CaptiveSpray: A New Ionization Technique to Maximize Speed, Sensitivity, Resolution and Robustness in LCMS Protein Biomarker Quantitation Kerry Nugent, Yixin Zhu and Peter Kent Michrom Bioresources, Inc, Auburn, CA Brett Phinney and Rudy Alvarado UC Davis Proteomics Center
Introduction NanoLC coupled with NanoSpray MS has become the gold standard for proteomics LCMS over the past decade due to the high sensitivity that can be achieved at nanoliter flow rates, but this technique lacks the speed and robustness required for high throughput protein biomarker quantitation. Although many potential proteins biomarkers have been discovered over the past few years, a large number of analyses must be run to validate these proteins as biomarkers and an even larger number of analyses will be required to analyze these biomarkers in clinical applications. This study introduces a new nano-capillary LC protocol coupled with CaptiveSpray MS, which combines the ease of use, speed and robustness of LC-ESI/MS with the high sensitivity and high resolution of nLC-NSI/MS for protein biomarker quantitation. After identifying potential biomarker proteins in tissue samples, serum samples were depleted of high abundance proteins (HAP), digested with trypsin and then analyzed by HT cLC/CSI/MRM-MS, to assess the LOD, LOQ, dynamic range, precision and robustness of this new methodology.
Experimental Samples
Human Prostate Tissue Samples Control and Prostate Cancer Patient Serum
Sample Preparation
Advance Bio-Cool Liquid Sample Handler
HPLC Separations
Paradigm MS4-NC MDLC
CSI Interface
Advance CaptiveSpray Source
MS Detection
LTQ MS (Full Scan MSMS) TSQ MS (MRM MS)
Evolution of Ionization for Proteomics LC/MS 1985
1995
High Flow Gas
HPLC Column
HV
Ambient Lab Air
Low Flow Gas MS Inlet
MS Inlet
HPLC Column
2005
HV
HPLC Column
MS Inlet HV
Conventional ESI-MS
Nanospray NSI-MS
CaptiveSpray CSI-MS
In the 80s, conventional ESI revolutionized biomolecule analysis using LC/ESI-MS (50-5000 ul/min). ESI utilizes a high sheath gas flow to desolvate ions but this excessive gas dilutes the sample ions such that only a small percentage of the sample gets into the MS.
In the 90s, Nanospray (NSI) made nLC/NSI-MS (10-1000 nl/min) the gold standard for proteomics research. Like conventional ESI, Nanospray is concentration dependant, so sensitivity improves as the LC flow rate decreases, but low flows limit capacity, throughput and robustness.
The Advance source provides the next step in the evolution of LC/MS (0.1-100 ul/min), using the vacuum of the MS to pull in gas around the spraytip, desolvating and funneling all the sample ions into the MS. It’s “Plug and Play” operation provides ESI robustness with NSI sensitivity.
Advance CaptiveSpray Source The Advance CaptiveSpray source is a revolutionary LCMS interface with a patented design that delivers unmatched sensitivity, resists plugging, and provides robustness for even the most complex proteome samples. CaptiveSpray provides a gas vortex that sweeps the emitter spray tip, desolvating and funneling the ions into the MS. A vacuum seal to the MS helps draws all of the sample ions into the MS independent of LC flow rate, providing the robust operation of ESI with the sensitivity of Nano Spray.
High Voltage (from MS)
HPLC Column
Gas Inlet
Spray Tip MS Inlet
CaptiveSpray Operation
Conventional Spray
CaptiveSpray
Unfocused spray from the emitter allows some ions into the MS.
Gas vortex around the spray concentrates and focuses ions into the MS.
CaptiveSpray Bridges the Gap Detection Range (fmol)
1000 100
ElectroSpray
10 1.0
CaptiveSpray
0.10 0.010
NanoSpray
0.001
0.05
0.50
5.00
50.0
Flow Rate (µL/min)
500.0
5000
NSI vs CSI: ~5-10x Throughput Gain 100
1 ug E. coli Digest
90
75µ x 150 mm 3µ 200Å Magic C18
70
Relative Abundance
5-45B in 120 min
NSI
80
Flow = 200 nl/min
60 50 40 30 20 10
867 Proteins ID
0 0
20
40
60
80
100 120 T i m e (m i n )
140
160
180
200
100
10 ug E. coli Digest
90
200µ x 150 mm 3µ 200Å Magic C18
70
Flow = 2 µl/min 942 Proteins ID
Relative Abundance
5-45B in 30 min
CSI
80
60 50 40 30 20 10 0 0
5
10
15
20 25 T i m e (m i n )
30
35
40
ESI vs CSI: ~50-100x Sensitivity Gain 500 fmol BSA
ESI
2.1 x 50 mm 1.9µ 100Å C18 5-45B in 8 min Flow = 200 µl/min 8600 PSI S = 1.3 x 10E6
1 0 0
10 fmol BSA
CSI
9 0
8 0
5-45B in 8 min Flow = 4 µl/min 3200 PSI
7 0 Relative Abundance
0.2 x 50 mm 2.7µ 90Å C18
6 0
5 0
4 0
3 0
2 0
1 0
S = 2.4 x 10E6
0 0
1
2
3
4
5
6 T im
e
7 ( m
in )
8
9
1 0
Identification of Protein Biomarkers Although comprehensive (shotgun) proteomics has been useful for identifying thousands of proteins in a broad range of samples, the field of functional proteomics has grown rapidly in recent years as researchers look at specific cellular pathways to learn more about the role of specific proteins in both normal and diseased cell states. Techniques like affinity chromatography and immunoprecipitation (IP) are often used to isolate specific proteins prior to LCMS analysis. Using a hypothesis driven approach for targeted proteomics, specific proteins of interest can be isolated, identified, characterized and quantified by LCMS. Relative quantification techniques including ICAT, ITRAC, SILAC and TMT can be used to compare control and diseased tissues and identify up and down regulated proteins that may be useful as biomarkers. Since these potential biomarkers are often present at low abundance in complex proteome samples, LCMS analysis requires high sensitivity, high selectivity and high throughput for optimum results.
TMT LCMS of Human Prostate Cells 45
40
35
Relative Abundance
The upper trace shows the TIC of a combined human prostate cell sample, where the proteins in the normal cells were labeled with one Tandom Mass Tag (TMT) and the proteins in the cancerous cells were labeled with a second TMT.
50
30
25
20
15
10
5
0 0
2
4
6
8
10
12
14
16
18
20 22 Tim e (m in)
24
26
28
30
32
34
36
38
40
5 .0
The lower trace shows the LC-MRM/MS of targeted peptides from the up and down regulated proteins (two peptides per protein and two transitions per peptide) in the mixed TMT prostate cell sample.
4 .5
4 .0
3 .5
3 .0
2 .5
2 .0
1 .5
1 .0
0 .5
0 .0 0
5
10
15
20 T im e (m in )
25
30
35
40
Validation of Potential Protein Biomarkers Once potential biomarkers have been identified and characterized, a significant effort is required to validate these proteins to insure they are good candidates for diagnostic and therapeutic applications. Since it is difficult to obtain tissue samples from patients in clinical applications, physiological fluids such as serum, plasma, urine or bile are generally used for biomarker validation. For this study, we chose two signature peptides from each of two upregulated proteins found in the human prostate cancer cells. These signature peptides were then screened in prostate cancer patient serum to test the validity of our methodology. A high sensitivity and high throughput LC-CSI/MRM-MS assay was developed to test linearity, detection limit, selectivity and robustness. The results of this validation phase indicated that although this method could be used to screen serum from prostate cancer patients, sample preparation and stable isotope standard addition would be required for a quantitative assay.
HT LC-CSI/MRM-MS of Biomarker Peptides
R T : 2 .2 6 AA: 60465
100 0
0
R T : 2 .2 6 AA: 49113
100 0
R T : 2 .9 5 AA: 29423
100
100
0 2
3
4
1
T i m e (m i n )
R T : 1 .7 7 AA: 169574
100 0
R T : 2 .2 6 AA: 83174
100 0
0
3
100 0 100
2 T i m e (m i n )
3
4
2 3 T i m e (m i n )
0
RT : 2 .2 7 AA: 50270
100
1
2
3
4
0
1
4
2
0
3
1
2 3 T i m e (m i n )
1
100 0
4
0
4
R T : 1 .7 8 AA: 99846
100 0
R T : 2 .2 8 AA: 51552
100 0
RT : 2 .9 6 AA: 36408
100
3
R T : 1 .4 1 AA: 63378
100
RT : 2 .2 8 AA: 59184
100
2 T i m e (m i n )
R T : 2 .9 6 AA: 29108
100
0 0
R T : 2 .9 6 AA: 34485
0
4
RT : 1 .7 8 AA: 108106
0
RT : 2 .9 5 AA: 32581
100
R T : 2 .2 7 AA: 56543
100
100
RT : 1 .4 1 AA: 104802
100
RT : 2 .2 7 AA: 63655
100
0
T i m e (m i n )
RT : 1 .7 7 AA: 131686
100
R T : 1 .7 7 AA: 129017
100
0 0
RT : 1 .4 0 AA: 96066
0
0
0
RT : 2 .9 6 AA: 34899
100
0
1
100
0
0
0
R T : 2 .9 4 AA: 35031
0
RT : 1 .7 7 AA: 108280
0
RT : 2 .9 4 AA: 34417
100
0
1
0
T i m e (m i n )
R T : 2 .2 6 AA: 44548
100
0 0
100
4
R T : 1 .7 6 AA: 96248
0
R T : 2 .9 5 AA: 53588
100
2
R T : 1 .3 9 AA: 102528
100 Relative Abundance
Relative Abundance
0
RT : 2 .2 5 AA: 52094
T i m e (m i n )
R T : 1 .4 1 AA: 131315
100
0
0 0
Relative Abundance
1
100
100
0
0
RT : 1 .7 5 AA: 143496
0
R T : 2 .9 5 AA: 37315
100
0
Relative Abundance
0
R T : 1 .7 7 AA: 126898
R T : 1 .4 1 AA: 99197
100
Relative Abundance
100
0
RT : 1 .4 0 AA: 89901
100
Relative Abundance
R T : 1 .7 7 AA: 123771
RT : 1 .3 9 AA: 101082
100 Relative Abundance
0
Relative Abundance
Relative Abundance
R T : 1 .4 0 AA: 92174
100
Relative Abundance
R T : 1 .4 0 AA: 91443
100
0
0
1
2 T i m e (m i n )
3
4
0
1
2
3
4
T i m e (m i n )
These ten analyses of depleted serum protein digests from prostate cancer patients show the presence of the two potential protein biomarkers at varying levels. These results indicate that this 5 minute LCMS assay could be used to rapidly screen serum for the potential biomarker proteins.
LCMS Quantitation of Protein Biomarkers For routine quantitation of protein biomarkers in physiological fluids, a robust, high throughput, high sensitivity quantitative method is required. As with quantitation of small molecule drugs and metabolites, some degree of sample preparation is required to minimize sample matrix effects and isotopic internal standards are necessary for reproducible results. For this study, we used an affinity column (Sigma Proteoprep-20) to remove the top 20 most abundant proteins in human serum prior to digestion and LCMS analysis. Prior to the affinity cleanup, the four signature peptides were added to female control human serum at levels from 500 amol/ml to 5 pmol/ml (0.01 – 100 ng/ml protein) and stable isotope versions of each peptide were added to the spiked controls and samples. This high sensitivity assay only required 10 µl of serum sample, resulting in minimal matrix interference and a very robust method.
HT LCMS Biomarker Quantitation R T : 1 .3 6 AA: 118
100 0
Female (0.01ng/ml)
RT : 1 .3 3 AA: 1055
100 0
R T : 1 .3 6 AA: 51650
100 0
0
R T : 1 .8 8 AA: 197
RT : 1 .8 5 AA: 166
0
Relative Abundance
100 R T : 1 .8 8 AA: 16367
100 0
R T : 2 .2 7 AA: 142
100 0
0
RT : 1 .8 8 AA: 16285
100 0
RT : 2 .2 4 AA: 201
100 0
R T : 2 .2 7 AA: 9861
100
Male (Normal PSA)
RT : 1 .3 6 AA: 51712
100
100
RT : 2 .2 7 AA: 9805
100
0
0
R T : 2 .9 8 AA: 377
100 0
RT : 2 .9 8 AA: 858
100 0
R T : 2 .9 8 AA: 19145
100
RT : 2 .9 8 AA: 19093
100 0
0 0
1
2
3
4
0 .0
5
0 .5
1 .0
1 .5
2 .0
T i m e (m i n )
R T :
0 .0 0
R T :
- 5 .0 2 R T : 1 .3 6 A A : 1 3 7 2 1 5
1 0 0
Female (10 ng/ml)
5 0
0 .0 0
R T : 1 .3 3 A A : 1 8 2 8
3 .5
4 .0
4 .5
5 .0
Male (Elevated PSA)
5 0
R T : 1 .3 6 A A : 5 1 7 1 2
1 0 0
2 .5 3 .0 T i m e (m i n )
- 5 .0 2
1 0 0
0
0
R T : 1 .3 6 A A : 5 1 6 5 0
1 0 0 5 0
5 0
0
0
R T : 1 .8 7 A A : 8 1 3 9 1
1 0 0
R T : 1 .8 8 A A : 2 1 8 8
1 0 0 5 0
5 0
R T : 3 .4 2 A A : 3 3 4
0
0
R T : 1 .8 8 A A : 1 6 2 8 5
1 0 0
Relative Abundance
Relative Abundance
The traces at the right show HT LC/MRM-MS analysis of the LLOQ and ULOQ spiked controls, as well as two samples from males previously screened using PSA. Initial results indicate that these potential biomarkers could be better suited for prostate cancer detection than PSA, but significantly more samples will need to be run using this HT assay to validate these proteins as prostate cancer biomarkers.
RT : 0 .0 0 - 5 .0 2
R T : 0 .0 0 - 5 .0 2
5 0 0
R T : 2 .2 5 A A : 4 2 8 6 5
1 0 0
0
R T : 2 .2 7 A A : 1 4 0 8
5 0
R T : 3 .4 0 A A : 1 4 5
0
5 0
1 0 0
5 0
0
R T : 2 .2 7 A A : 9 8 0 5
1 0 0
R T : 1 .8 8 A A : 1 6 3 6 7
1 0 0
R T : 2 .2 7 A A : 9 8 6 1
1 0 0 5 0
5 0
0
0
R T : 2 .9 9 A A : 2 5 8 4 9 3
1 0 0 5 0
5 0
R T : 0 .9 4 A A : 2 8 1
0
R T : 2 .9 8 A A : 4 6 6 0
1 0 0
0
R T : 2 .9 8 A A : 1 9 0 9 3
1 0 0
R T : 2 .9 8 A A : 1 9 1 4 5
1 0 0 5 0
5 0 0
0 0
1
2
3 T im e
(m i n )
4
5
0
1
2
3 T im e
(m i n )
4
5
Conclusions ¾ CaptiveSpray (CSI) Provides Higher Sensitivity Than ESI ¾ CSI Offers High Throughput and Robustness vs NSI ¾ Paradigm LC Provides Fast Gradients at Low Flow Rates ¾ Biomarker Identification Throughput Enhanced by CSI ¾ Biomarker Validation Can be Improved Using CSI-MS ¾ Biomarker Quantitation is Possible With LC-CSI-MRM/MS
