TO DOWNLOAD A COPY OF THIS POSTER, VISIT WWW.WATERS.COM/POSTERS ©2013 Waters Corporation

INTRODUCTION

Most proteomic samples, when digested,

generate peptides with similar distributions of

hydrophobicities [1]. The complexity of these

samples requires orthogonal methods of

separation (in LC or MS) to identify and quantify

all detectable peptides in a sample. Changing

the pH between 10 and 2.6 has been shown to

dramatically change the selectivity of a reverse-

phase separation [2]. The need for multiple dimensions of separation must be balanced by

the analysis time for each sample. For this

reason, a high throughput 2D RP/RP method

was developed and used in combination with ion

mobility to analyze peptides from human

placental samples.

INCREASED THROUGHPUT FOR 2D LC IN THE ANALYSIS OF HUMAN PLACENTAL SAMPLES

Martha Stapels1, Keith Fadgen1, J. Will Thompson2, M. Arthur Moseley2, and James Langridge1 1Waters Corporation, Milford, MA; 2Duke University School of Medicine, Durham, MA

METHODS

LC/MS

Prototype nanoACQUITY UPLC® System with 2D Technology SYNAPT™ G2-S HDMS™

Sample Preparation

Proteins were treated using a protocol to generate TRIzol-

soluble and sonication–soluble fractions and then reduced,

alkylated, and digested in-solution with trypsin.

First Dimension

Column: 300 µm x 5 cm XBridge™ C18 (5 µm)

Gradient formation

Discontinuous step gradient

Eluent A: 20 mM ammonium formate pH 10.0 Eluent B: acetonitrile

Second Dimension

Trap: 180 µm x 2 cm Symmetry C18 (5 µm)

Column: 75 µm x 15 cm HSS T3 C18 (1.8 µm)

Eluent A: 0.1% formic acid in water Eluent B: 0.1% formic acid in acetonitrile

Online Dilution with RP/RP

To maximize sample recovery on the second dimension trap

column from the organic-containing fractions, an aqueous flow was delivered with the 2nd dimension pump, and mixed with

the eluted fraction prior to trapping.

MS Data processing

Data were processed and searched with ProteinLynx Global

Server (PLGS 2.5.2) with IdentityE informatics.

RESULTS

Figure 1. Fluidic layout for the high throughput RP/RP analysis

with online dilution. Loading on one trap column takes place

during elution of the other.

Figure 7. Summed integrated TIC for the 28 fraction experiment.

The cuts for a 3 fraction experiment are highlighted.

In order to determine optimum fractionation with the high-low

pH RP/RP system, a 28 fraction

experiment was performed to map out the separation space. First dimension cuts of 5, 7, 8-28 in 1% steps, 30, 32, 35, 38,

and 50% acetonitrile were chosen to properly interrogate the

separation space. Chromatograms for each fraction are in Fig-

ure 4.

For each fraction, the total ion chromatogram (TIC) was inte-

grated across its entire separation space. The results were

summed, normalized, and plotted against the percent ACN used for each cut. Figure 7 shows the optimum fractionation

strategy for the RP/RP separation space. As an example. the

cuts for a 3 fraction experiment are depicted.

Using a total analysis time of 4 hours, human placental sam-

ples were run with 3 or 6 fractions in the first dimension. Rep-

resentative chromatograms are shown in Figures 8 and 9. As

shown in Figures 11 and 13, the 6 fraction method consis-

tently yielded more identifications.

Figure 4. BPI Chromatograms for the 28 fraction experiment

used for optimization.

ACKNOWLEDGEMENTS

The authors would like to thank Dr. Monique Chireau of Duke

University for the human placenta samples used in this study.

CONCLUSION

Optimized prediction strategy for high/low

pH RP/RP separations was developed

Using high throughput 2D, more proteins

and peptides are identified per hour

Increasing peak capacity with orthogonal

methods is beneficial for identifying and

quantifying proteins in complex mixtures

Ion mobility resolves interfering peptides

and requires no additional analysis time

Increased fractionation improves

identifications when total analysis time is

held constant

REFERENCES

1. Geromanos S. et. al, Proteomics. 2011, 11, 1189-211

2. Gilar M. et. al, J. Sep. Sci. 2005, 28, 1694-703

Figure 13. Number of identified proteins and peptides from

both a sonicated and detergent extracted placental sample,

using either a 3 or 6 fraction 2D method over 4 hours.

Figure 9. Example 6 fraction results for human placenta

samples in 4 hours.

Figure 11. Identified proteins from the two protein fractions,

using either 3 or 6 fraction 2D analyses in 4 hours. 85% of the

proteins overlapped between the methods, but the 6 fraction analysis yield more identifications.

8

6

4

2

0

Tota

l Io

n C

urr

ent

(10

6)

50403020100

Time (min)

75µm x 150 mm HSS T35 to 50% ACN in 45 min500 nL/min

3.5

3.0

2.5

2.0

1.5

1.0

0.5In

tegra

ted T

IC (

10

9)

5040302010

%ACN - First Dimension

Ion mobility

13.3% ACN

50% ACN

19% ACN

Figure 8. Example 3 fraction results for human placenta

samples in 4 hours.

Sonication Fraction1609 Proteins

TRIzol Fraction1466 Proteins

3 Fraction(1214 Proteins)

6 Fraction(1429 Proteins)

180 1034 395

3 Fraction(1121 Proteins)

6 Fraction(1293 Proteins)

173 948 345

Figure 2. Alignment of low energy and elevated energy ions by

LC retention time (top, MSE) and also with the orthogonal

dimension of separation by mobility (bottom, HDMSE).

Figure 3. A plot showing the three resolving axes used in this

analysis: LC retention time, m/z, and ion mobility drift time.

Each of these axes has a given resolving power (N) and the overall system resolving power is given by the product of all

three and the fraction of occupied bins. The incorporation of

ion mobility into this analysis increases the peak capacity of

the system by at least an additional order of magnitude.

Figure 6. The integrated response as a function of %ACN

used for the 1st dimension steps.

Figure 5. For each fraction, the response from all of the ions

detected was integrated into a single value.

Figure 10. Left: Interfering peptides are often detected at

same retention time and m/z. Right: Interfering peptides are

resolved by orthogonal gas phase separation.

Figure 12. Observed peptide splitting between fractions for

triplicate injections of 2 human placenta samples. On average,

91% of the peptides were observed in only 1 fraction.