Fluorescence Correlation Spectroscopy and Frster Resonance Energy - - PowerPoint PPT Presentation

Fluorescence Correlation Spectroscopy and Förster Resonance Energy Transfer

Thorsten Wohland

Interaction of Light with Matter

luminescence

Internal Conversion

Ground State S0

Energy

Excited Singlet State S1 Triplet State T1

Absorption Fluorescence VR: vibrational relaxation IC: Internal Conversion VR VR IC

Intersystem Crossing

Ground State S0

Energy

Excited Singlet State S1 Triplet State T1

Absorption Fluorescence VR: vibrational relaxation ISC: Intersystem Crossing VR VR ISC Delayed Fluorescence Photochemical reaction Phosphorescence

Lifetimes, rate constants, and quantum yield

nr

k 1    

Excitation rate kex~I 

knr nr f

k     

Lifetime Quantum yield

Fluorescence Properties

• Wavelength (absorption and emission)
• Lifetime (of various states)
• Quantum yield
• Polarization

6

http://micro.magnet.fsu.edu

F-techniques

7

Fluorescence Anisotropy Fluorescence Lifetime Single Particle Tracking Fluorescence Recovery after Photobleaching

Förster Resonance Energy Transfer - FRET

8

The actual formula for the FRET rate

     

        d F n N r Q r k

A D A D D T 4 4 5 6 2

128 10 ln 9000 ) (





        

Note: Formula is now in SI units! 6

1 ) (        r R r k

D T



Spectral overlap

• rientation

distance Donor lifetime environment

T D T

k k E  

1



6 6 6

r R R E  

R0 is the so-called Förster radius. It is the distance at which a FRET pair exhibits 50%

• FRET. It is a constant for any FRET-pair.

Förster Distance Calculator

 Donor: Alexafluor 488 TFP (AF488) labelled protein layer  Acceptor: DiI labelled lipid bilayer

10

Far Donor – Acceptor

 Low FRET  High fluorescence intensity  High average donor lifetime

Near Donor – Acceptor

 High FRET  Lower donor fluorescence intensity  Lower average donor lifetime

FRET to monitor conformational changes of a virus

11

Lifetime experiment

5 10 15 20 1x105 2x105 Overall Decay

Time (ns) Intensity (cnts)

5 10 15 20 1x105 2x105 3x105 Overall Decay

Intensity (cnts) Time (ns)

5 10 15 20 100 200 300 400

Intensity (cnts) Time (ns)

ROI Decay Fitted Curve IRF 5 10 15 20 1x104 2x104 ROI Decay Fitted Curve IRF

Time (ns) Intensity (cnts)

30 60 90 120 150 180 0.0 0.2 0.4 0.6 0.8 1.0

Intensity (kcts) Time (s)

30 60 90 120 150 180 0.0 0.2 0.4 0.6 0.8 1.0

Time (s) Intensity (kcts) Intensi

25˚C 37˚C

5 10 15 0.0 0.2 0.4 0.6 0.8 1.0

Intensity (a.u.)

IRF Trace at 25 C Trace at 37 C

Time (ns)

Higher FRET at 25˚C

2.2 2.4 2.6 2.8 3.0 3.2 37 C

avg (ns) Temperature

DV2 at 25 C DV2 at 37 C 25 C 0.55 0.60 0.65 0.70 0.75

Low FRET population

37 C

Temperature

25 C 0.25 0.30 0.35 0.40 0.45

2 (High FRET population)

𝜐𝐸 = 1 Γ + 𝑙𝑜𝑠 + 𝑙𝑈

12

24 26 28 30 32 34 36 38 2.4 2.6 2.8 3.0 3.2 3.4 3.6

25C to 37C (Donor only) 25C to 37C (Dual lablelled)

avg (ns) Temperature (C)

DV2 (NGC) transition vs temperature in absence of MgCl2

37C to 25C (Dual lablelled)

24 26 28 30 32 34 36 38 0.20 0.25 0.30 0.35 0.40 0.55 0.60 0.65 0.70 0.75 DV2 (NGC) transition vs temperature in absence of MgCl2

25C to 37C

a1(low FRET population) Temperature (C) a2(High FRET population)

37C to 25C

25 to 37ºC

Temperature dependence of lipid bilayer-protein coat distance

13

State 1 State 2

Imaged under TIRF microscope

Single particle spectroscopy

Mg2

+

14

Donor Acceptor Donor Acceptor

iSMS

Overlay

Single particle spectroscopy

15

20 40 60 80 100 3 4 5

Donor

Background Donor Intensity

Time(s)

20 40 60 80 100 5.0 5.5 6.0 6.5 7.0 7.5

Acceptor

Background Acceptor Intensity

Time(s)

20 40 60 80 100 0.2 0.4 0.6

Overlay Time(s)

Acceptor intensity Donor Intensity 20 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0

Time(s) EFRET

20 40 60 80 100 3 4 5 6 7

Time(s)

20 40 60 80 100 5 6 7 8 9

Time(s)

20 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0

Time(s)

20 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0

Time(s)

20 40 60 80 100 3 4 5

Time(s) Donor

20 40 60 80 100 5.0 5.5 6.0 6.5 7.0 7.5

Time(s) Acceptor

20 40 60 80 100 0.2 0.4 0.6

Time(s)

20 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0

Time(s)

Molecule 1 Molecule 2 Molecule 3

Single particle spectroscopy

16

Single particle spectroscopy

(ref: Sean A. McKinney1 et. al. 2002)

1 2 3 4 400 800 1200 1600 Dwell time histogram of HJ from Closed to open state

Counts Time(s) k21Close  Open= 4.3 0.1 s-1 n = 29

1 2 3 4 500 1000 1500 2000 2500 Dwell time histogram of HJ from Open to closed state

Counts Time(s) k21Open  Closed= 3.4 0.1 s-1 n = 29

• 0.2

0.0 0.2 0.4 0.6 0.8 1.0 1.2 200 400 600 800 1000 1200 1400 1600 FRET Efficiency population Graph

Observations FRET Efficiency (E) k12 k21

20 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0

EFRET Time(s)

HMM Fitting

20 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0

Time(s)

20 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0

Time(s)

Summary 1

• FRET can measure distances in the range of

~10 nm

• It can be measured either be observing the

emission wavelength or best by lifetimes

• It can be done in ensembles or on a single

molecule level

• Here we demonstrated its application to viral

conformations and holiday junction dynamics

17

Fluorescence Correlation Spectroscopy (FCS)

• What are fluctuations?
• What are correlations?
• How to calculate correlations?
• Fluorescence Correlation Spectroscopy (FCS)
• FRET-FCS
• Fluorescence Cross-Correlation Spectroscopy (FCCS)
• Imaging FCS/FCCS

Fluctuations

19

A + B AB

Time [AB] kinetics equilibrium fluctuations fluctuations

FCS: General idea

• What is a correlation
• Predicting the future
• Self-similarity

Correlations

1  g 1  g

b a b a  

b a b a g  

1  g

No correlation Anti-correlation Correlation

• X. Shi and T. Wohland, “Fluorescence Correlation

Spectroscopy”, in Nanoscopy, CRC Press, 2010

Example

22

0.25 0.25 0.25 0.25 Probability <A> = 0.5 <B> = 0.5 <AB> = 0.25 A B <AB> <A><B> = 1

1 1 1 1

Example

23

0.3 0.2 0.2 0.3 Probability <A> = 0.5 <B> = 0.5 <AB> = 0.2 A B <AB> <A><B> = 0.8

Example

24

0.1 0.4 0.4 0.1 Probability <A> = 0.5 <B> = 0.5 <AB> = 0.4 A B <AB> <A><B> = 1.6

Correlations

• 1. Correlated variables

a 1 1 0 1 0 0 1 1 1 0 1 0 1 0 1 1 0 0 0 0 b 1 1 0 1 0 0 1 1 1 0 1 0 1 0 1 1 0 0 0 0 b a b a  

2 1

 b a ;

4 1 2 1 2 1

   b a

• 2. Anticorrelated variables

a 1 1 0 0 1 0 1 0 1 1 1 0 1 0 0 0 1 0 1 0 b 0 0 1 1 0 1 0 1 0 0 0 1 0 1 1 1 0 1 0 1 b a b a    b a ;

4 1 2 1 2 1

   b a

• 3. Uncorrelated variables

a 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 b 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 b a b a  

2 1

 b a ;

2 1 2 1

1    b a

Autocorrelations

       

t a t a t a t a  

       

      t a t a t a t a

               

2

t F t F t F t F t F t F t F G         

         

       t a t a t a t a G

Stationary Processes

Short time shifts 

Blue: F(t) Yellow: F(t+) Time

       

     t F t F t F t F ?

   

t F t F 

   

1

   t F t F 

 

2

   t F t F 

 

3

   t F t F

The intensity peaks always overlap to some extent and thus

       

      t F t F t F t F

Long time shifts 

Blue: F(t) Yellow: F(t+) Time

       

     t F t F t F t F ?

   

t F t F 

   

1

   t F t F 

 

2

   t F t F 

 

3

   t F t F

The intensity trace contains a random pattern of intensity peaks. Therefore an

• verlap of all/many peaks is only achievable at short times.

       

      t F t F t F t F

Periodic signals

Blue: F(t) Yellow: F(t+) Time

       

     t F t F t F t F ?

   

t F t F 

   

1

   t F t F 

 

2

   t F t F 

 

3

   t F t F

The intensity trace contains a regular pattern of intensity peaks (i.e. it is repeated). Therefore an overlap of all/many peaks is achievable periodically and the correlation function will show that periodicity.

ACF: Autocorrelation Function (the correlation of a variable with itself) CCF: Cross- correlation Function (the correlation between two variables)

How is an ACF calculated practically?

Intensity values recorded every nanosecond … To calculate the correlation for the range

• f seconds you would need 1 billion

values … … … … If we make the time bins larger then we lose the information at short times.

So best would be to use a varying time scheme.

• T. Wohland, R.Rigler, H. Vogel, Biophys. J. 2001, 80(6), 2987-2999.

Correlation Time Schemes

The typical scheme used is called the semi-logarithmic time scale. The first n channels have a time D. The second group contains n/2 channels with 2 D. The next group n/2 channels with 4 D.

D

n=16

2D

n=8

4D

n=8

…

1) Each time a new measurement of length D comes in, calculate all ACF values for lag times 0 to 16D. 2) After 2 measurements of D,correlate the last two newest measurements with all channels in group 2. Then take the last two channels of group 1 and combine them into one channel with width 2D

• f group 2 and shift.

. . . . . .

Confocal FCS setup

33 Thompson, Topics in Fluorescence Spectroscopy Techniques vol 1 (1991) Haustein and Schwille, Biophysics Textbook Online Krichevsky, Bonnet, Rep. Prog. Phys. 65, 251–297 (2002)

FCS: Characteristic Parameters

2.0 1.8 1.6 1.4 1.2 1.0 10

• 6

10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

Time G(t)

34

~F(N) ~F(1)

2.0 1.8 1.6 1.4 1.2 1.0 10

• 6

10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

G(t) Time 3 D

M  

 

N G 1 ~

Correlation Functions

36

3 2 2 2

4

eff D

N C V C w z w D z K w      

 

1 2 1 2 1 2 3 2 2 2 2 2

1 4 4 4 1 1 1 1 D D D G C w z w w z     

  

                      

 

1 1 2 2

1 1 1

D D

G G N K     

  

               

Number of particles Correlation time Structure factor

FCS

Parameters: Width, Amplitude, Shape Amplitude: concentration Accessible range: 50 pM to 1 mM Width: characteristic time Time scales accessible lie between: 1 ns and 1 s

• U. Meseth, T. Wohland, R. Rigler, H. Vogel, Biophys. J. 1999, 76(3), p. 1619-1631.
• T. Wohland, R.Rigler, H. Vogel, Biophys. J. 2001, 80(6), 2987-2999.

Shape: Type of process

Live cell measurements

Differences in correlation width

38

Experimental process

Construction: EGFR-GFP, EGFR-YFP, EGFR-mRFP FCS,FCCS setup

Transfection Calibration Z-scan EGFR-mRFP Free mRFP

Liu et al. Biophys. J. (93): 684-698 (2007).

Comparison of cytosolic free FPs and membrane fusion EGFR-FPs

1.4 1.3 1.2 1.1 1.0 G() 10

• 6

10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

correlation time (s)

Triplet state at microsecond Photodynamics at hundred microseconds Diffusion of Cytoplasmic GFP 1 milisecond

1.15 1.10 1.05 1.00 G( 10

• 6

10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

correlation time (s)

Triplet state at microsecond Photodynamics at hundred microseconds Diffusion of membrane- localised EGFR-GFP at tens of miliseconds

EGFR-GFP GFP

Lignad-Receptor Binding

How to use amplitude and width of and autocorrelation fucntion

41

Example: The 5HT3 Receptor

a) Determination of binding constants b) And stoichionetry …

The ligand GR65630 was labeled with Cy5

Stoichiometry of ligand binding

Ligand-Receptor Interactions

Ligands: 0.5 – 1.1 kDa C12E9 micelle: 60 - 70 kDa 5HT3As-R + micelle: ~320 kDa

nM K RBA

d

. 2 . 18  

GR-Cy5:

nM K FCS

d

. 8 7 . 15  

Mass: 500±300 kDa Stoichiometry: 1:1

FRET-FCS

Other fluctuations …

45

46

FRET-FCS

Proximity ratio calculated

p =

𝑱𝑩 𝑱𝑩+𝑱𝑬

Gp Gp(t) (t) =

〈δ𝑸 𝒖𝟏 ∗δ𝑸 𝒖𝟏+𝒖 〉 〈𝑸 𝒖𝟏 〉𝟑

ACF of Proximity ratio:

1E-5 1E-4 0.001 0.01 0.1 1 0.00 0.02 0.04 0.06 0.08

Correlation time (s) Gp () Effective relatxation time fit for Hairpin at 50 C

tau 3.72228E-5 1.51036E-5 B 0.26418 0.01835 Gp 0.16887 0.01901

1E-5 1E-4 0.001 0.01 0.1 1 0.00 0.02 0.04 0.06 0.08

Effective relatxation time fit for Hairpin at 25 C Gp () Correlation time (s)

tau 7.00066E-5 1.73267E-5 B 0.35928 0.02305 Gp 0.1396 0.01224

25 degree = 70 µs 50 degree = 37 µs Fitting using Stretched Exponential Function Fluorescence acquisition in donor and acceptor channels

47

24 26 28 30 32 34 36 38 40 42 0.8 1.2 1.6 2.0 2.4 Temperature dependence of DV2 Effective relaxation time

Effective Relaxation Time () (ms) Temperature (C)

25C to 40C 40C to 25C

1E-5 1E-4 1E-3 0.01 0.1 1 2 3 4 5 6 7 DV2 (NGC) Proximity ratio curves

25C 40C 25C Reverse

Gp () Correlation time (s)

Dengue Virus 2 envelop dynamics

Summary 2

• Autocorrelation functions provide a measure for

the self-similarity of a signal

• For a signal in time it provides the capability to

predict the future (statistically)

• FCS provides information on dynamics of

processes

• In the case of diffusion it provides diffusion

coefficients and concentrations

• But any fluctuations can be measured and

correlated (e.g. FRET-FCS)

48

Fluorescence Cross-correlation Spectroscopy (FCCS)

2.2 2.0 1.8 1.6 1.4 1.2 1.0 G() 10

• 6

10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

10 [s] 2.2 2.0 1.8 1.6 1.4 1.2 1.0 G() 10

• 6

10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

10 [s]

Fluorescence Cross-correlation Spectroscopy (FCCS)

2.2 2.0 1.8 1.6 1.4 1.2 1.0 G() 10

• 6

10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

10 [s]

Green (G, GR) + crosstalk (R) + background Red (R, GR) + crosstalk (G) + background GreenRed (GR) + crosstalk (G, R) + background

SW-FCCS

51 Ricka and Binkert, Phys Rev A, 39(5) :2646-52 (1989) Hwang and Wohland, ChemPhysChem 5, 549-551 (2004) Hwang and Wohland, JCP, 122, 114708 (2005)

Fluorophores: Quantum dots Tandem dyes (energy transfer dyes) Organic dyes Fluorescent proteins

~2000 counts per second and particle

52

How to determine the Kd

    

d

G R K GR 

    

d

G R K GR 

Line through origin with a slope of Kd

     

G R GR  

Normalized Frequency 0.15 0.10 0.05 0.00

ln KD (ln nM)

8 7 6 5 4 3 2

[G]·[R] [GR]

Kd distribution

53

Cdc42 and IQGAP1

GTP: guanosine triphosphate; GDP: guanosine diphosphate; GEF: guanine nucleotide exchange factors; GAP: GTPase-activating proteins; GDI: guanine nucleotide exchange inhibitors.

Cdc42 is a regulator of membrane trafficking and cytokeletal organization IQGAP1 is involved in regulation

• f cell motility and morphology

and is supposed to bind the active GTP bound form of Cdc42 Dominant negative (GDP bound) Constitutively active (GTP bound)

54

SW-FCCS in zebrafish embryos

Cytosolic protein interaction (cdc42/IQGAP1)

Examples of Applications

55

EGFR activation and signaling

Liu et al., Biophys. J. (93): 684-698 (2007). Ma et al Front. Biosci. Jan 1;3:22-32 (2011). Shi et al., Biophys. J. (97)2:678-686 (2009). Sudhaharan et al., JBC 284: 13602-13609 (2009).

Ligand-receptor binding (Nodal Factor binding to activin receptor II)

Wang et al. (eLife, 2016)

IMAGING FCCS

56

57

FCS in a confocal system

 [s] G() 1) Measurements are not simultaneous 2) Cell damage by long illumination times

Alternative illumination schemes

58

TIR – Total Internal reflection VAI – Variable Angle Illumination SPIM – Single Plane Illumination Microscopy The z-sectioning of the illumination together with the xy-sectioning provided by the pixels of a camera define multiple observation volumes.

Kannan et al., Anal. Chem. (79): 4463-4470 (2007). Sankaran et al., Biophys. J. (97): 2630-2639 (2009).

Imaging FCS

59

2.0 1.8 1.6 1.4 1.2 1.0 10

• 6

10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

Time G(τ)

Diffusion/mobility

2.0 1.8 1.6 1.4 1.2 1.0 10

• 6

10

• 5

10

• 4

10

• 3

10

• 2

10

• 1

G(τ) Time

Concentration D N

Examples

60

DLPC/DSPC bilayer on glass GFP-GPI on SH-SY5Y cells

Bag et al. Methods Appl. Fluoresc. 4 (2016) 034003

Imaging Fluorescence Cross- Correlation Spectroscopy

(TIRF illumination)

Auto- (red, green) and Cross- (blue) correlation functions

545/35 nm 615/45 nm

Neuroblastoma cell labeled with DiI-C18 (pos. control). 514 nm, 300 mW excitation.

Cross- correlation functions 61

527-563 nm 592-638 nm

Imaging FCCS on EGFR

62

Degree of dimerization

q = GGR(0)/Min{GG(0), GR(0)}

Acknowledgements

63

Group members

Sun Guangyu* Wang Xi* Ng Xue Wen Anand Paratap Singh* Nirmalya Bag* Radek Machan*

Collaborators

Vladimir Korzh and Cathleen The (IMCB) Karuna Sampath (TLL, U of Warwick) Christoph Winkler (NUS) Jörg Langowski and Jan Krieger (DKFZ) Timothy Saunders (NUS)

Biomedical Research Council (BMRC) Science and Engineering Research Council (SERC) Singapore Bioimaging Consortium (SBIC) Singapore Stem Cell Consortium (SSCC) Academic Research Fund (ARF) Baden- Wuerttemberg- Singapore LSI initiative

Jagadish Sankaran Kamal Kant Sharma Huang Shuangru* Sibel Javas* Angela Koh Andreas Karampatzakis Sarala N. Tantirimudalige Jonathan Foo Sapthaswaran Veerapathiran Anjali Gupta