBA forms a complex with DNA in the dark and, on UVA irradiation, reacts with pyrimidine bases of DNA yielding monoadducts only involving its furan side double bond. Two main photoproducts form: they derive from a C4-cycloaddition to thymine and cytosine, respectively, and account for 56% and 39% of the total photoreaction yield. Both show cis-syn configuration. Two other isomers, one with thymine and one with cytosine, formed with so much lower yield (ca 3 and 1%, respectively) that their structure could not be assigned. Furthermore, in spite of its angular structure, THBA induces a small number of crosslinks in DNA.

DNA damage induced by 4,6,8,9-tetramethyl-2H-furo[2,3-h]quinolin-2-one, a new furocoumarin analog: photochemical mechanisms.

Some photochemical and photobiological properties of 4,6,8,9-tetramethyl-2H-furo[2,3-h]quinolin-2-one (HFQ) were studied in comparison with its isomer 1,4,6,8-tetramethyl-2H-furo[2,3-h]quinolin-2-one (FQ) and 8-methoxypsoralen (8-MOP). The HFQ photobinds to DNA forming furan-side monoadducts (MAHFQ) that have molecular structure very similar to those of FQ (MAFQ). Unlike MA8-MOP and MAFQ, MAHFQ no longer photoreact. The HFQ, like FQ, produces moderate amounts of singlet oxygen but no superoxide anions. The HFQ and FQ induce numbers of DNA-protein cross-links (DPC), much more plentiful than those of 8-MOP (about two and seven times, respectively) but no interstrand cross-links. The mechanism of DPC formation was studied in vivo in mammalian cells by alkaline elution and in vitro using a new test mixing histones and DNA from calf thymus. The latter is a very useful technique for the double irradiation protocol. The DNA (or histones) are separately exposed to a first UVA dose in the presence of the sensitizer; then, after its unbound molecules have been removed, histones (or DNA) are added to assemble the chromatin-like complex that is irradiated again. According to in vitro and in vivo methods, DPC appear to be formed by FQ and 8-MOP by a biphotonic process that starts with monoadduct induction in DNA, followed by their conversion into DPC. In the resulting lesions, the sensitizer molecule forms a covalent bridge between the two macromolecules (DPC at length greater than zero). Instead, HFQ induces DPC by a monophotonic process; thus, HFQ is probably not a physical part of the bridge between DNA and proteins, which may be linked together directly, like DPC at zero length induced by UVC.

4-tert-butylperoxymethyl-9-methoxypsoralen as intercalating photochemical alkoxyl-radical source for oxidative DNA damage.

We describe the synthesis of a novel psoralen peroxide 1 that generates on irradiation (350 nm) alkoxyl radicals, namely tert-butoxyl radicals, as confirmed by electron spin resonance studies with the spin trap 5,5-dimethyl-pyrroline-N-oxide. The radical source intercalates into the DNA, which has been demonstrated by linear-flow-dichroism measurements. Thus, the alkoxyl radicals are formed advantageously directly in the DNA matrix. In supercoiled pBR322 DNA, the generation of strand breaks by the photochemically or metal-catalyzed generated alkoxyl radicals is demonstrated. Photosensitization by the psoralen chromophore was excluded because similar substances that do not release radicals caused no DNA damage, nor were the photoproducts of the peroxide 1 active. With calf thymus DNA, 8-oxoGua and small amounts of guanidine-releasing products, e.g. oxazolone, were observed. However, in these reactions the photoproduct also displayed some DNA-oxidizing capacity.

In vitro phototoxic properties of new 6-desfluoro and 6-fluoro-8-methylquinolones.

A representative set of potent antibacterial 6-desfluoro-8-methylquinolones, in which the C-6 fluorine atom is replaced by -NH(2) or -H, and their 6-fluoro counterparts, were investigated to evaluate their phototoxic potential and to explore the mechanism behind their phototoxicity. The capacity to photosensitize biological substrates (lipids, proteins, DNA) has been analyzed, as well as their photocytotoxicity on red blood cells and 3T3 murine fibroblasts. The results obtained show that the quinolones studied are able to photosensitize red blood cell lysis in an oxygen-dependent way and induce a high decrease in cell viability after UVA irradiation. A major correlation with phototoxicity lies in the structure of the individual antibacterials and their hydrophobicity; in particular, 6-amino derivatives are less phototoxic than corresponding unsubstituted and fluorinated compounds. Cellular phototoxicity was inhibited by the addition of free radical and hydroxyl radical scavengers (BHA, GSH and DMTU), suggesting the involvement of a radical mechanism in their cytotoxicity. A good correlation was observed between lipid peroxidation and phototoxicity, indicating that the test compounds exert their toxic effects mainly in the cellular membrane. Preliminary experiments on pBR322 DNA show that these derivatives do not photocleave DNA, differently from the two photogenotoxic fluoroquinolones, ciprofloxacin and lomefloxacin, used as reference compounds.

Antiretroviral activity of furocoumarins plus UVA light detected by a replication-defective retrovirus.

The replication defective retrovirus, pXM5(N2), was used for an easy, safe and reproducible test for the screening of furocoumarins with antiretroviral activity. High titer viral supernatants have been photomodified by UVA light (20 kJ m-2) in the presence of different concentrations of two psolarens (8-methoxypsoralen, 8-MOP and 4,5',8-trimethylpsoralen, TMP) and one angelicin (4,6,4'-trimethylangelicin, TMA). At low concentrations (100-250 ng ml-1) 8-MOP and TMA did not show any significant antiviral activity, while TMP demonstrated a reduction of virus infectivity by one log at 250 ng ml-1. At the highest concentration (5 micrograms ml-1), TMA and TMP reduced the virus titer by one and more than two logs, respectively, being, therefore, two and four times more active than 8-MOP. The most active compound, TMP, was further tested on HIV-1 viral supernatants. Total inactivation of the HIV-1 (200 SFU) was obtained in the presence of 1 microgram ml-1 of TMP and 20 kJ m-2 of UVA light. Our results support the validity of the N2 system to detect the antiretroviral activity of furocoumarins and suggest the potential of TMP in combination with UVA light against HIV-1.

Benzoangelicins: new monofunctional DNA photobinding agents.

4,6-Dimethylbenzoangelicin, obtained by fusing a benzene ring at the furan side of 4,6-dimethylangelicin, was studied in terms of crystal structure and interactions with DNA in both ground and excited states. 4,6-Dimethylbenzoangelicin has a planar structure and forms a molecular complex with DNA, undergoing intercalation inside the double helix. Under UVA irradiation, it photoconjugates covalently with the macromolecule, showing a DNA photobinding rate slightly lower than that of 8-methoxypsoralen, involving however only its 3,4 double bond, i.e. behaving as a pure monofunctional agent. The parameters of dark binding and photobinding were determined, and two C4 cycloadducts with thymine were isolated and characterized.

6,4,4'-trimethylangelicin photoadduct formation in DNA: production and characterization of a specific monoclonal antibody.

The photochemical reactions of 6,4,4'-trimethylangelicin (TMA) with calf thymus DNA and an octanucleotide containing a single thymine have been characterized. HPLC analyses of enzymatically hydrolyzed TMA-DNA showed that isomeric forms of 4',5'-furan-side monoadducts were the major products. To develop monoclonal antibodies Balb c mice were immunized with the TMA-DNA complexed with methylated bovine serum albumin. The resultant antibodies were characterized by enzyme-linked immunosorbent assays (ELISA). The most sensitive antibody (7E3) has high specificity for TMA-DNA, very low cross-reactivity with DNA modified with either 4',5'-dimethylangelicin or 4'-methylangelicin and no cross-reactivity with non-modified DNA or with DNA modified with either 4'-aminomethyl-4,5',8-trimethylpsoralen or 8-methoxypsoralen. To characterize further this antibody, oligonucleotides containing specific TMA photoadducts were isolated from the photoreaction mixture by polyacrylamide gel electrophoresis and used as competitive inhibitors in the ELISA. Autoradiography of the gel showed an intense band corresponding to the 4',5'-monoadduct and two weaker unidentified bands. Antibody 7E3 reacted only with the 4',5'-monoadduct band as would be expected since this photoadduct was the principal photoadduct in the original antigen.

Photosensitized cross-linking and cleavage of pBR322 and M13 DNA: comparison of 4,4',6-trimethylangelicin and 3-carbethoxypsoralen.

The furocourmarins 3-carbethoxypsoralen (3-CP) and 4,4',6-trimethylangelicin (TMA) were generally believed to be incapable of cross-linking DNA upon irradiation with ultraviolet light. Denaturation of photosensitized pBR322 DNA, either supercoiled or previously linearized with a restriction enzyme, proved that 3-CP was indeed monofunctional, but that TMA produced cross-links. Identical conclusions were reached with double stranded M13 DNA which had been linearized with EcoR 1. Both sensitizers also induced partial DNA cleavage. In contrast to 3-CP, photosensitization with TMA made the DNA resistant to enzymatic cleavage.

In vitro studies of the phototoxic potential of the antidepressant drugs amitriptyline and imipramine.

Amitriptyline and imipramine, two tricyclic antidepressant drugs, have been studied to evaluate their phototoxic potential using various models. Reactive oxygen species production was investigated. A negligible production of singlet oxygen was observed for both compounds whereas a significant production of superoxide anion was noted for amitriptyline in particular. Moderate red blood cell lysis under UVA light (365 nm) was induced in the presence of the two drugs at a concentration of 50 microM. Cellular phototoxicity was investigated on a murine fibroblast cell line (3T3). The two drugs were phototoxic causing cell death at a concentration of 100 microM and a UVA dose in the range of 3.3-6.6 J/cm2. Furthermore, the two drugs photosensitized the peroxidation of linoleic acid, as monitored by the formation of dienic hydroperoxides. The presence of BHA and GSH, two free radical scavengers, significantly reduced the lipid oxidation photoinduced by the drugs, suggesting a predominant involvement of radical species. Finally, the involvement of nucleic acids in the phototoxicity mechanism was also investigated using a pBR322 plasmid DNA as a model.

New benzoquinolizin-5-one derivatives as furocoumarin analogs: DNA-interactions and molecular modeling studies.

Three derivatives of 1H,5H and 3H,5H-benzo[ij]quinolizin-5-one (BQZ1), previously prepared by chemical synthesis with the aim of obtaining furocoumarin analogs, have been studied. These are able to intercalate inside DNA and by subsequent irradiation with UVA light, to photoreact with DNA. Compound I (10-methoxy-7-methyl-1H,5H-benzo[ij]quinolizin-5-one) has a potentially photoreactive 2,3 double bond because of its conjugation with the pyridine ring of quinolinone, while compounds II (10-acetoxy-7-methyl-3H,5H-benzo[ij]quinolizin-5-one) and III (10-methoxy-7-methyl-3H,5H-benzo[ij]quinolizin-5-one) have a potentially photoreactive 1,2 double bond conjugated with the benzene ring of quinolinone. Compounds I and III, having a tricyclic planar structure, intercalate inside the DNA, while compound II cannot intercalate efficiently because of the steric hindrance of the acetoxy group in 10, lying outside the plane of the molecule and rotated by an angle of 77.6 degrees with respect to the tricyclic plane. The photoreaction of BQZ with DNA structure, as already known for psoralen and angelicin derivatives, consists of a [2 + 2] photocycloaddition reaction with the pyrimidine bases. The main photoadduct between the 2,3 double bond of I and the 5,6 double bond of thymine has been isolated and characterized by NMR, showing a cis-anti structure. Theoretical calculations, using AM1 Hamiltonian, have been carried out to describe the photocycloaddition reaction mechanism better. From a theoretical point of view, in the case of BQZ both the 1,2 or 2,3 double bonds and the 6,7 double bond may be involved in the [2 + 2] photocycloaddition. Spin densities and molecular orbital symmetries of compound I, in its triplet state, suggest that the 2,3 double bond interacts favorably with the 5,6 double bond of thymine moiety. On the contrary, the acetoxy substituent in position 10 of II seems to play a negative role in the DNA intercalation process.

Pyrroloquinolinone methylderivatives, furocoumarin analogues: interaction with biomolecules and computer-aided studies.

Pyrroloquinolinones, furocoumarin analogues, contain a divinilbenzene moiety, suggesting possible photoreactivity. Quantum mechanics calculations indicate that the pyrrole-side double bond exhibits strong photoreactivity, while the pyridone-side double bond is only poorly photoreactive. Intercalation models obtained by molecular mechanics calculations suggest that, in the cis-syn intercalation arrangement, the pyridone-side double bond is well aligned with the nearby thymine, supporting possible C4-cycloaddition with the 5,6 double bond of thymine, while the pyrrole-side double bond assumes an unfavourable position for photobinding. These data suggest that photoreaction between the pyridone-side and thymine double bonds may takes place, although with very low yield. Experimental evidence concerning DNA-photobinding exhibited by 2,6-dimethyl-9-methoxy-4H-pyrroloquinolinone (Compound I) confirms theoretical predictions. The formation of C4-cycloadducts between the pyridone side double bond and thymine also takes place with very low yield. Compound I shows marked BSA photobinding, suggesting that pyrroloquinolinones may photoreact with proteins. The three pyrroloquinolinones examined show high yields of singlet oxygen generation, suggesting that photobiological effects may be obtained through this photodynamic pathway, rather than through DNA photobinding.

Azapsoralens: new potential photochemotherapeutic agents for psoriasis.

New bioisoters of psoralen, obtained by replacing carbon 8 of the central benzene ring with a nitrogen, were studied from the photochemical, photobiological and phototherapeutic points of view. In particular, 4,4'-, 4',5'-dimetyl, 4,4',5'-trimethyl and 3,4,4',5'-tetramethylazapsoralen were studied. The crystal and molecular structure of 4,4',5'-trimethylazapsoralen, obtained by X ray diffraction, was also reported. Like psoralen, these compounds form a molecular complex with DNA, undergoing intercalation inside the double helix of the macromolecule. When irridiated with long ultraviolet light (365 nm), the intercalated drug photoconjugates covalently to the macromolecule, forming mono- and diadducts. The photobinding rate show the following order of magnitude: 4,4',5'-trimetylazapsoralen (4,4',5'-TMAP) = 3,4,4',5'-tetramethylazapsoralen (3,4,4',5'-TMAP) greater than 4',5'-dimethylazapsoralen (4',5'-DMAP) = 4,4'-dimethylazapsoralen (4,4'-DMAP). The DNA photobinding rate of 8-methoxypsoralen (8-MOP), taken as reference compound, is similar to that of the two dimetylazapsoralens but lower than tri- and tetramethyl derivatives. The ability of azapsoralens to form cross-links in DNA is lower than that of 8-MOP. However, capacity to induce cross-links does not parallel the DNA photobinding rate; it is higher for trimethyl derivate and lower for tetramethylazapsoralen. Azapsoralens show evident antiproliferative activity. The trimethyl derivative is the most active, followed by tetrametyl, both these compounds showing activity slightly higher than that of 8-MOP. The two dimethylderivatives are less active. The mautagenic activity of azapsoralens on E. coli WP2 TM6 is lower than that of 8-MOP in the same conditions. The new compounds do not show any skin phototoxicity on guinea pig skin. On the basis of its DNA photobinding, antiproliferative activity, mutagenicity and lack of skin phototoxicity, 4,4',5'-TMAP was chosen for clinical evaluation. Clinical results obtained by topical treatment of psoriatic plaques reveal evident therapeutic effectiveness and clearing is between good and moderate, although 8-MOP, used as reference compound, is more effective.

1-Thiopsoralen, a new photobiologically active heteropsoralen. Photophysical, photochemical and computer aided studies.

1-Thiopsoralen (7H-thieno[3,2-g]benzofuran-7-one) 1, a lead compound of a series of heteropsoralens, was investigated. The electronic transitions involved were studied. Fluorescence quantum yield is very low, while laser flash photolysis showed that the triplet state is practically the sole transient of 1. Fluorescence quantum yield (phi F) and triplet lifetime (tau F) as well as triplet quantum yield (phi T) and lifetime (tau T) were determined. The production of singlet oxygen was also evaluated by photophysical measurements. Photophysical data suggest that DNA photobinding of 1, owing to short fluorescence lifetime value and high triplet quantum yield, occurs likely through triplet mechanism. Interactions between 1 and DNA were studied both in the ground and the excited state. In the ground state 1 undergoes intercalation inside duplex DNA. This fact is also supported by molecular modeling studies. By UVA-light activation 1 photobinds covalently to DNA forming mono and diadducts. The furan side 1-thymine monoadduct, isolated from DNA photomodified by thiopsoralen, shows a cis-syn stereochemistry, in agreement with quantum mechanics studies. Compound 1 photobinds also with linolenic acid, component of lecithins, giving a C4-cycloaddition, and supporting that this compound also induces photolesions at the level of cell membrane, like psoralen. Compound 1 exhibits strong skin-phototoxicity.