emple Chemistry Department Philadelphia, PA www.chem.temple.edu Biological Photochemistry: The fate of electronic excited states in proteins, DNA, and the role of quenching Robert J. Stanley DOE Workshop on Aqueous Scintillators January 19, 2010 Electronic excited states in Biology • Chemiluminescence – Bioluminescence – charge transfer? radicals? • Photoinduced electron transfer – Photosynthesis – DNA repair • Photochemistry – DNA damage – photosensors DNA…a polymer of nucleotides connected by phosphodiester linkages 5’ Nucleic acid bases A, T, C, & G 3’ Voet and Voet, Biochemistry, 2nd Ed. Wiley, New York, 1995 B-DNA is double-stranded (ds) DNA, forming the famous double helix (1954 - Watson, Crick, Franklin) Watson-Crick base pairing (complementarity) DNA absorbs UV radiation * transition Absorbance (corrected) 0.8 5'-CTCCPACTTGC-3' 5'-GCAAGTTGGAG-3' dsDNA 0.6 0.4 0.2 P=6MAP 0.0 240 260 280 300 320 340 Wavelength (nm) 360 380 400 Quenching of excited states can be desirous or devastating in living systems: DNA • UV light absorbed by DNA is rapidly transformed into heat – Conical intersections in the potential surfaces of excited and ground state nucleic acid bases leads to ultrafast degradation of light into heat (10-12 sec.) …GOOD! • Excited native DNA bases (Guanine, Adenine, Thymine, Cytosine) can be either excited state donors or acceptors – – – – sequence dependent reaction *G8-oxo-G T-T T<>T pyrimidine dimerization Cancer, apoptosis…BAD UV light damages DNA Bad photochemistry 2+2 photo-cycloaddition O O O O CH 3 CH 3 H 3 C HN O NH N N T-T h O < 320 nm CH 3 HN O NH N N T<>T or CPD O If DNA damage is left unrepaired then mutations, cell death, and cancer can develop http://toms.gsfc.nasa.gov/ery_uv/euv.html Pathways involving energy transfer D = G*, A*, C*, T* D*A Bright Dark Bright or Dark A = G, A, C, T ISC 3 D * A DA hD DA* D 3 A or D 1 A hA Triplet Energy Transfer Förster or Dexter Transfer (singlets) DA Fluorescence “Structural” quenching pathways D*A Bright Dark DhotA hD Intramolecular vibrational relaxation Conical Intersection DA Fluorescence Pathways involving electron transfer D*A Bright Dark Bright or Dark D A or D A hD D A or D A hEX? Photoinduced Electron Transfer (PET) Exciplex (EX) formation (charge transfer) DA Fluorescence Enzymatic Repair of CPDs by DNA Photolyase uses blue light as an energy source (Good photochemistry) • Repair of the thymidines is direct: T<>T T-T without modifying the DNA backbone • Wide spread: E. coli, Frogs, Rice, Kangaroos…Humans (no!) Possible Applications: • Photosomes® (AGI Dermatics) • transgenic crops (wheat?) Mees, A., et al (2004) Science 306, 1789-1793. Sancar, A. Structure and function of DNA photolyase. Biochemistry 33, 2-9 (1994). DNA Photolyase (PL) is a flavoprotein (Vitamin B2) that binds and repairs CPDs • PL functions efficiently with only FAD (required for repair and binding • PL binds the CPD with high affinity (no light required): KA = 109 M-1 for dsDNA with CPD Park, H.-W., Kim, S.-T., Sancar, A., and Deisenhofer, J. (1995) Science 268, 1866-72. Flavin Structure and Oxidation States • Flavins can transfer 1 or 2 electrons (unlike nicotinamide) and are used in a large number of redox reactions in the cell FADH— — Biochemistry 2nd Ed., Voet and Voet, J. Wiley & Sons • Surprisingly, flavins are a major biological chromophore (DNA repair, circadian rhythms, phototropism, etc.) Photolyase functions by Photoinduced Electron Transfer from the FAD to the CPD • A large separation between the FADH- and the CPD (~16 Å) would give a slow electron transfer rate (keT, from Marcus theory) There’s a cavity in the protein keT e r 2 e ( G ) 2 / 4kT Orbital overlap x Driving force FAD • Slow electron transfer would compete poorly with 1FADH— deactivation (about 5 ns) but repair > 0.7! What happens to substrate conformation upon binding to Photolyase? Minor disruption AA Photolyase Moderate disruption Base Flipping T<>T Severe disruption Fluorescent reporter approach to probing double helical structure 5’-probe approach: 5’ 3’ Base Flipping 5’ 3’ 5’ 3’ 3’ 5’ 3’-probe approach: 5’ 3’ Base Flipping 5’ 3’ 5’ 3’ The fluorescence quantum yield of the reporter decreases when base stacked…but why? 3’ 5’ 6MAP is an attractive new fluorescent adenosine analogue 4-amino-6-methyl-8-(2-deoxy--D-ribofuranosyl)-7(8H)-pteridone CH3 O H H N H3C N 6 N H 3 C 1' N O 7 2 8 O N R N 6MAP Thymine Properties:1 fl = 0.2 ex = 330 nm ( ~ 8,500 M-1cm-1) em= 430 nm (large Stokes shift) et al, “Synthesis and Fluorescence Characterization of Pteridine Adenosine Nucleoside Analogs for DNA Incorporation.” Anal. Biochem.298, 231-240 (2001). Absorbance (corrected) N 4 0.03 ss-3'-6MAP ds-3'-6MAP-TT ds-3'-6MAP-CPD ss-5'-6MAP ds-5'-6MAP-TT ds-5'-6MAP-CPD 0.02 0.01 0.00 310 320 330 340 350 360 370 380 390 400 410 Wavelength (nm) 1Hawkins, K. Yang, S. Matsika, and R.J. Stanley, Biochemistry 2007 5’-GCAAGTTGGAG-3’ 3’-CGTTCAFCCTC-5’ 3.5x10 5 5'-6MAP 5'-6MAP/T<>T 5'-6MAP/T-T a) Fluorescence (corrected) Base flipping of the CPD monitored by 6MAP 3.0x10 5 2.5x10 5 2.0x10 5 1.5x10 5 1.0x10 5 5.0x10 4 5'-6MAP/PLox b) 5'-6MAP/T<>T/PLox 5'-6MAP/T-T/PLox +PL -PL Base Flipping 0.0 5’-GCAAGTTGGAG-3’ 3’-CGTTCFACCTC-5’ Why is the intensity pattern sequence-dependent? Fluorescence (corrected) 350 6 2.0x10 1.5x10 400 450 a) 3'-6MAP 500 550 350 400 3'-6MAP/T<>T Wavelength b)(nm) 3'-6MAP/T-T 450 3'-6MAP/PL 500 550 ox 3'-6MAP/T<>T/PLox 3'-6MAP/T-T/PLox 6 +PL -PL 1.0x10 6 5.0x10 5 Base Flipping 0.0 350 400 450 500 550 350 400 Wavelength (nm) 450 500 550 These data are consistent with disruption of base stacking due to base flipping of the CPD by Photolyase Photolyase ? Mees et al, Science v. 306, 1789-1793 (2004) Is the fluorescence quantum yield modulation of 6MAP due to PET? AMP DMF GMP CMP TMP 2.0 ACN 6MAP -15 -25 6MAP 0 Current ( A) I0/I Stern-Volmer quenching of 0 1.5 6MAP by G,A,C, and T: 15 -15 DMAP what is the rate of quenching, 1.0 0 kq? 15 -15 submitted to Biochemistry -15 6MI 6MAP 0 0 -5 10 -425 10 -3 10 -2 10 10 -1 0 25 DMF -25 0 25 15 -15 0 -25 ACN 15 nt ( A) What are the redox potentials? Cyclic voltammetry of 6MAP in aprotic organic solvents -25 Quencher Concentration (M) 3MI 0 25 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -25 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 DMAP 0 Volts (NHE) 25 The quenching of 6MAP* proceeds through nucleobase oxidation: 6MAP*:NMP6MAP¯:NMP+ 9 7x10 9 6x10 FBA 9 5x10 kq (M s ) -1 -1 9 4x10 AMP GMP TMP NB GET(eV) Eact(eV) G -0.63 0.000 A -0.16 0.003 C 0.021 0.048 dT -0.009 0.032 9 3x10 6MAP 9 2x10 CMP (Scandola-Balzani relation) 9 10 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 G°NBO (eV) submitted to Biochemistry 0.1 0.2 What’s the mechanism for base analog quenching? Pathways for energy transduction in a model FBA oligo 5’-NF*N-3’ h Conical Intersection -1 0.5 0.0 (M cm ) Bright Dark -1 1.0 2AP 3'-CCC2APGC-5' 2AP+4C+G T = 77K =55 -0.5 280 5’-NF +N--3’ 300 320 340 360 380 Wavelength (nm) Absorption Stark spectra of ssDNA with 2AP (), a hexamer with 2AP () , and a mix of the individual bases (). Fluorescence Exciplex Stark and MRCI calculations (Matsika) and (norm.) 5’-NFN-3’ 1.2 1.0 0.8 0.6 0.4 0.2 0.0 -0.2 -0.4 -0.6 6-MI 300 IF and IF (norm.) Photoinduced Electron Transfer 350 400 450 Wavelength (nm) Stark absorption and emission spectra of 6-MI (), a guanine analog, compared with their absorption and emission spectra (). Another possibility: 6MAP emission overlaps the absorption of the FAD: FRET from 6MAP*FAD? 1800000 ss-6MAP/PLox 1600000 ss-6MAP/TT/PLox 1400000 ss-6MAP/T<>T/PLox 1200000 1000000 800000 600000 400000 200000 0 380 400 420 440 460 480 500 520 540 Wavelength (nm) Yang et al, JPC B (2007) Fluorescence Energy Transfer Efficiency ET 6 R0 6 6 R0 rDA R0 the Förster distance where ET = 0.5 rDA the distance between a donor (fluorescent analogue) and an acceptor (FAD in photolyase) The Förster distance R0 (Å) = 0.211 ( n D J ) 2 4 1/ 6 2 : the orientation factor; n : the refractive index of the medium; D : the fluorescence quantum yield of the donor; J : the overlap integral. The Overlap Integral F J D ( ) A ( ) d 4 F D ( )d A FD 350 400 450 500 Wavelength (nm) FD(): the fluorescence intensity of the donor as a function of wavelength. εA(): the molar extinction coefficient of the acceptor at that wavelength; The Orientation Factor (cos T 3 cos D cos A ) 2 2 O H2N mD N 4 3 N 2 6 H3C 1 N R 10a rDA O N 10 9 4 9a 4a 8 N H3C 5a R 6MAP in 3'-6MAP NH N N 8 7 2 O 5 7 6 H3C FADox in Photolyase θT: mD, mA θD: mD , rDA θA: mA, rDA mA The transition dipole moment direction 6MAP was calculated from TD-DFT Yang et al, JPC B (2007) Orientation factors and ET between Probes and FADox From the crystal structure, lit. and TDDFT calcs experiment crystal structure Yang et al, JPC B (2007) FRET efficiency vs. orientation 1.0 xtal ( 6 MAP ) 0.8 0.6 FRET 0.4 3'-6MAP/FAD (m1) 0.2 5'-6MAP/FAD (m1) 0.0 60 70 80 90 (deg.) Yang et al, JPC B (2007) 100 110 I [I(6MAP/T<>T/PL)-I(6MAP/T<>T)] NO FRET! • The FAD is quenched 100x in the protein (acceptor is dark) 0.04 • A work-around : timeresolved FRET? 0.02 0.00 • Quenching mechanism is different for the two probes -0.02 -0.04 400 450 500 550 Wavelength (nm) 600 • photoinduced electron transfer vs. ultrafast internal conversion? • Does FAD* undergo PET to tryptophan??? Yang et al, JPC B (2007) Can we identify the kinetics and mechanism of repair? Two color pump probe femtosecond spectroscopy: 2 1PL red • What is the electron transfer lifetime (eT)? : T<>T • Does repair proceed by a concerted or sequential mechanism? eT PLsq• : T<>T • kic, krad c 1 3 PLsq• : T|_|T • h 2 4 krec kbeT kdiss PLred : T-T PLred + T-T 1 PLred : T<>T 7 PLsq• : T-T • 5 6 MacFarlane and Stanley (2003) Biochemistry 42, 8558-8568 Transient absorption measurement layout F1 BBO M9 Mode and wavelength monitor Laser control ISO Ti:sapphire M6 M10 M11 M13 P1 W1 M2 M3 M W2 M8 W3 B1 F2 L4 CaF2 L3 Delay stage M12 controller M7 L1 M4 YLF laser CW Nd:YAG L5 Ti:Sapphire amplifier L6 M15 M5 M14 Sample Chopper Controller Synchronization Delay Generator L7 L8 Monochromator L2 M1 CCD PET to the CPD substrate quenches the FADH excited state in ~ 30 ps 0.003 0.002 fl _ 1:5 PLred :(T<>T)5 A265 0.001 krad krad k ET (3 ns) 1 ~ 0.01 1 1 (3 ns) (0.032 ns) 0.000 -0.001 -0.002 _ PLred eT 32 20 ps -0.003 0 0 1000 20 40 2000 60 80 100 3000 Time (ps) MacFarlane and Stanley (2003) Biochemistry 42, 8558-8568 What’s are the intermediates? A unidirectional sequential model: 1PL A(,t) = ci(t)i() = C(E - 0) red : T<>T 2 keT PLsq• : T<>T • where Ei() = True spectra of the intermediates 0() = Ground state absorption spectrum • Construct C(t) = C0eKt (from the K matrix) 3 krepair h krad PLsq• : T-T • 4 hv hv hv k et K 0 k et 0 0 • • Calculate Minimize 0 0 k repair 0 k rec 0 0 k rec krec 1 PLred : T<>T or T-T Ei () = C-1A(,t) {A(,t) – C(E- 0)} using K matrix Pl-red+(TTT<>TT) 0.03 0.025 delta A 0.02 0.015 0.01 0.005 0 0 1000 700 600 2000 500 3000 400 Time (ps) Wavelength (nm) Pl-red+(TTTTT) 0.035 0.03 0.025 delta A The broadband transient absorption data: 0.02 0.015 0.01 0.005 0 700 0 1000 600 2000 3000 Time (ps) 500 400 Wavelength (nm) Spectrotemporal intermediates in the repair reaction: E spectra 4.5 x 10 4 1PL Intermediate Spectra: PLred-CPD red : T<>T 53 ps 2 4 PLsq• : T<>T • Extinction (M-1 cm-1) 3.5 PLSQ 3 3 2.5 h 2 620 ps 540 ps PLsq• : T-T • 4 1.5 1 2753 ps 0.5 0 400 450 500 550 600 Wavelength (nm) 650 700 1 PLred : T<>T or T-T • Fitting the data does not rule out a sequential bond breaking mechanism... • More complicated kinetics cannot be ruled out! In conclusion… Quenching is a simple term for many possible mechanisms to shunt electronic energy in excited molecules D*A D A or D A h DA Bright Dark Bright or Dark D A or D A Photoinduced Electron Transfer (PET) Fluorescence A battery of approaches need to be used to explore all possible pathways The Charge Separation Investigation Team Madhavan Narayanan •Ultrafast spectroscopy •Protein Chemistry Dr. Zhanjia Hou •Ultrafast spectroscopy •Single molecule spectroscopy Goutham Kodali • Stark spectroscopy • Computational chemistry • “Vector dude” Dr. Alex MacFarlane IV •Ultrafast spectroscopy •Electric field effects Salim Siddiqui, M.D., Ph.D. •Stark spectroscopy •Computational chemistry The Group Gone, but not forgotten.. Funding NSF Molecular Biosciences, REU Petroleum Research Fund Collaborators Prof. Aziz Sancar (UNC) Mary Hawkins (NIH) Prof. Spiridoula Matsika A closer look at the damage… 5’-GCTTAATTCG-3’ 3’-CGAATTAAGC-5’ A A 5’ 3’ 2.4Å 1.9Å Watson-Crick base pairing is distorted Base stacking is weakened Crystal structure: Park et al, PNAS 99, 15965-15970 (2002). DNA Photolyase (PL) binds its CPD substrate by base flipping CPD Flavin Adenine Dinucleotide Mees, A., et al (2004) Science 306, 1789-1793. Spectral overlaps of probes and FAD S0S2 S0S1 0.8 0.6 Does FRET explain the intensityFADpattern (A) 3'-6map (D) difference? Normalized Absorbance Normalized Fluoresence 1.0 ox 0.4 0.2 0.0 350 400 Wavelength (nm) 450 500