1 Gel Filtration (GF) or Size Exclusion Chromatography (SEC)
1
Gel Filtration (GF) or
Size Exclusion Chromatography
(SEC)
2
Gel Filtration chromatography (GF)
Principles of GF
Fractionation range
Parameters for resolution optimization
Use of GF
Column performance
Troubleshooting
Examples
3
What is gel filtration?
Gel filtration is a technique of liquid chromatography which separates molecules according to their ratio (sizes, oligomeric state) Beads with pores of well-defined sizes: fractionation range Mobile phase: almost all kind of buffers
4
Gel structure Agarose
Dextran
A hypothetical structure for Superdex
Beads of defined porosity: fractionation range
The degree of cross-linking determines the size of the pores and therefore the
fractionation range of the resin
Unlike many other chromatographic procedures, size exclusion is not an adsorption
technique.
Void volume Vo
Volume of the gel matrix Vs
Pore volume Vi
5
Terms and explanations
Vo= Void volume: volume of the solution outside the beads, or
elution from very large molecules
Ve = the volume from the time the protein is placed until it
appears in the effluent
Vi = volume of the solution inside the beads = Vc - Vs - Vo
Vc = Total (geometric) volume of the column
Vt = Elution volume for very small molecules
2 3
Void volume Vo
Volume of the
gel matrix Vs
Pore volume Vi
1
Vo
Ve
Vt
Vc
https://www.youtube.com/watch?v=oV5VB5kO3tQ https://www.youtube.com/watch?v=E3z1wIImvHI https://www.youtube.com/watch?v=rPRbqYWlSEo
6
Steric exclusion
Molecules are excluded from the gel bead to different extents according to their sizes.
Gel bead
Largest molecules - excluded from pores, travel with the mobile phase, elute rapidly from column • The volume at which large molecules elute is called the void volume, Vo (same as the volume of solution that surrounds the beads)
Smallest molecules – enter the pores of the beads, are included in the matrix and retarded in their movement, spend most of the time in the stationary phase, elute last • The volume at which small molecules elute corresponds to Vt (total volume of solution surrounding (Vo) and inside the beads, Vs) Vt = Vo + Vs
Intermediate size molecules – spend different amounts of time both inside and outside the beads (partition between the mobile and stationary phase) • The volume at which intermed.molecules elute is called the elution volume (Ve) and depends on the partition of the molecule between the Vo and Vs which is proportional to the distribution coefficient (K) Ve = Vo + KVs
7
Constructing a selectivity curve
Kav 1
0 log (Mr)
Run standards and determine the elution volume for each
Calculate Kav values
Plot log (Mr) for each standard against the calculated Kav
Selectivity curve is usually moderately straight over the range Kav=0.1 to Kav=0.7
Extrapolate MW of your protein according to his Ve
ot
o
VV
VVeavK
8
Shape effects
Different fractionation
ranges for:
•Native, globular proteins
•Partially folded molecules
•Proteins inside detergent
micelle: MW of protein +
MW of micelle
Denatured proteins
Kav 1
0.7
0.1
log (Mr)
Native proteins
9
How to choose GF type
Kav 1
log (Mr)
Molecules with different shapes have different
selectivity curves
Linear polysaccharides
Globular proteins
RESIN FR Glob Prot FR Dextrans
Sephacryl S100 1-100 kDa ND
Sephacryl S200 5-250 kDa 1-80 kDa
Sephacryl S300 10-1500 kDa 2-400 kDa
Sephacryl S400 20-8000 kDa 10-2000 kDa
Protein 1: 30kDa Protein 2: 80kDa
Vo Vt
10
Gel Filtration chromatography (GF)
Principles of GF
Fractionation range
Parameters for resolution optimization
Use of GF
Column performance
Troubleshooting
Examples
11
Incr
easi
ng
excl
usi
on
lim
it
Results depend on selectivity (fractionation range)
Sephacryl S-100
40 80 120 ml
Sephacryl S-300
Sephacryl S-200 BSA Cyt C
IgG b-L Cytidine
AU280
Better for larger proteins
Better for smaller proteins
Best for these proteins
RESIN FR Glob Prot FR Dextrans
Sephacryl S100 1-100 kDa ND
Sephacryl S200 5-250 kDa 1-80 kDa
Sephacryl S300 10-1500 kDa 2-400 kDa
Sephacryl S400 20-8000 kDa 10-2000 kDa
12
High efficiency can compensate for low selectivity.
If selectivity is high, low efficiency can be tolerated (if large peak volume is acceptable).
Resolution depends on efficiency and selectivity
Low selectivity
High selectivity
high efficiency
low efficiency
high efficiency
low efficiency
13
14
Efficiency
Efficiency depends on:
Particle size of matrix
Particle size distribution of matrix
Packing quality of the column
Sample (volume, purity and viscosity)
Flow rate (more important for bigger beads)
15
Peak width depends on particle size
Superdex Peptide 13-15 µm Superdex 30 prep grade 24-44 µm
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0 10 20 30 40 50
Retention time (min)
AU214
0
0.01
0.02
0.03
0.04
0.05
0.06
0 10 20 30 40 50
Retention time (min)
AU214
16
1 x Superdex® Peptide HR 10/3000 2 x Superdex® Peptide HR 10/30
Resolution depends on column length Increasing column length increases resolution
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0 5 10 15 20 25
Retention time (min)
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0 10 20 30 40 50
Retention time (min)
AU214
0
50
100
150
mAU
40.0 45.0 50.0 55.0 60.0 65.0 70.0 75.0 ml
0
50
100
150
mAU
60.0 70.0 80.0 90.0 100.0 110.0 ml
Superdex® Peptide 60 x 1.6cm ~ 120ml Superdex® Peptide 100 x 1.6cm ~ 200ml
SUMO-Atox1
SUMO-Atox1
SUMO
SUMO
Atox1
Atox1
Michal Shoshan from Edith Tshuva lab.
17
Column: Superdex® Peptide HR 10/30
Resolution depends on sample volume
25µl
0
0.02
0.04
0.06
0.08
0.1
0 5 10 15 20 25
Retention volume (ml)
AU214
200 µl
200 µl
400 µl
0
0.02
0.04
0.06
0.08
0 5 10 15 20 25
Retention volume (ml)
AU214
400 µl
0
0.02
0.04
0.06
0.08
0.1
0 5 10 15 20 25
Retention volume (ml)
A214
25 µl
18
Why use gel filtration?
Group separations: Desalting, Buffer exchange, Removing reagents (replace dialysis)
Purification of proteins and peptides: complex samples, monomer/dimer
QC: Estimation size & size homogeneity
Protein-Protein Interaction
19
Desalting proteins
Desalting in a simple column
Column:
Sample:
Buf fer:
PD-10
HSA, 25 mg
NaCl 0.5M
HSA NaCl
volume
Volume for desalting: up to 25% column volume
20
Use in group separations
Adjusting pH, buffer type, salt
concentration during sample
preparation, e.g. before an assay.
Removing interfering small
molecules: EDTA, Gu.HCl, etc
Removing small reagent molecules,
e.g. fluorescent labels, radioactive
markers.
Gravity Desalting Columns
Multi Spin Desalting Columns
FPLC Desalting Columns
Spin Desalting Columns
21
HiTrapDesalt10ml001:1_UV1_280nm HiTrapDesalt10ml001:1_Cond HiTrapDesalt10ml001:1_Fractions HiTrapDesalt10ml001:1_Inject HiTrapDesalt10ml001:1_Logbook
0
100
200
300
400
500
600
700
800
mAU
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 ml
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
HiTrapDesalt10ml002:1_UV1_280nm HiTrapDesalt10ml002:1_Cond HiTrapDesalt10ml002:1_Fractions HiTrapDesalt10ml002:1_Inject HiTrapDesalt10ml002:1_Logbook
0
200
400
600
800
1000
mAU
0.0 5.0 10.0 15.0 ml
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Waste 17 18 Waste
Desalting in the presence of buffer + 250mM NaCl
Desalting in the presence of buffer + 100mM NaCl
OD 280nm
Conductivity
Use buffer that avoid protein precipitation
Gali Prag
22
Fractionation of multiple components
Separate multiple components in a sample on the basis of differences on their size
Best results with samples that contains few components or partially purified samples
(polishing step) : Not recommended for proteins close by MW
Limited sample volume (0.5-4% of total column volume). Not so suitable if the
sample volume is large
Flow-rate limitation : Time consuming
Removes higher oligomeric states and other aggregates
Protein elutes with the column equilibration buffer
23
Separating dimer and oligomers from monomer
Column: Superdex 75 HR 10/30 Sample: A special preparation of rhGH in distilled water
0.025
0.05
Oligomer
Monomer
ime (min)
Dimer
T 10 20 V O V C
280 nm
A
11/26/2015 24
Case study: HLT-p53CT- Affinity start with pellet of 1.5L culture
Ni-Sepharose FF 14ml
HLTp53CTNiNTA16ml004:1_UV1_280nm HLTp53CTNiNTA16ml004:1_UV2_260nm HLTp53CTNiNTA16ml004:1_Conc HLTp53CTNiNTA16ml004:1_Fractions HLTp53CTNiNTA16ml004:1_Inject HLTp53CTNiNTA16ml004:1_Logbook
0
500
1000
1500
2000
2500
mAU
200 250 300 350 400 450 ml
F4 Waste 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Load + 10cv 0%B + 3cv 8%B + 4cv 15%B + 4cv 100%B
POOL 17-22: 3.5OD x 35ml ~ 276mg
11/26/2015 25
HLT-p53CT- Cation Exchange after TEV
protease cleavage ON 4ºC SP-Sepharose FF 5ml HLTp53CTHiTrapSP5mlml005:1_UV1_280nm HLTp53CTHiTrapSP5mlml005:1_UV2_260nm HLTp53CTHiTrapSP5mlml005:1_Cond HLTp53CTHiTrapSP5mlml005:1_Conc
HLTp53CTHiTrapSP5mlml005:1_Fractions HLTp53CTHiTrapSP5mlml005:1_Inject HLTp53CTHiTrapSP5mlml005:1_Logbook
0
50
100
150
mAU
500 550 600 650 700 ml
F3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44
26
p53C-Terminal - Polishing column Column: Sephacryl S100 prep. 960 x 26mm (~500ml) 6ml/fract.
Ni column – TEV protease cleavage ON – dilution – CEIX – concentration & GF
HLTp53CTSephacrylS100of500ml004:1_UV1_280nm HLTp53CTSephacrylS100of500ml004:1_UV2_260nm HLTp53CTSephacrylS100of500ml004:1_Fractions HLTp53CTSephacrylS100of500ml004:1_Inject HLTp53CTSephacrylS100of500ml004:1_Logbook
0
50
100
150
mAU
150 200 250 300 350 400 ml
F3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
Fractions around 23 and 32 are higher MW impurities
POOL 7-14
Ronen Gabizon from Assaf Friedler lab.
27
Quality control: Use for size estimation / oligomeric state
Gives an estimate of molecular size in native or denative solution (Guanidine HCl, urea, detergents)
MW of globular protein using native buffers (Precision is not so good)
Oligomeric state of the protein / homogeneity / complex
Complementary information to PAGE-SDS
Size exclusion chromatograph in line with multi angle light scattering ,
added value to characterize proteins mass and shape in native solution
conditions
Calculating Mw and radius from the light scattering equations – much
more accurate.
Calculate the Mw during the elution peaks- detect homogeneity sample.
Detect low amount of aggregation – large molecules amplify the
intensity of LS.
Useful for protein/protein or protein/ligand interaction
SEC-MALS: Size Exclusion Chromatography - Multi Angle Light Scattering
29
Refolding by Dilution with Non Detergent Sulfo Betaine
0.0
5.0
10.0
15.0
mAU
0.0 5.0 10.0 15.0 ml
Refolding with 1MNDSB at 0.1mg/ml
0.0
10.0
20.0
30.0
40.0
mAU
0.0 5.0 10.0 15.0 ml
refolding with 1M NDSB at 0.2mg/ml
0.0
5.0
10.0
15.0
20.0
mAU
0.0 5.0 10.0 15.0 ml
refolding with 1M NDSB at 0.5mg/ml
0.0
10.0
20.0
30.0
mAU
0.0 5.0 10.0 15.0 ml
Refolding with 1M NDSB at 1mg/ml
Multimer Monomer
30
Complex formation: Leptin and Leptin Receptor
Superdex75prep002:1_UV3_220nm Superdex75prep002:1_Fractions Superdex75prep002:1_Inject Superdex75prep002:1_UV3_220nm1 Superdex75prep002:1_UV3_220nm2 Superdex75prep002:1_Logbook
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
mAU
100 150 200 250 ml
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Waste
Superdex 75 160x1.6cm column - Buffer: 20mMTrisHCl pH8.0 50mMNaCl 0.02%NaN3
Receptor alone
Leptin alone
Complex: Leptin + Receptor
31
Gel Filtration chromatography (GF)
Principles of GF
Fractionation range
Parameters for resolution optimization
Use of GF
Column performance
Troubleshooting
Examples
32
Increasing resolution
Choose appropiate fractionation range
Increase column volume (Connect two columns)
Reduce the flow rate
Change to a gel with smaller beads (higher efficiency)
Reduce the sample volume / protein quantity
Check the column efficiency
Clean and/or re-pack
SUMO alone and ATOX1
ATOX1 alone
Optimization of ATOX1 purification
Michal Shoshan from Edit Tshuva lab
SUMO-Atox1 IMAC purification
ATOX1: SUMO protease treatment after IMAC purification. Concentration by UF instead of AS precipitation. Load on 200ml
Superdex 30 column (20mM MES pH 6.0 + 150mM NaCl)
ATOX1: SUMO protease treatment after IMAC purification. Ammonium Sulphate precipitation. Load on 120ml Superdex 30
column (20mM MES pH 6.0 + 150mM NaCl)
ATOX1: SUMO protease treatment after IMAC purification. Concentration by UF in the presence of 4M NaCl. Load on 200ml
Superdex 30 column (20mM MES pH 6.0 + 250mM NaCl)
SUMO alone
ATOX1 alone
0
50
100
150
mAU
60.0 70.0 80.0 90.0 100.0 110.0 ml
chelating2mlAir Imp7Beta001 18 12 12001:1_UV1_280nm chelating2mlAir Imp7Beta001 18 12 12001:1_UV2_260nm chelating2mlAir Imp7Beta001 18 12 12001:1_Conc chelating2mlAir Imp7Beta001 18 12 12001:1_Fractions chelating2mlAir Imp7Beta001 18 12 12001:1_Logbook
0
20
40
60
80
100
%B
120.0 130.0 140.0 150.0 160.0 170.0 180.0 190.0 ml
F4 1 F2 2 3 4 5 6 7 8 9 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 3132
Fractions 16-22 Peak-833 mAU
100%B
20%B
15%B
Chelating2mlAir, Imp7598 ImpBeta1-442 from 1L 17C ON 19.12.12
Imp7BetaSuperose12prepar200ml 18 12 12001:1_UV1_280nm Imp7BetaSuperose12prepar200ml 18 12 12001:1_UV2_260nm Imp7BetaSuperose12prepar200ml 18 12 12001:1_Fractions Imp7BetaSuperose12prepar200ml 18 12 12001:1_Inject Imp7BetaSuperose12prepar200ml 18 12 12001:1_Logbook
0
20
40
60
80
100
120
140
mAU
60 80 100 120 140 160 ml
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
Extra ImpBeta Peak - 106ml
Complex Peak - 87ml
Aggregate Peak - 65ml
Superose12prepar200ml Imp7 598 ImpBeta1-442 + Tween20 0.01% after Ni 20/12/12 Imp7BetaSuperdex75prep320ml 25 04 12001:1_UV1_280nm Imp7BetaSuperdex75prep320ml 25 04 12001:1_UV2_260nm Imp7BetaSuperdex75prep320ml 25 04 12001:1_Fractions Imp7BetaSuperdex75prep320ml 25 04 12001:1_Logbook
-10.0
0.0
10.0
20.0
30.0
40.0
50.0
60.0
mAU
80 100 120 140 160 180 200 ml
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Peak2: 182 ml
Peak1: 137 mlPeak agregate: 115 ml
Imp7 (598-C) + Imp Beta (1-442) after Ni and conc - Superdex75prep320ml 25.04.12
Optimization of Complex Formation
Strategy: Co-purification of two proteins (separately
expressed) Complex stabilization: 0.01% Tween-20
Capture: IMAC column Nadav Komornik / Oded Livnah lab
Superose 12 200ml column Superose 12 320ml column 0.01% Tween-20 through purification
Aggregate
Complex Free protein
Increasing resolution - Example: Pegylated protein 120617M1605Superose12Anal001:1_UV1_280nm 120617M1605Superose12Anal001:1_UV2_260nm 120617M1605Superose12Anal001:1_Fractions 120617M1605Superose12Anal001:1_Inject 120617M1605Superose12Anal001:1_Logbook
0
50
100
150
mAU
5.0 10.0 15.0 20.0 ml
F3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Waste
120618Superose12Prep3columns500ml001:1_UV1_280nm 120618Superose12Prep3columns500ml001:1_UV2_260nm 120618Superose12Prep3columns500ml001:1_Fractions 120618Superose12Prep3columns500ml001:1_Inject 120618Superose12Prep3columns500ml001:1_Logbook
0
50
100
150
mAU
100 150 200 250 300 350 ml
2 3 4 5 6 7 8 9 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70 72 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93Superose 12 analytical 30 x 1cm = 23ml column
Load : ~1mg protein Superose 12 preparative
3 tandem columns 250 x 1.6cm = 502ml column Load : ~25mg protein / 5ml
Vo
1 2 3
How can we get better separation between 2 and 3 ??
Can we scale-up protein loading to separate 1 from 2 and 3 ??
36
Column size
Desalting and other group separations
Volume four times the expected sample volume
Length is not so important
Fractionation
Volume ~0.5-4% times the expected sample volume
Sample volume can be increase if resolution is OK
Length 30-100 cm or more (depends of the resolution)
37
Why choose gel filtration?
Advantage
Separates by size. Complementary to IEX and HIC
Very gentle, high yields
Works in any buffer solution
Removes aggregates
Fast for buffer exchange
Mostly use in a final polishing step
Mandatory for QC
Complementary results than PAGE-SDS
Disadvantage
Limited sample volume
Poor resolution in a complex mixture
Flow-rate limitation – time consuming
Elution of diluted sample
Poor selectivity compared with SDS-PAGE
Not efficient in capture or intermediate steps
38
Troubleshooting
Lower yield than expected
Protease degradation of the protein
Adsorption to filter, valves or top of the column
Non-specific adsorption
Sample precipitate
MW of protein is not as expected
Oligomerization state of the protein is different
Protein bounds to another protein or complex
Unfolded or naturally unfolded protein
Protein has changed during storage
Ionic or Hydrophobic interactions between protein and matrix
Protein precipitate
Very broad peak elution
Different oligomeric states or protein aggregation
Sticky protein
Non specific adsorption to matrix
Protein is part of complex with different sizes
Overloading
Peak of interest is poorly resolved
Sample volume is too high
Short column
Poor selectivity or efficiency of the column
Flow rate too high
Column is dirty or not well packed
Viscous sample
39
Some molecules are highly hydrophobic, making them incompatible with fractionation via size-
exclusion chromatography (SEC). Field flow fractionation (FFF) separates macromolecules and
nanoparticles by size without a stationary phase, eliminating most of the non-ideal surface
interactions prevalent in SEC.
In an Asymmetric-Flow FFF separation channel,
macromolecules and nanoparticles are gently
pushed against a semipermeable membrane by
crossflow. Smaller particles diffuse back up towards
the center of the channel. Laminar channel flow
induces a parabolic flow velocity profile, causing
smaller particles to elute earlier.
Field flow fractionation (FFF)
40
Inhibition of HIV-1 Replication by Modulating the Oligomerization Equilibrium of the Viral Integrase
Zvi Hayouka. et al. PNAS 104 (20): 8316-8321 (2007) Friedler lab
The effect of ligand binding on the oligomeric state of Integrase
IN
IN + LEDGF/p75 361-370
IN + LEDGF/p75 361-370 (2h incubation)
IN + DNA LTR
IN + DNA LTR +
LEDGF/p75 361-370
Analytical gel filtration(superose 12)
200
125
9565Molecular weight
(gr/mol)
41
HTL435aaSuperdex200prep500mlB005:11_UV3_220nm HTL435aaSuperdex200prep500mlB005:11_Logbook
0
50
100
150
200
mAU
100 150 200 250 300 350 400 ml
NATIVELY UNFOLDED PROTEIN Case study: HTL - A natively unfolded proline-rich domain in ASPP2 that regulates its protein interactions by intramolecular binding to the Ank-SH3 domains.
Shahar Rotem et al. JBC Friedler lab
660 440 232 156 67 kDa
HTL 435aa
FoldIndex©: a simple tool to predict whether a given protein sequence
is intrinsically unfolded. Jaime Prilusky, Clifford E. Felder, Tzviya
Zeev-Ben-Mordehai, Edwin Rydberg, Orna Man, Jacques S.
Beckmann, Israel Silman, and Joel L. Sussman, 2005, Bioinformatics.
435aa
MW: 49.7kD
Superdex 200 prep. 100x2.6cm :
trimer ??
Analytical ultracentrifugation:
monomer