FD-based AFM: The tool to image and simultaneously map multiple properties of biological systems

Transcription

1 force-distance curve-based atomic force microscopy FD-based AFM: The tool to image and simultaneously map multiple properties of biological systems Technical Journal Club 1. Sept 215 Valeria Eckhardt

2 Overview 1. Atomic force microscopy (AFM) Brief intro AFM Imaging mode: contact / oscillation Force spectroscopy mode 2. FD-based AFM 3. Applications FD-based AFM Paper 1: Native proteins Paper 2: GPCR-ligand interaction 4. Summary & Outlook

3 1. Atomic force microscopy (AFM) AFM was invented in 1986 by Gerd Binnig, Calvin Quate and Christoph Gerber AFM Imaging mode For high-resolution surface imaging Laser HeNe LASER Photodetector POSITION SENSITIVE DETECTOR Contact mode AFM (l..6 nm v..1 nm) Oscillation mode AFM Cantilever LEVER Mirror Sample SAMPLE Modified from Meyer & Amer, 1988

4 1. Atomic force microscopy (AFM) AFM imaging mode: Advantages of oscillation - vs. contact mode 1. Reduces contact time, friction and lateral forces 2. Reduced vertical forces allow imaging of soft materials: DNA, proteins, cells oscillation mode is the most commonly used AFM method (easy, less invasive) AFM imaging mode: no quantification of biological, chemical and physical properties

5 1. Atomic force microscopy (AFM) AFM imaging mode: Contact/ Oscillation AFM force spectroscopy mode 1. Micro- and nanomanipulation 2. Force spectroscopy mode: Quantify inter- and intramolecular interaction forces: electrostatic, van der Waals, hydrophobic forces Approach & retraction: record Vertical displacement of AFM tip & deflection of the cantilever Force-displacement! force-distance curve describe dependence of interaction forces between the AFM tip and the sample from the distance 3. Probe mechanical properties Indentation-retraction experiments electrostatic properties, deformation, pressure, adhesion

6 Overview 1. Atomic force microscopy (AFM) Brief intro AFM Imaging mode: contact / oscillation Force spectroscopy mode 2. FD-based AFM 3. Applications FD-based AFM Paper 1: Native proteins Paper 2: GPCR-ligand interaction 4. Summary & Conclusion

7 2. FD-based AFM Force-distance curve-based atomic force microscopy FD-based AFM = AFM imaging + AFM force spectroscopy Photodiode Laser Lateral resolution 1 nm Vertical resolution.1 nm Distance Force x y Oscillating cantilever with Silicon nitride tip pixel by pixel manner pixel size < 1 nm 2 The value of every pixel of the final sample topography is determined by the tip-sample distance and the present imaging force.

8 2. FD-based AFM Force-distance curve-based atomic force microscopy Photodiode Laser b 15 F i Imaging force 3 Distance Force Approach x y 5 4 Retraction 5 5 Distance (nm) 1 15 Pfreudenschuh et al., nm = contact point tip-sample Approach FD curve 1. Noncontact 2. Initial contact 3. Repulsive contact Retracting FD curve 4. Adhesion 5. Noncontact

9 c 2. FD-based AFM Multiple physical forces can be derived from the approaching and retracting FD curves: F i F Low D Def Approach d F i D Modulus F Adh Imaging force Modulus fit region Retraction Energy dissipation 5 Distance (nm) Distance (nm) 1 15 Deformation: D Def = D FLow D Fi D Mechanical flexibility/stiffness Adhesion: minimal force Energy dissipation Young s modulus Pfreudenschuh et al., Reduced Young s modulus: E* = (F i F Adh ) (RD 3 Modulus ) 1 2 4

10 Multiple physical forces can be derived from the approaching and retracting FD curves: Physical forces of interactions: Coulomb forces van der Waals forces hydrophobic attraction solvation forces Biochemical forces: Covalent bonds Ligand-receptor pairs Biopolymers Nucleic acids Membrane and water-soluble proteins Cellular membranes Lipid bilayers

11 Overview 1. Atomic force microscopy (AFM) Brief intro AFM Imaging mode: contact / oscillation Force spectroscopy mode 2. FD-based AFM 3. Applications FD-based AFM Paper 1: Native proteins Paper 2: GPCR-ligand interaction 4. Summary & Conclusion

12 3. Applications FD-based AFM Multiparametric high-resolution imaging of native proteins by force-distance curve based AFM Moritz Pfreundschuh 1, David Martinez-Martin 1, Estefania Mulvihill 1, Susanne Wegmann 2 & Daniel J Muller 1 1Department of Biosystems Science and Engineering, ETH Zurich, Basel, Switzerland. 2 Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Charlestown, Massachusetts, USA. Correspondence should be addressed to D.J.M. (daniel.mueller@bsse.ethz.ch). Published online 17 April 214; doi:1.138/nprot Aim: Contour the surface and quantify biophysical and biochemical properties of native proteins at high resolution Samples: 2 water soluble proteins: fibrils of human tau protein, neurofibrillary tangles in AD Assess highly variable polymorphic structure and biophysical properties of amyloid-like fibrillar aggregates by FD-b AFM Bacteriorhodopsin from the purple membrane of H.salinarum (functionally & structurally best studied protein)

13 3. Applications FD-based AFM 1. Set up the AFM Isolate from acoustic, mechanical and electrical noise (noise analyzers) Equilibrate the AFM: Focus the laser beam onto the tip to adjust the photodiode signal 2. Sample immobilization 3. Cantilever selection 4. Record FD curves 5. FD curve analysis Photodiode Force Distance x Laser y

14 3. Applications FD-based AFM 1. Set up the AFM 2. Sample immobilization Better hydrophilic supporting surface than hydrophobic Mica: negatively charged at neutral ph, hydrophilic, good electrical insulator HOPG = highly ordered pyrolytic graphite: hydrophobic, good conductor Native membrane proteins should be present in the membrane Glue metal disc, hydrophobic teflon foil, mica sheet Cleave the mica sheet with Scotch tape

15 3. Applications FD-based AFM 1. Set up the AFM 2. Sample immobilization Sample preparation bacteriorhodopsin preparation: Dilute purple membrane stock solution in adsorption buffer to ~1 μg/ml and place ~3 μl of it onto the freshly cleaved mica for 15 3 min preparation of tau fibrils: Dilute the fibril solution in adsorption buffer to a final concentration of ~3 μg ml 1 and place ~1 2 μl of it onto the freshly cleaved mica for 1 2 min Remove adsorption buffer, apply imaging buffer and mount sample on AFM. Immerse cantilever in imaging buffer.

16 3. Applications FD-based AFM 1. Set up the AFM 2. Sample immobilization 3. Cantilever selection Soft AFM cantilever to measure interaction forces between single biomolecules, ranging typically from 5 to 25 pn. High resonance frequencies >1kHz are needed to detect fast biomolecular interactions. Shape and size of the AFM stylus determine the lateral resolution. Sharp styluses with a small tip radius (ca 2 nm) for high resolution.

17 3. Applications FD-based AFM 1. Set up the AFM 2. Sample immobilization 3. Cantilever selection When measuring the mechanical flexibility or stiffness, the spring constant (stiffness) of the cantilever should be similar to that of the sample. Intermediate stiffness ca.1n/m Sensitivity for mechanical flexibility is decreased, if stiff cantilevers >1N/m are used. Mechanical properties of biological systems are heterogeneously distributed and can change dynamically.

18 3. Applications FD-based AFM 1. Set up the AFM 2. Sample immobilization 3. Cantilever selection

19 3. Applications FD-based AFM 1. Set up the AFM 2. Sample immobilization 3. Cantilever selection Mechanical properties of biological systems are heterogeneously distributed and can change dynamically. Mechanical properties should be precisely assigned to structural details. Functional state at which mechanical measurements were performed should be well defined.

20 3. Applications FD-based AFM 1. Set up the AFM 2. Sample immobilization 3. Cantilever selection a 4. Record and analyze FD curves Approach Approach Retraction Retraction b c Approach Retraction Distance (nm) buffer solution and clean mechanical support = mica surface sharp transition in contact area Distance (nm) buffer solution and a clean and mechanically flexible sample smooth transition in contact area Distance (nm) buffer solution with contaminated AFM stylus or mica surface

21 4. Record & analyze FD curves densely packed patches of Bacteriorhodopsin, lipid bilayer Adhesion (pn) 4 c 4 nm.5 (nm) Deformation b 4 nm Adhesion 4 (pn) Topography a imaging force of 14 pn, a cantilever amplitude of 4 nm, a frequency of 2 khz and a scanning frequency of 1 Hz per line 5 1 (nm) 2 nm

22 4. Record & analyze FD curves Topography Young s modulus: force needed to stretch / compress the sample Deformation (nm) Deformation 5 18 (MPa) DMT modulus 1 (nm) Topography a c e Raw data Average b 1 nm 5 nm d 1 nm 5 nm f 1 nm 5 nm imaging force of 45 pn, a cantilever amplitude of 14 nm, a frequency of 2 khz and a scanning frequency of.77 Hz per line

23 Overview 1. Atomic force microscopy (AFM) Brief intro AFM Imaging mode: contact / oscillation Force spectroscopy mode 2. FD-based AFM 3. Applications FD-based AFM Paper 1: Native proteins Paper 2: GPCR-ligand interaction 4. Summary & Conclusion

24 3. Applications FD-based AFM Imaging G protein coupled receptors while quantifying their ligand-binding free-energy landscape David Alsteens 1,5, Moritz Pfreundschuh 1,5, Cheng Zhang 2,4, Patrizia M Spoerri 1, Shaun R Coughlin 3, Brian K Kobilka 2 & Daniel J Müller 1 Aim: image single native GPCRs in membranes and quantify their dynamic binding strength to native and synthetic ligands GPCR PAR1: Protease-activated receptor-1 Receptor for thrombin, important in coagulation cascade

25 GPCR PAR1: Protease-activated receptor-1 Receptor for thrombin, important in coagulation cascade TRAP: Thrombin receptor-activating peptide, binds to heptahelical bundle Quantify how tethered ligands bind PAR1 Force N S FL Thrombin N S FL TRAP L i RN ii iii L RN Extracellular Lipid bilayer Inactive PAR1 C C C Thrombinactivated PAR1 Intracellular

26 Functionalize the AFM tip with the TRAP: physiological condition If brought to contact, ligand and receptor can bind Retraction breaks the specific bond, the required force is measured by the deflecting AFM cantilever b Laser Photodiode c Distance PEG linker Peptide F.25 khz Covalently bind TRAP to PEG poly ethylen glycol spacer, which is chemically attached to the cantilever PAR1 Peptide Native PAR1 terminus Optimal conditions: Tip oscillation at.25 khz & amplitudes of 5 nm y Lipid bilayer d x Sample Sample: human PAR1 in proteoliposomes

27 Tip-sample interaction Force Time Force Applied force Distance Adhesion Approach Retraction Deformation Force vs. time curve Force vs. distance curve Adhesion: Minimal force of the retraction FD curve Deformation: The difference between two applied imaging forces proteoliposome was too short to allow ligand binding o Topography Height ii 1 nm Adhesion d e f Adhesion ii 1 pn Height Adhesion ii iv iv i i i iii iii iv iii 25 nm Topography: Membrane patches protrude 4.5 nm +/-.7 nm Topography = Height measurement is congruent with adhesion map

28 Free energy landscape of the receptor: Assess ligand binding free energy a Free energy Bound state Unbound state x u Transi(on state (F cos ) x k off k off Reaction coordinate (x) F = G F > G bu Activation free energy to cross the transition state free-energy difference between bound & unbound state external force stressing a bound, reduces the activation energy barrier towards unbound state Apply external forces and measure how much is needed to break the ligand-par1 bond

29 Force required to separate the ligand from PAR1 is plotted against the loading rate Feq = 49.9 ±.9 pn 14.1 ± 4.7 pn e Loading rate: describes the force applied c over timefx eq==.3 a xu =.6 ±.9 Å ±.1 Å.5.2 ms N SFLLRN 15 N R R L L L F S L F S a 1, 1, 5 koff = 3,621 ± 885 s ms N 5c 2 R L 1, 1, 1,, 1,, b b Gbu = ±.42 kcal mol 1 d 2 bu G 8.61 ±.82 koffbu==2,767 ± 361 s 1 mol 1 Loading ra Noise level 25 pn 1, Gbu = 5.73 ± 3.76 kcal mol 1 f d 1, 1, off f off 1 bu 1 1, 1 R Y pf-f A hr Cha 5 Noise level 25 pn Cha 1, Loading rate (pn s 1) Noise level 25 pn 1,, 1 1, 1, 1, Loading rate (pn s 1) NA 15 hr 2 1,, R L L F S 1 pf-fy A hr Cha 5 R pf-f Noise level 25 pn 1, 1, 15 1 A 1, Loading rate (pn s 1) 1,, 5 5 Figure 4 Loading rate dependent interaction forces of single ligand-receptor bonds quantitate the ligand-binding energy landscape of PAR1. 1 (a d) For four different peptides SFLLRN (a), SFLLAN (b), SALLRN (c), and A(pF-F)RChahRY (d), the force required to separate the ligand from PAR1 is plotted against the loading rate. (e,f) For SFLLRN (e) and SFLLAN (f), the force required to separate the ligand from PAR1 complexed with the antagonist Noise 5 level 25 pn G +Vorapaxar Y 15 L F S L NA 1, u= Gbu = 8.47 ± x1.12 u 1.5 xu =.3 ±.2 ÅFeq eq u Feq = 27.1 ± 1.8 pn 1,, Loading rate (pn s 1) eq bu 1, 1, Noise level 25 pn F = 51.1 pn 1, 1,, 1,± 1.3 1, = ) (pn spn xu =.6F±eq.1 ÅLoading±rate1.3 1 bu Noise level 25Nois pn L ±pn.1kcal Å mol u F= =.6 F = 27.1 ± 1.8 pn Gbu = x ±±.58 x =.6 ±.1 Å d f x =.3 ±.2 Å 1 1 G = ±.58 kcal mol 2 koff = 3,121 Gbu±= ±.58 kcal mol G = 8.47 ± 1.12 kcal s 2 G = 8.61 ±.82 kcal mol mol 2 k = 3,121 ± 37 s 2 k = 2,767 ± 361 s.2 ms.2 ms k = 3,121 ± 37 s.2 ms Feq = 29.5 ± 1.4 pn xu =.4 ±.1 Å NR 1 ARTICLES Loading rate (pn s 1) 1,,eq 1, Loading rate (pn s 1) x G ±.1± Å.82 kcal mol 1 b u =.4 = Loading rate (pn s 1) kcal SFLLRN: and Arginine important for specificity 2 Phenylalanine off k off 2,767 ms ± 361 s 1.5 = ms Replace by Alanine: SALLRN and SFLLAN NA ms L SALLRN: Gbu = 5.73 ± 3.76 Nkcal L Fm S ol A L mol 1 SFLLAN: Gbu = 8.61 ±.82 k cal L F S 1 1 reduced free energy difference, abolished high-affinity binding Noise level 25 pn 1, 15 L A S15 Feq = 26.4 ± 1.6 pn xu =.3 ±.1 Å Gbu = 8.38 ± 1.4 kcal mol 1 2 N 5 1, L A S e Feq = 14.1 ± 4.7 pn xu =.3 ±.1 Å Gbu = 5.73 ± 3.67 kcal mol 1 R Noise 15 level 25 pn 25 pn L Noise level L L F S Loading rate (pn s 1) 1 5 Feq = 29.5 ± 1.4 pn Fxeq = 29.5 ± 1.4 pn u =.4 ±.1 Å L L A S L FS + Vorapaxar 1 1 1, 1, 1, 1, 1, 1, 1,, 1, 1,, 1 Loading rate (pn s ) NR 1 Gbu = ±.42 kcal mol , 1, 1 NR SALLRN 15 Feq = 49.9 ±.9 pn xu =.6 ±.9 Å 2 Noise level Noise level 25 25pN pn xu =.3 ±.1 Å 2 Gbu = 8.38 ± 1.4 off 1 k G off bu==3, ±± skcal mol 1 k.5 =.2 ms± 885 s 1 3,621 e u u Feq = 14.1 ± 4.7 pn Gbu =Å 5.73 ± 3.67 kcal mol 1 2 xu =.3 ±.1 Gbu = 5.73 ± 3.67 kcal mol c = 49.9 ±.9 pn 2 F eq 1 G = bu.6 x = ±.9 Å ±.42 kcal mol a Feq xu = Feq = 26.4 ± 1.6 pn G Noise level 5 25 pn

30 Free energy binding landscape of three different peptide-ligands b d Free energy x u SFLLRN.6 ÅSFLLAN SFLLRN.4 ÅSALLRN SFLLAN.3 Å SALLRN G bu 5.7 kcal mol kcal mol kcal mol 1 Bound state Unbound state Reaction coordinate (unbinding) Free energy difference between the ligand-bound and unbound state Less energy needed to break bond Lower affinity to PAR1

31 Overview 1. Atomic force microscopy (AFM) Brief intro AFM Imaging mode: contact / oscillation Force spectroscopy mode 2. FD-based AFM 3. Applications FD-based AFM Paper 1: Native proteins Paper 2: GPCR-ligand interaction 4. Summary & Outlook

32 4. Summary & Outlook FD-based AFM allows to: Structurally quantify the physical (mechanical), chemical and biological properties of native proteins Is applicable for most membrane and watersoluble proteins. Image human PAR1 in proteoliposomes at high resolution and simultaneously quantify their dynamic binding strength to different ligands.

33 4. Summary & Outlook Limitation: Limited to the characterization of single native proteins in vitro Outlook: Characterize single proteins in their native environment of the living cell or tissue to determine how proteins work in the cellular context and how cells control proteins to function as required. Image cells at subnanometer resolution

34