The Impact of Column Peak Capacity on the Multi- dimensional Chromatography of Complex ...€¦ ·...
22
The Impact of Column Peak Capacity on the Multi- The Impact of Column Peak Capacity on the Multi- dimensional Chromatography of Complex Peptide Mixtures dimensional Chromatography of Complex Peptide Mixtures Martin Martin Gilar Gilar Life Sciences Chemistry, Waters Corporation Life Sciences Chemistry, Waters Corporation ISPPP 2003 ISPPP 2003 November 12, 2003 November 12, 2003 © 2003 Waters Corporation
Transcript of The Impact of Column Peak Capacity on the Multi- dimensional Chromatography of Complex ...€¦ ·...
The Impact of Column Peak Capacity on the Multi-The Impact of Column Peak Capacity on the Multi-dimensional Chromatography of Complex Peptide Mixturesdimensional Chromatography of Complex Peptide Mixtures
Martin Martin GilarGilar
Life Sciences Chemistry, Waters CorporationLife Sciences Chemistry, Waters Corporation
ISPPP 2003ISPPP 2003
November 12, 2003November 12, 2003
© 2003 Waters Corporation
Outline
Highly complex peptide samples
Column peak capacity definition
1D RP-HPLC - where are the limits?
Peak capacity prediction
Peak capacity in 2D and MD-HPLC
Productivity of MD separation
Future development
© 2003 Waters Corporation
What is the required peak capacity?
Proteomic samples = peptides
10-50 thousand proteins
100-500 thousand peptides
Current MD-HPLC systems: peak capacity > 10 000orthogonal selectivity - 1D x 2D x …MDSCX and RP-HPLC
MS = additional separation dimension
© 2003 Waters Corporation
Column peak capacity
Column peak capacity: Maximum number of peaks that canbe separated on a column within a given gradient time.
0 Minutes 25
w13.4 %
= 0.17h w13.4 %
= 0.16w13.4 %
= 0.15 w13.4 %
= 0.17h w13.4 %
= 0.16w13.4 %
= 0.15 w13.4 %
= 0.17h w13.4 %
= 0.16w13.4 %
= 0.15
wt
P g+=1
© 2003 Waters Corporation
1D HPLC peak capacity
Column type
Lx ID (mm)
Particle
size (mm)
Void volume
(ml)
Gradient time
(s=0.0123)
Gradient time
(s=0.0246)
Gradient time
(s=0.0492)
Symmetry300 C18
50 x 4.6 5 0.66 37.5 17.8 8.9
Symmetry300 C18
150 x 4.6 5 1.86 100 50 25
Symmetry300 C18
150 x 4.6 3.5 1.90 100 50 25
Symmetry300 C18
150 x 4.6 7 1.80 100 50 25
Symmetry300 C18
300 x 4.6 5 3.46 186 93.1 46.5
Gradient time Gradient time Gradient time
PolySULFOETHYL
aspartamide 50 x 4.6 5 0.7 20 40 80
© 2003 Waters Corporation
1D HPLC peak capacity
Minutes0 100
12
3
45678
9
12 3
4 56
78
9
12 3
4 5 67 8
9
Minutes0 100
12
3
45678
9
12 3
4 56
78
9
12 3
4 5 67 8
9
25 Minute GradientP = 157
50 Minute GradientP = 249
100 Minute GradientP = 351
Symmetry300 C18, 150 x 4.6 mm, 5µm
© 2003 Waters Corporation
1D HPLC peak capacity
Strategies for increasing the column peak capacity
Long gradient time (shallow gradients)
Extend the column length + gradient time
Use small particle size sorbent
01.05015.00 =⋅=⋅∆=
gttcs
© 2003 Waters Corporation
Peak capacity prediction
1)(41
0 +⋅∆⋅∆⋅
⋅+=gttcBcBNP
NN tgtg
0
100
200
300
400
0 5000 10000 15000 200000
100
200
300
400
500
0 50 100 150 200
Column efficiency N
Peak
cap
acity
Peak
cap
acity
Gradient time (minutes)
N = 10000t0=2.4B=55∆c=0.5
t0=2.4tg=60B=55∆c=0.5
© 2003 Waters Corporation
Peak capacity prediction
1)(41
0 +⋅∆⋅∆⋅
⋅+=gttcBcBNP
Column efficiency N, Dm = 3x10-10 m2/s
5 µm
3.5 µm
2.5 µm
0
5
10
15
20
25
30
0 0.2 0.4 0.6 0.8 1
flow rate
H (u
m)
50 x 4.6 mm I.D.© 2003 Waters Corporation
Peak capacity prediction
1)(41
0 +⋅∆⋅∆⋅
⋅+=gttcBcBNP
ln k
organic content %/100
small molecule
protein
peptides
Peptides B~45-65
© 2003 Waters Corporation
Predicted versus experimental peakcapacity
181
262
351
424455
132
203
249
300337
78
136156
188220
0
50
100
150
200
250
300
350
400
450
500
0 2000 4000 6000 8000 10000 12000 14000
Column efficiency
Peak
cap
acity
B=55Dm = 3x10-10 m2/s
Five columns, 3 gradient slopes
s = 0.0123
s = 0.0246
s = 0.0492
© 2003 Waters Corporation
Peak capacity prediction
02550
75100125150175200
0
20000
40000
60000
80000
100000
0
200
400
600
800
1000
1200
1400
1600
1800
Col
umn
peak
cap
acity
Gradient time
Theoret
ical p
lates
150 x 4.6 mm column, decreasing dp
© 2003 Waters Corporation
2D-HPLC peak capacity
Orthogonal selectivity - P multiplication
SCX + RP-HPLC
P = P1D x P2D 2D
1D
© 2003 Waters Corporation
SCX-HPLC peak capacity
SCX peak capacity ~ 50 % of RP-HPLC values or less(polysulfoethyl aspartamide, PolyLC Inc.)
SCX step gradient elution - peak capacity = no. offractions
Column L x ID(SCX)
Gradient time (minutes)
Peak capacity
50 x 4.6 20 65
50 x 4.6 40 83 50 x 4.6 80 113
© 2003 Waters Corporation
2D-HPLC peak capacity
SCX step gradient elution - extreme demand on 2D
Fast 1D, very long 2D
low peak/min productivity
extensively long total analysis time
complex sample enters MS
limited number of identified peptides / dynamic range
great demand on data interpretation (software)
© 2003 Waters Corporation
2D-HPLC peak capacity
1D fractions SCX
RP column length (mm)
2D grad time RP (minutes)
Peak capacity
Time (hours)
10 150 60 2700 10
10 250 120 3800 20 10 500 240 5300 40 40 150 60 10800 40
* 5 µm sorbent
Productivity?
© 2003 Waters Corporation
Productivity of separation
1050
90130
170210
250
5
20
35
50
65
0.0
5.0
10.0
15.0
20.0
25.0
sepa
rate
d pe
aks
/ min
ute
Column length (mm)
grad
ient t
ime
Productivity =peaks / min
Best for short,efficient columnsusing fast gradient
1040
70100
130160
190220
250
5
20
35
50
65
0
50
100
150
200
250
300
350
400
450
Pea
k ca
paci
ty
Column length (mm)
grad
. tim
e gr
adien
t tim
e (m
in.)
© 2003 Waters Corporation
2D-HPLC productivity
1D fractions SCX
RP column length (mm)
2D grad timeRP (minutes)
RP-HPLC peak capacity
Total peak capacity
Time (hours)
ProductivityPeaks/min.
20 250 (3.5 µm) 240 657 13000 80 2.7
40 150 (3.5 µm) 60 354 14000 40 5.9
80 50 (3.5 µm) 15 175 14000 20 11.6
80 20 (1.8 µm) 6 178 14000 8 29.6
Short + efficient columns = best productivity
© 2003 Waters Corporation
Validity of the model
Uses assumption of porous sorbents
Dm, and B = ? Changes with peptide MW.
Model overestimates P for small dp and fast gradients
Extra-column peak broadening?
Developed for 4.6 mm I.D. (lower P for nano-columns)
Peak capacity ≠ no. of separated peaks100 components injected → 37 observed peaks (Giddings)1D & 2D will resolve ~ 37 % components (assuming that no. ofcomponents injected = peak capacity of the system)
© 2003 Waters Corporation
© 2003 Waters Corporation
Conclusions
RP-HPLC peak capacity ~ 200-400
SCX-HPLC peak capacity ~50-100
2D-HPLC peak capacityon-line ~5000-10,000off-line ~10,000-40,000
limited productivity of current setups
long analysis time (second dimension)
What is the expected sample complexity?
What is the best pre-fractionation?
Future outlook
Protein pre-fractionation prior to 2D-HPLC (peptides)
Abundant protein depletion (serum proteomics)
~ 1 µm particles
ultra-HPLC
monolithic columns
New orthogonal modes of 2D-HPLC
Progress in MS/MS and data acquisition and handling
Fast 2D-HPLC systems ~10,000 peak capacity
© 2003 Waters Corporation
Amy E. DalyAmy E. Daly
Mariana Mariana KeleKele
UweUwe D. D. NeueNeue
Tad Tad DourdevilleDourdeville
Kenneth J. FountainKenneth J. Fountain
Ying-Ying-Qing Qing YuYu
John C. GeblerJohn C. Gebler
Acknowledgements
© 2003 Waters Corporation
