Investigation of multiphoton-induced fluorescence from solutions of 5-hydroxytryptophan Roger H. Bisby,* a Morfakis Arvanitidis, a Stanley W. Botchway, b Ian P. Clark, b Anthony W. Parker b and Darren Tobin a a Biosciences Research Institute, University of Salford, Salford, UK M5 4WT. E-mail: r.h.bisby@salford.ac.uk b Lasers for Science Facility, CCLRC Rutherford Appleton Laboratory, Chilton, Didcot, Oxon, UK OX11 0QX Received 16th July 2002, Accepted 9th December 2002 First published as an Advance Article on the web 8th January 2003 It is reported (J. B. S. Shear, C. Xu and W. W. Webb, Photochem. Photobiol. 1997, 65, 931) that multiphoton near infrared excitation of 5-hydroxytryptophan results in a transient product with green fluorescence. Visible fluorescence from multiphoton excitation enables detection of 5-hydroxytryptophan with extremely high sensitivity and also has potential applications in imaging of biological systems and investigation of protein dynamics. The characteristic fluorescence at 500 nm has now also been observed in a two laser experiment whereby 308 nm photolysis of the solution is followed by an excitation step at 430 nm. Fluorescence was observed in aerated and deaerated solutions and in the presence of ascorbate. Enhancement of fluorescence was observed on addition of ethanol. Transient absorption experiments with 308 nm photolysis showed the formation of three transient species. In the presence of ascorbate the radical formed by photoionisation was quenched, revealing a long-lived species (τ > 1 ms) with a similar absorption spectrum, which is ascribed to the fluorescing species. Fluorescence induced by multiphoton excitation had a lifetime of 910 ± 10 ps and was also unaffected by ascorbate. In the presence of organic solvents there was an increase in fluorescence lifetime, but a decrease in overall fluorescence intensity. The fluorescence intensity and fluorescence lifetime both decreased in acidic solution (pH < 3). The results indicate that the fluorescence does not originate from the 5-indoxyl radical as previously suggested but from one or more other transient products which require further characterisation. Introduction Multiphoton excitation of chromophores in biological imag- ing 1,2 and quantitative fluorescence measurements 3 reduces background luminescence and has the potential to increase sensitivity and selectivity. Multiphoton excitation is usually used to excite the same fluorescent species as normally obtained under conventional one photon ultraviolet or visible illum- ination. In the case of 5-hydroxytryptamine (serotonin, 5HT), multiphoton excitation using the output of a titanium sapphire laser in the spectral region of 720 to 800 nm gives rise to an additional luminescent signal with a maximum at about 500 nm (described as ‘hyperluminescence’), compared with the normal UV-excited fluorescence of 5HT and 5-hydroxtrptophan (5HTrp) at around 340 nm. 4 This visible emission has allowed as little as 70 zmol of 5HTrp and 5HT to be detected. 5 In addition, the visible fluorescence from multiphoton excitation of proteins containing 5HTrp in place of tryptophan itself allows them to be imaged separately from other proteins. 6 The fluorescence lifetime of multiphoton-induced hyperlumines- cence from 5HT in aqueous solutions has been determined to be about 0.8 ns and the intensity of hyperluminescence shown to depend on oxygen concentration and the presence of buffer solutes. 7 The origins of the visible fluorescence of 5HT and 5HTrp are unknown. Webb and co-workers have shown that the emission from 5HT arises from an overall 6-photon process with 830 nm photons, and may be resolved into a 4-photon excitation of 5HT to the transient fluorescing intermediate, and a 2-photon excitation of fluorescence. 4 Luminescence in the region of 550 nm has also been reported from 5HT in acidic solutions. 8 Shear et al. 4 suggest that multiphoton-induced hyperluminescence from 5HT might arise from the 5-indoxyl radical, known to be formed by photoionisation of 5-hydroxyindoles by 248 nm laser flash photolysis, 9 and which has also been studied by pulse radiolysis. 10–13 In neutral solutions the 5-indoxyl radical has an absorption maximum at 410–420 nm, 10,13 consistent with two-photon excitation of hyperluminescence by ca. 800 nm laser pulses. In the present experiments we have attempted to charac- terise the fluorescence from photo-excited 5HTrp with both nanosecond ultraviolet laser excitation and on multiphoton excitation with femtosecond near infrared laser excitation. The results suggest that the fluorescence does not arise from the 5-indoxyl radical. Materials and methods 5-Hydroxytryptophan and other chemicals were obtained from Sigma-Aldrich and used as received. Solutions were prepared in water purified by ion-exchange and filtration in a Barnstead unit, and adjusted to the indicated pH values with phosphate buffer or HCl. The fluorescence induced by nanosecond laser excitation of 5HTrp was measured using a pump-probe technique with the same apparatus as previously employed in nanosecond time- resolved resonance Raman experiments. 14 Photochemically- generated transients in 5HTrp solutions were produced by a 308 nm pump pulse (ca. 1 mJ, 10 ns, 10 Hz) from a XeCl laser (Lumonics). This was followed after a set time interval by a 430 nm probe laser pulse (ca. 1 mJ, 5 ns) from a dye laser (Con- tinuum YAG-pumped Sirah). The two laser beams were focused colinearly to ca.100 μm diameter at the sample. Luminescence excited by both lasers was dispersed by a Spex Triplemate DOI: 10.1039/b206848f Photochem. Photobiol. Sci., 2003, 2, 157–162 157 This journal is © The Royal Society of Chemistry and Owner Societies 2003