Starting with a fighting chance: Happy protein Shane Seabrook PEW - - PowerPoint PPT Presentation

Starting with a fighting chance: Happy protein

Shane Seabrook – PEW 2014 #PEW2014 @CSIROC3 @CSIROnews

CSIRO FMF Biophysics Group

• Structural Biology Campaigns
• Small Molecule Therapy Development
• Biotherapy Formulation Development
• Antibody Screening
• Mechanistic Biology+Nano Particle Interactions
• Bio-remediation, bio-cataysis & bio-fuels

Topics

• Protein Formulation
• Stability
• Aggregation

Why are you so unhappy, Protein?

• 1. Expression
• 2. Purification
• 4. Structure

Protein Production Protein Crystallization

• 3. Crystallisation

• 1. Obtaining a population of identical protein

– Same shape (fold) – Same activity

• 2. Keeping that population stable (folded/active)

Challenge: Happy Protein

Goal: Uniformity

Building insight

• 1. Use a different construct.
• 2. Change the expression/purification process.

OR

• 3a. Alter the solvent surrounding your protein.
• 3b. Use ‘binders’ to tighten the structure.

...providing armour and weaponry; giving it a fighting chance!

Formulations

What is a Formulation

• A mixture of two of more chemicals

– Buffer – Salt – Additive

• Subtle changes in the formulation chemistry can have a

significant effect on the behaviour / stability of your protein

Learn the chemistry that makes your protein happy

Protein stability ƒ(formulation)

...strongly affected by the formulation...

Seabrook, S. A.; Newman, J., High-Throughput Thermal Scanning for Protein Stability: Making a Good Technique More Robust. ACS Combinatorial Science 2013.

• 1. Crowther, G. J.; He, P.; Rodenbough, P. P.; Thomas, A. P.; Kovzun, K. V.; Leibly, D. J.; Bhandari, J.; Castaneda, L. J.; Hol, W. G. J.; Gelb, M. H.; Napuli, A. J.; Van Voorhis, W. C., Use of thermal melt curves to assess the quality
• f enzyme preparations. Analytical Biochemistry 2010, 399 (2), 268-275.
• 2. Crowther, G. J.; Napuli, A. J.; Thomas, A. P.; Chung, D. J.; Kovzun, K. V.; Leibly, D. J.; Castaneda, L. J.; Bhandari, J.; Damman, C. J.; Hui, R.; Hol, W. G. J.; Buckner, F. S.; Verlinde, C.; Zhang, Z. S.; Fan, E. K.; Van Voorhis, W. C.,

• 3. Dupeux, F.; Rower, M.; Seroul, G.; Blot, D.; Marquez, J. A., A thermal stability assay can help to estimate the crystallization likelihood of biological samples. Acta Crystallographica Section D 2011, 67 (11), 915-919.
• 4. Ericsson, U. B.; Hallberg, B. M.; DeTitta, G. T.; Dekker, N.; Nordlund, P., Thermofluor-based high-throughput stability optimization of proteins for structural studies. Analytical Biochemistry 2006, 357 (2), 289-298.
• 5. Falconer, R. J.; Marangon, M.; Van Sluyter, S. C.; Neilson, K. A.; Chan, C.; Waters, E. J., Thermal Stability of Thaumatin-Like Protein, Chitinase, and Invertase Isolated from Sauvignon blanc and Semillon Juice and Their Role

• 6. Froese, D. S.; Healy, S.; McDonald, M.; Kochan, G.; Oppermann, U.; Niesen, F. H.; Gravel, R. A., Thermolability of mutant MMACHC protein in the vitamin B12-responsive cblC disorder. Molecular Genetics and Metabolism

• 7. Gast, K.; Damaschun, G.; Misselwitz, R.; Zirwer, D., Application of dynamic light scattering to studies of protein folding kinetics. European Biophysics Journal 1992, 21 (5), 357-362.
• 8. Good, N. E.; Winget, G. D.; Winter, W.; Connolly, T. N.; Izawa, S.; Singh, R. M. M., Hydrogen Ion Buffers for Biological Research*. Biochemistry 1966, 5 (2), 467-477.
• 9. He, F.; Becker, G. W.; Litowski, J. R.; Narhi, L. O.; Brems, D. N.; Razinkov, V. I., High-throughput dynamic light scattering method for measuring viscosity of concentrated protein solutions. Analytical Biochemistry 2010, 399

Formulation Workflow

Formulation(s) Optimization

Assays Common initial strategy is to take a punt on a combination of the following:

Base condition: Tris, PO4, HEPES (cheap, readily available) 50mM to 1M NaCl Additives: 0 to 10% polyol (> solubility of an unhappy protein) Reducing agent (DTT, TCEP, bME) Co-factors (divalent cations, organics) Other stuff – sufactants, aa’s, sugars(polyols)

A CHEMICAL DEGUSTATION

Formulation Workflow (HTP)

HTP Screen Screen Production

HTP Assay

Aggregation / Activity Optimization

Systematic Screen Design

pH 5.0 pH 7.0 pH 9.0 All buffers at 50mM with 50mM and 200mM NaCl

Water (milli Q) pH 7.0 – imidazole pH 5.0 – sodium acetate pH 7.0 – MOPS (3-(N-morpholino)propanesulfonic acid) pH 5.5 – piperazine pH 7.5 – HEPES pH 6.0 – MES (N-morpholino)ethanesulfonic acid) pH 7.5 – Na2H/KH2 PO4 pH 6.0 – citric acid pH – 8.0 tris chloride pH 6.5 – bis-tris pH 8.5 – glycyl-glycine pH 6.5 – ADA (N-(2-Acetamido)iminodiacetic Acid) pH 9.0 – CHES (N-Cyclohexyl-2-aminoethanesulfonic acid)

Crystallisation friendly buffer screen:

• Diverse pH, variable [NaCl]
• Duplicate pH with different buffers
• Controls (is the result real?)

Additive/Ligand Screening

• Find a good ‘base condition’, then

– Hampton Research Solubility and Stability Screen – (conditions available on c6.csiro.au)

• Or be specific:

– Family of known co-factors and/or inhibitors – Specific chemical groups (e.g. Polyols, Surfactants)

Summary - Formulations

I. Look at a wide range of chemical space, your protein could be unhappy by only 0.5 pH unit, or 50mM NaCl II. Think about ways to assess the formulations in HTP (we will look at an accessible method next)

• III. Make a list of potential chemical agents that could enhance

stability – co-factors, surfactants, polyols

• IV. Think about controls – be sure you’re improving the stability

and not just complicating the process

Thermal Stability

Assumption!

...thermal stability is a proxy for overall stability

The more energy (heat) required to unfold a protein in a given chemical condition = the less prone it will be to denaturation in that chemical condition over time... ...this means you will have more of the same (uniform) protein to use for your campaign of interest.

Assumption!

...that proteins are only “gently” mis-folded

– i.e. Disruption to the quaternary / tertiary structure

• Not for refolding completely denatured protein

The Classic Thermal Melt

Intensity Max (Imax) Intensity Min (I0)

Tm (T50) T10 T (Tonset – Tmax)

DSF - theory

Temperature

• Based on interaction between the

hydrophobic core of proteins and an environmentally sensitive dye (SYPRO)

• ex = 490nm
• em = 570nm
• The peak of -dRFU/dT = Th

DSF – in practice

• Mass dilution using automation (no dialysis or buffer exchange)
• Add small amount of Sypro dye (in excess)
• Put into a RT-PCR machine, increment 0.5C every 5 seconds
• Et voila! About 1 hour later, 96 melt curves

DSF – lots of curves

DSF – Controls

Blue = Lysozyme, the machine control; we did what we thought we were going to Black = Protein without dye; the protein isn’t fluorescent active at em/ex Orange = Formulation with dye, no protein; the formulation isn’t fluorescent active at em/ex

DSF – melt analysis (1D, Th)

Case study – Endogluconase (cellulase)

pH 5.0 pH 7.0 pH 9.0

25 30 35 40 45 50 55 60 65 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0

pH

Th (ºC)

DSF – protein behaviour

Endogluconase Transferase Amidase

• Proteins behave REALLY differently … e.g.

DSF – Good Curves

Blue = bis-tris pH 6.5 + 50mM NaCl Pink = ches pH 9.0 + 200mM NaCl Green = 200mM NaCl

UNIFORM SAMPLE

DSF – Bad Curves

Blue = bis-tris pH 6.5 + 50mM NaCl (Th 43C) – even though lower Th, best curve shape Red = citrate pH 6.0 + 50mM NaCl (Th 45C) Green = Acetate pH 5.0 + 50mM NaCl (n/a)

DSF – Good Curves or Bad Curves?

Multimodal melt curves ... One of three things:

• 1. Protein forms oligomers (common)
• 2. Protein has multiple domains (antibody)
• 3. There is more than one protein present (eek!)

UNIFORM SAMPLE...?

Reducing confusion

Automated Processing of DSF experiments – saving you time, making robust decisions

• UROP Students (undergrad research opportunities program)
• Biomedical Research Victoria (Bio21 Cluster)
• Marko Ristic & Nick Rosa
• Find the best Tm
• Of the good curves, which had the highest Tm
• Number of curves which have been kept
• Throws out the ‘Bad Curves’ (using a curve fit classifier)
• Tms vs buffer, for both salt concentrations
• Which chemical/pH/[salt] increased/decreased Tm?
• The control checks are output one by one
• Did our process work
• Protein in original Formulation
• What was your benchmark

Example I

Starting formulation: Hepes pH 7.5 + 150mM NaCl + 5% Glycerol – Th = 53C ADA pH 6.5 + 200mM NaCl – Th = 57.5C MES pH 6.0 + 200mM NaCl – Th = 55.5C (Better shape curve)

Final Formulation: MES pH 6.5 + 200mM NaCl

Returned huge/reproducible crystals that diffract to 2.0 Å

Example II

Starting formulation: TBSA – Th = 60.5C Bis-tris pH 6.5 + 50mM NaCl – Th = 75.0C Note! Very sensitive to high pH CHES pH 9.0 – Th = 37C Returned beautiful crystals (Crystals diffract to ≈3.0 Å):

Assisting protein purification; e.g.

Problem: Construct unhappy in standard purification system Solution: Use DSF to find a better purification formulation

Expression etc Affinity Chromatography Size Exclusion Chromatography

Additives / Ligands

• Assume that a bound ligand

will increase the stability

Native Protein Native Protein With Ligand Bound Folded Molten Globule Unfolded

ENERGY

Ligand Binding Energy (~Gbind) ~Tm Molten Globule Melted Protein Unfolded Protein Native Protein   Native Protei  

• Thermal denaturation curves
• AstC (Th = 43.5°C)
• AstC plus PLP cofactor (Th = 46.0°C)
• Crystals of holo-AstC diffracted to a higher resolution than

the apo-protein (2.20 Å)

• Dimer of holo-AstC, with PLP clearly bound in the

cofactor site at the dimer interface.

Newman J, Seabrook S, Surjadi R, Williams CC, et al. (2013) Determination of the Structure of the Catabolic N-Succinylornithine Transaminase (AstC) from Escherichia coli. PLoS ONE 8(3): e58298.

DSF – Searching for binders

• TM values from a series of DSF curves for Arginase vs. Ligand.
• 10x ligand concentration was enough for a measurable

increase in TM (>1ºC) for active binders.

• Can be rapidly applied to a wide range of protein:ligand

systems in an effort to narrow down potential candidates

What about ‘sticky’ proteins?

Differential Static Light Scattering (DSLS)

Temperature of Aggregation (Tagg)

Io IӨ

Intensity of scattered light (IӨ) is proportional to the degree of aggregation

DSLS – in practice

• Mass dilution using automation (no dialysis or buffer exchange)
• Requires a little more protein, 0.1 to 0.5mg/mL final conc.
• Put into a DSLS machine, choose ramp rate
• Et voila! About 1 hour later, 383 melt curves

Intensity Max (Imax)

Tm (T50) T10 T (Tonset – Tmax)

Intensity of Scattered Light

Intensity Min (I0)

DSLS - Output

Low pH High pH

50mM NaCl 200mM NaCl 50mM NaCl 200mM NaCl

Summary - Melts

I. You probably have an RT-PCR machine and SYPRO dye in your lab, you can do this easily with your protein II. You may not have a DSLS machine, but we do

• III. If you can’t find automation, dilute out your protein and the

dye so that you can use a multi-channel pipette

• IV. Don’t just look at the melt temp, pay close attention to the

shape of the melt curves V. Use this as a rough and quick alternative to look for ligands that might bind to your protein

Aggregation

DLS - theory

The measurement of Rayleigh light scatter to determine Hydrodynamic Size (sphere: Rh)

• Basically, LARGE things diffuse slowly and scatter light

differently to small things, which diffuse quickly

• The larger the particle, the more light it scatters

Stokes-Einstein relationship

D = kT 6  Rh

Rh ≈/ MW

DLS – in practice

• Clean sample (no dust)
• Clean cuvette (no dust)
• Clean room (no dust)

Theme? No dust!!

• >0.1mg/mL (ideally >1mg/mL)
• Volume limited by cuvette architecture
• Temp limited by machine capacity
• Add some inert oil to inhibit evaporation

Good Data vs. Bad Data

Same protein – 3 formulations

Measure aggregation caused by altering protein formulation

Endogluconase

• Least aggregate (Best Th)
• A bit of aggregate (control Th)
• A lot of aggregate (Worst Th)

Acetate pH 5.0 (best Th): 99% (mass) monomer

• Excellent candidate
• High purity (low %Pd)

CHES pH 9.0 (worst Th): 63% (mass) “low-mer”

• Terrible – not a monomer anymore
• 47% large aggregate

Same protein – 3 formulations

Control 3.6nm %PD = 25% MgCl2 2.9nm PD = 23% LiNO3

3.8nm %PD = 40%

Temperature of Aggregation (Tagg)

Time of Aggregation

3 hrs 6 hrs 1 Day

Summary

I. You can probably find one of these at your Institute, they’re fairly easy to use II. Quickest way to tell if you have a UNIFORM protein sample without aggregates

• III. Very sensitive to dust and particles – be clean and filter

everything that you possibly can

• IV. Easy to do a stability versus time study, though this can

generate an overwhelming amount of data to process

Overall Summary

I. Be adventurous, try some different formulations and save yourself purifying a new batch of protein II. Think about the physical behaviour and biochemical pathways ... what might help your protein?

• III. Talk to people in your lab and institute, find out what kit you

have and how you can make use of it

• IV. Talk to me
• Everything I’ve talked about is accessible within FMF Biophysics

Thanks!

#PEW2014 @CSIROC3 @CSIROnews

crystal.csiro.au