Fluorescence Spectroscopy

 


Return to the Chemistry Home Page


Fluorescence

Schematic state energy level diagram

S is singlet and T is triplet. The S0 state is the ground state and the subscript numbers identify individual states.

Timescale

S0 Æ Sn absorption

Sn Æ S1 internal conversion (10-11 - 10-14 sec)

S1 Æ S0 + hn fluorescence (10-7 - 10-9 sec)

S1 Æ Tn intersystem crossing (10-8 sec)

S1 Æ S0 internal conversion (10-5 - 10-7 sec)

T1 Æ S0 + hn phosphorescence (10 - 10-3 sec)

T1 Æ S0 internal conversion (10 - 10-3 sec)

 

Characteristics of Excited States

  1. Energy
  2. Lifetime
  3. Quantum Yield
  4. Polarization

 

Phosphorescence occurs at longer wavelength than does fluorescence

Often, the emission band is red-shifted relative to the absorption band: "Stokes shift"

Excited states decay exponentially with time

I = I0e-t/t

I0 is the initial intensity at time zero,

I is the intensity at some later time t,

and t is the lifetime of the excited state.

Also, kF = 1/ t, where kF is the rate constant for fluorescence.

 

Quantum Yield = F

FF = number of fluorescence quanta emitted divided by number of quanta absorbed to a singlet excited state

FF = ratio of photons emitted to photons absorbed

Quantum yield is the ratio of photons emitted to photons absorbed by the system: FF = kF / kF + kISC + knr + kq + kr

 

Polarization

Molecule of interest is randomly oriented in a rigid matrix (organic solvent at low temperature or room temperature polymer). Plane polarized light is used as the excitation source.

The degree of polarization is defined as follows:

where I|| and I^ are the intensities of the observed parallel and perpendicular components, respectively. a is the angle between the emission and absorption transition moments. If a is 0° than P = +1/2, and if a is 90° than P = -1/3.

Figure 26, Becker, pg. 84

Polarization of fluorescence of phenol in propylene glycol at -70°C shows that the transition moments of the corresponding absorption bands are mutually perpendicular.

Phosphorescence is usually slow (seconds) therefore quenching by impurities including oxygen usually makes phosphorescence difficult to observe. Low temperature glasses and rigorous exclusion of oxygen is usually necessary to observe phosphorescence. Since this condition is not biological, fluorescence is the primary emission process of biological relevance.

Experimental Measurements

Steady-state measurements: F, I

Time-Resolved measurements: t

Emission spectra are obtained when the excitation monochrometer M1 is fixed and the emission monochrometer M2 is scanned.

If M2 is fixed and M1 is scanned the result is an excitation spectrum. Excitation and absorption spectra should be identical.

Relative quantum yields are determined by using a standard such as quinine sulfate in 1 N H2SO4 (fF = 0.70), or fluorescein in 0.1 N NaOH (fF = 0.93). The area under the emission band of the standard relative to the sample are compared. It is of course important that the absorption at lex are matched.

Excited-state decay rates can be measured by exciting the sample with a short pulse of light and monitoring the emission as a function of time.

Figure 8-14, Cantor & Schimmel, pg. 442.

Fluorescence decay of a pure sample showing a single exponential decay. The dark line shows the excitation pulse.Time correlated single photon counting was used to obtain this data. This technique counts the number of emitted photons hitting a detector at times, t, following excitation

One critical difference between steady-state and kinetic measurements of fluorescence is that the value of tF is not a function of concentration of the sample while the value of FF is concentration dependent. Only at low concentration is the value of FF linearly dependent on concentration. The reason is the so-called inner filter effect.

 

The inner filter effect:

At low concentration the emission of light is uniform from the front to the back of sample cuvette. At high concentration more light is emitted from the front than the back. Since emitted light only from the middle of the cuvette is detected the concentration must be low to assure accurate FF measurements.

Fluorescence characteristics of chromophores found in proteins and nucleic acids. Generally, quantum yields are low and lifetimes are short.

 

 

Absorption

Fluorescence

 

Sensitivity

Substance

Condition

lmax

(nm)

emax

10-3

lmax

(nm)

fF

tF

(nsec)

emaxfF

10-2

Tryptophan

H2O, pH 7

280

5.6

348

0.20

2.6

11

Tyrosine

H2O, pH 7

274

1.4

303

0.14

3.6

2.0

Phenylalanine

H2O, pH 7

257

0.2

282

0.04

6.4

0.08

Adenine

H2O, pH 7

260

13.4

321

2.6 10-4

<0.02

0.032

Guanine

H2O, pH 7

275

8.1

329

2.6 10-4

<0.02

0.024

Cytosine

H2O, pH 7

267

6.1

313

0.8 10-4

<0.02

0.005

Uracil

H2O, pH 7

260

9.5

308

0.4 10-4

<0.02

0.004

NADH

H2O, pH 7

340

6.2

470

0.019

0.40

1.2

Figure 8-15, Cantor & Schimmel, pg. 444

Fluorescence emission spectra of human serum albumin (solid line), tryptophan alone (dashed line), and an 18:1 molar ratio of tyrosine to tryptophan (gray line): Excitation at 245 nm.

The 18:1 sample approximates the relative occurrence of these amino acids in the protein. Note that the spectrum of the protein closely resembles that of pure tryptophan because tyrosine sensitivity is low and its emission is most likely quenched by tryptophan (via energy-transfer mechanism).

 

Commonly, fluorescent probe molecules are used to characterize protein and nucleic acids.

Sensitivity is higher.

lmax is also different from biomolecule so selective excitation is possible.

Fluorescence generally is much more sensitive to the environment of the chromophore than is light absorption. Therefore, fluorescence is an effective technique for following the binding of ligands or conformational changes.

The sensitivity of fluorescence is a consequence of the relatively long time a molecule stays in an excited singlet state before deexcitation. Absorption, or CD, is a process that is over in 10-15 sec. On this time scale, the molecule and its environment are effectively static. In contrast, during the 10-9 to 10-8 sec that a singlet remains excited, all kinds of processes can occur, including protonation or deprotonation reactions, solvent-cage relaxation, local conformational changes, and any processes coupled to translational or rotational motion.

A number of fluorescent molecules have a very convenient property

in aqueous solution their fluorescence is very strongly quenched, but in a nonpolar or a rigid environment (like in a protein or nucleic acid) a striking enhancement is observed.

Figure 8-17, Cantor & Schimmel, pg. 447.

In addition protein protects the probe from quenchers such as oxygen.

F0/F = fo/f = [kF + kIC + kISC + kq(Q)]/(kF + kIC + kISC) = 1 + kqt0(Q),

where F = fluorescence in the presence of quencher, F0 = fluorescence in the absence of quencher.

Therefore a plot of F0/F versus concentration of Q will yield a value for kq.

Quenching of tryptophan fluorescence by collision with oxygen:

Tryptophan: kq = 12 109 M-1 sec-1 (diffusion controlled)

Carbonic anhydrase: kq = 2.6 109 M-1 sec-1

 

Singlet-Singlet Energy Transfer

If the emission band of a molecule (D) coincides with the absorption band of another (A) two processes occur: emission of D is quenched, emission of A is sensitized.

Figure 8-18, Cantor & Schimmel, pg. 449

Förster theory can be used to determine distance between chromophores

The rate of energy transfer = kT = (1/tD)(R0/R)-6.

tD = lifetime of D in the absence of A.

R0 = characteristic transfer distance = 9.7 103 (J k2 n-4 fD)1/6 cm,

where J = Ú eA(n)fD(n)n-4 dv

J is a measure of the spectral overlap between donor emission and acceptor absorption (shaded region in figure). FD is the normalized fluorescence of the donor; n is the refractive index of the medium between donor and acceptor; fD is the quantum yield of donor in the absence of acceptor; and k2 is a complex geometric factor that depends on the orientation of donor and acceptor. If both donor and acceptor are free to tumble rapidly on the time scale of fluorescence emission, k2 approaches a limiting value of 2/3.

If the efficiency of energy transfer is expressed as E = kT/(kT + 1/tD) than

E = R06/(R06 + R6).

Figure 8-20, Cantor & Schimmel, pg. 455

For dansyl-(L-proline)n-a-naphthyl for n = 1 to 12. For this pair R0 = 50Å.

 

Energy transfer plays a large role in determining the emission spectrum of normal proteins.

The fluorescence of tyrosine is overlapped by the absorption of tryptophan. The R0 for this donor-acceptor pair is about 9 Å. This is short enough, given the average size of globular proteins, such that most tyrosines in proteins are quenched by singlet-singlet energy transfer by tryptophan.

The result is that observed emission comes from primarily tryptophan (see above).

Fluorescence Anisotropy/Polarization

Figure 16-14 and legend, pg. 529, Marchell

 

Steady-state experiment

continuous irradiation with polarized light of molecules in solution.

Polarization =

I|| and I^ are time-independent steady state values for fluorescent intensity polarized parallel and perpendicular.

Po is the maximum P which occurs when the rotational motion is very slow compared to the singlet excited state lifetime.

trot = rotational correlation time = the characteristic lifetime of rotational diffusion. For large proteins trot is large.

If trot << tF than the polarization, P, approaches zero (i.e., the steady-state fluorescence is completely depolarized so that by the time the fluorescence occurs, the direction of oscillation of the emission dipole is completely random).

The relationship between P and trot is:

trot is related to the rotational diffusion constant, Drot, according to:

h = solution viscosity

k, R = gas constant per molecule and per mole, respectively

r = molecular radius for spherical molecule

V = 4/3 pr3No = molar volume

T = temperature in °K

Therefore:

This is the Perrin Equation

If the temperature, T, is low or the viscosity, h, is large, than T/h approaches zero and molecular motion will be slowed down to the limit trot >> tF and the polarization, P, reaches its maximum value, Po.

The slope of the plot of versus will be which yields a value for V since tF is known. This in turn yields a value for trot and an estimate for the radius of the molecule (note: these equations are valid only for spherical molecules).

 

Perrin plot for human macroglobulin (MW = 900,000 Da) and two fragments (MW = 180,000 and 50,000 Da) yields values for trot.

The protein is first covalently labeled with dansyl chloride which has strong fluorescence when bound to proteins and a tF = 12 ns.

900 kDa => trot = 80 ns

180 kDa => trot = 69 ns

50 kDa => trot = 58 ns

If the 50 kDa and 900 kDa were both rigid molecules one would expect a 260% reduction in trot, but only get 25% reduction. Therefore, the 900 kDa protein must be highly flexible.

 

Time-resolved Fluorescence Depolarization (Anisotropy).

A short pulse of vertically polarized light is directed at the sample;

the light is absorbed, promoting the molecule to an excited singlet state;

following vibrational relaxation, light is emitted (fluorescence) at lower energy;

if the molecule rotates during the time interval between absorption and emission, there will be a decrease in the polarization with time at a rate that reflects the rate at which the molecule rotates diffusionally.

Fluorescence Anisotropy =

at times, t, after the light pulse is turned off.

The overall fluorescent intensity [I|||(t) + 2I^(t)] will decrease exponentially in time according to the lifetime, tF, of the excited singlet state.

The A(t) will decrease with a time constant, trot, which represents the time it takes for a molecule to rotate diffusionally, i.e., A(t) = Aoexp(-t/trot).

A dansyl-labeled protein can be used to determine directly the trot in a time-resolved experiment.

Figure 16-15, pg. 532, Marshell

(a) dansyl-labeled protein in membrane

(b) free dansyl-labeled protein

A(t) is bi-exponential, i.e., two trot’s.

In (a) trotA = 3 ns, trotB = 700 ns.

In (b) trotA = 3 ns, trotB = 45 ns.

The fast component is due to local flexibility at the site of attachment of the label to the protein, and the slow component corresponds to rotational diffusion of the whole protein molecule.

Clearly, in a membrane the protein is highly immobilized.

Rotational Correlation Times, trot, of Proteins, Determined Either from Experimentally Measured Fluorescence Depolarization Decay Rates or from Theoretical Models of the Protein as a Sphere in a Continuous Medium.

Protein

Molecular

Weight

trot (from

fluorescence

depolarization)

trot (from

)

trot(expt) / trot(calc)

Apomyoglobin

17,000

8.3 nsec

4.4 nsec

1.9

b-lactoglobulin

18,400

8.5

4.7

1.8

Trypsin

25,000

12.9

6.4

2.0

Chymotrypsin

25,000

15.1

6.6

2.3

Carbonic anhydrase

30,000

11.2

8.0

1.4

b-lactoglobulin (dimer)

36,000

20.3

9.7

2.1

Apoperoxidase

40,000

25.2

10.5

2.4

Serum albumin

66,000

41.7

17.4

2.4

 

Fluorescence depolarization studies of an antibody to which a fluorescent hapten was bound provided evidence for internal flexibility in the immunoglobulin molecule.

Antibodies were grown specific to a dansyl-hapten.

Depolarization studies of the hapten-antibody complex reveals two trot’s ( 33 and 168 ns).

The fast component may be due to flexibility in the "hinge" region that joins the fragments and the slow component is most likely due to rotation of the whole complex.