f d f d h l f d dUse of Ammonium Formate to Modify Reversed‐Phase LC‐MS Analyses of Peptides and Tryptic Digests y y p yp gBarry Boyes1, 2; Darryl Johnson2; Ron Orlando2; Stephanie Schuster1; Jack Kirkland1

1Ad d M i l T h l I Wil i DE 2C l C b h d R h C U i i f G i A h GA1Advanced Materials Technology Inc., Wilmington, DE; 2Complex Carbohydrate Research Center, University of Georgia, Athens, GA

Effect of Ammonium Formate on Proteomic Analysis by LC/MSMobile Phase Modifiers for LC Separations Ammonium Formate as an Additive for LC/MS AnalysesObjectivesd d f f b l h f h h l

• TFA and Formic acid have disadvantages for LC/MS. A mix of Ammonium Formate/Formic Acid is an attractive mobile phase showing narrow and symmetrical peaks compared to Formic Acid alone

Halo Peptide ES-C18, 0.2 mm ID x 50 mm, Flow Rate 9 µL/min., T=25C, 3 pmol apoMyoglobin digest in 2 µL; 2-45% B in 15 minutes, B: 0.1% Formic acid in Acetonitrile

• Compare acid modifiers of mobile phases for high resolution reversed‐phase gradient separations of peptides.  E l t th ti b fit f dif i t i l f i id bil

Proteins from procyclic T. brucei were analyzed via LC‐MS/MS using the DDA optimized experimental 

Column: Halo Peptide ES-C18, 4.6 x 100 mm; Flow rate: 2.0 mL/min; T= 30°C;

attractive mobile phase, showing narrow and symmetrical peaks, compared to Formic Acid alone.

90

1009.27

752.4210.28

804.58NL: 8.04E5Base Peak F: ITMS + c ESI 90

1009.56

752.35NL: 5.18E5Base Peak F: ITMS + c

A: 0.1 % Formic Acid A: 0.1 % Formic Acid/10 mM Ammonium Formate• Evaluate the separation benefits of modifying typical formic acid mobile phases by addition of ammonium formate. 

• Examine the compatibility of ammonium formate as an additive to

conditions using Halo Peptide ES‐C18 0.2 x 150 mm columns (see also Poster Number MP 226).  Gradient elution with mobile phase combinations with and without ammonium formate (in Eluent A and B) were examined for peptide identifications The number of unique peptide identifications forA: Water/acid modifier; B: ACN/0.1% TFA or Formic Acid

Gradient: 1.5% to 26% B in 15 min.; Injection: 8 µL (800 ng) of synthetic peptides S1-S550

60

70

80

ve A

bund

ance

13.95908.93

Full ms [300.00-2000.00]

50

60

70

80

ve A

bund

ance 10.63

804.51

13.35690.43

Base Peak F: ITMS c ESI Full ms [300.00-2000.00] • Examine the compatibility of ammonium formate as an additive to 

formic acid mobile phase for LC‐MS analysis of tryptic digests.• Define the practical value of the use of ammonium formate modified Average Tailing Factor of S3 & S5 vs Column Load

and B) were examined for peptide identifications.  The number of unique peptide identifications for each experimental condition reflects a total collected from duplicate runs.  The combination of the Halo Peptide column with ammonium formate in both eluents yielded the most peptide identifications 

d f l d

0

10

20

30

40

Rel

ati

12.93690.27

10.93748.387.77

927.5914.77

908.595.75704.39

17.701952.54

2.741266.39

0

10

20

30

40

Rel

ativ

8.42704.49

11.28748.41 14.17

909.0319.49

1617.642.47

1127.677.97

684.2016.54

943.135.33

1549.90

• Define the practical value of the use of ammonium formate modified mobile phase in a proteomic workflow. 

Introduction3.50

4.00

ght

g g

0 1% Formic AcidS1,S2

S3S4 S5

using a 5% peptide false discovery rate.  

• Experiments were performed in duplicate using a 90 minute gradient0 2 4 6 8 10 12 14 16 18 20

Time (min)

00 2 4 6 8 10 12 14 16 18 20

0

DJ_Halo_stem_ApoMyoglobin_3pmol_2_051710_100517170709 #1914-1933 RT: 12.23-12.33 AV: 5 NL: 5.78E4F: ITMS + c ESI Full ms [300.00-2000.00]

100 754.09

DJ_Halo_stem_ApoMyoglobin_3pmol_AF_1_051810 #1945-1971 RT: 12.79-12.95 AV: 11 NL: 6.88E4F: ITMS + c ESI Full ms [300.00-2000.00]

100 754.52

IntroductionRecent developments in HPLC instruments and column packing materials are permitting faster separations, particularly for reversed‐phase analyses of peptides and protein m

0.1% TFA

2.50

3.00

ctor

at 5

% h

ei 0.1% Formic Acid

10 mM Ammonium Formate/0.1% Formic Acid20 mM Ammonium Formate/0.1% Formic Acid0.1% TFA

, Experiments were performed in duplicate using a 90 minute gradient• Protein IDs are number of protein groups using a 5% protein FDR• Total protein groups across all experiments: 131 (87 groups  commonly identified)

P id ID b f i id i 5% id FDR

50

60

70

80

90

e Ab

unda

nce

50

60

70

80

90

ve A

bund

ance

faster separations, particularly for reversed phase analyses of peptides and protein fragments.  The recent popularity of sub‐2 μm diameter particles and the new development of small diameter superficially porous particles designed for biomolecules is 

i i f i f id d i f @ 2

15 n

m

0.1% Formic Acid 1.50

2.00

Taili

ng F

ac • Peptide IDs are number of unique peptides using a 5% peptide FDR• Total unique peptides across all experiments: 1061 (287 peptides commonly identified) 

400 600 800 1000 1200 1400 1600 1800 20000

10

20

30

40

Rel

ativ

e

503.34395.19 780.691506.561231.28515.81 871.72 969.89 1992.691631.89 1831.49

400 600 800 1000 1200 1400 1600 1800 20000

10

20

30

40

Rel

ativ

e

941.18 1506.56395.19 494.21 1604.581037.00 1833.39690.02 1232.29 1963.06

permitting faster separations of peptides and protein fragments. 

Better materials and instruments for conducting high efficiency separations need to be 

mAU

@ 1.000 500 1000 1500 2000

Mass Injected (ng)

Average Peak Width of S3 & S5 vs Column Load Eluent A Eluent B Proteins IDs Peptide IDs400 600 800 1000 1200 1400 1600 1800 2000m/z

400 600 800 1000 1200 1400 1600 1800 2000m/z

g g y pcombined with appropriate methods of use. It has long been appreciated that peptide separations using typical “MS friendly” formic acid acidified mobile phases yield 

ti th t i f i t “MS f i dl ” t ifl ti id (TFA) difi d bilComparing Formic Acid PLUS Ammonium Formate versus Formic Acid for LC/MS analyses of peptides and 

f l0.1% Formic Acid/20 mM NH4 Formate 0.16

0.18

0.20

ht (m

in)

Eluent  A Eluent  B Proteins IDs Peptide IDs

separations that are inferior to “MS unfriendly” trifluoroacetic acid (TFA) modified mobile phases.  The Halo Peptide ES‐C18 column is prepared using fused‐core particles that are surface modified with an extremely stable steric‐protected C18 bonded phase, 

tryptic digests of proteins reveals:• Retention increases with ammonium concentration, accompanied by selectivity shifts, and improved 

k h0.08

0.10

0.12

0.14

h at

50%

hei

gh

99.9% H2O, 0.1% Formic Acid, 10 mM Ammonium Formate

80% ACN, 0.1% Formic Acid, 10 mM Ammonium Formate

118 825y p p ,

appropriate for the low pH conditions preferred for peptide separations. We show, consistent with many previous observations, that TFA is a most effective acid modifier for separations and that formic acid exhibits significant band broadening when used with

peak shape.• Improved sample mass load tolerance at 10 or 20 mM ammonium formate. Even at the minute 

quantities detected by online MS peak shapes are superior with the added salt0 5 10 15 0 00

0.02

0.04

0.06

Peak

Wid

th

99 9% H O 0 1% F i A id 99 9% ACN 0 1% F i A id 118 639separations, and that formic acid exhibits significant band broadening when used with the Halo Peptide ES‐C18 columns; this is universally observed for high performance column packing materials of use for peptide separations, and is thought to result from 

quantities detected by online MS, peak shapes are superior with the added salt.• Global MS signal is lower, but S/N is very similar. For a small percentage of peptides (c. 15%)  up to 10‐

fold differences in either direction of relative signal strength are observed• Band broadening and tailing using Formic Acid is driven by surface overload for basic peptides

0 5 10 15

Time (min)0.00

0 500 1000 1500 2000Mass Injected (ng)

99.9% H2O, 0.1% Formic Acid 99.9% ACN, 0.1% Formic Acid 118 639

the poor dissociation of this weak acid, particularly in acetonitrile/water mixtures.  Ammonium formate is an effective modifier, previously shown to reduce band broadening of bases including peptides The compatibility of this additive with LC MS has

fold differences in either direction of relative signal strength are observed. • Ionization charge state differences are dependent on flow rate (column ID) and are highly sequence 

dependent Determination of a systematic trend to higher or lower ionization state will require

Band broadening and tailing using Formic Acid is driven by surface overload for basic peptides, even at low quantities. Peptides with a blocked N‐terminus, and without basic residues, show very high load tolerance, narrow width and good symmetry, compared to the example of S1‐S5

99.9% H2O, 0.1% Formic Acid, 10 mM Ammonium Formate

99.9% ACN, 0.1% Formic Acid92 457

broadening of bases, including peptides. The compatibility of this additive with LC‐MS has not been fully explored for proteomic applications, and there has been no systematic analysis of the potential benefit to protein IDs for complex proteomic samples. 

dependent. Determination of a systematic trend to higher or lower ionization state will require analysis of a larger dataset.

very high load tolerance, narrow width and good symmetry, compared to the example of S1 S5 peptides or  β‐amyloid (1‐38). 

• Detection of trace impurities is enhanced with Ammonium Formate/Formic Acid mobile phase. Halo Peptide ES-C18 0 2 mm ID x 150 mm Flow Rate 4 µL/min T=25C 3 ug synthetic protein standard digest in 5 µL;

Materials and MethodsC l f HALO P tid ES C18 d d t Ad dM t i l T h l I

Column: Halo Peptide ES-C18, 2.1 x 100 mm; Flow rate: 0.5 mL/min; T= 60°C; A: Water/acid modifier; B: ACN/0.1% Formic Acid

G di t 20 M A i F t /0 1% F i A id 24% t 27% B i 20 i

Detection of trace impurities is enhanced with Ammonium Formate/Formic Acid mobile phase.

RT: 5.0 - 120.0100ce

76.160.5NL: 2.74E7

Halo Peptide ES C18, 0.2 mm ID x 150 mm, Flow Rate 4 µL/min., T 25C, 3 ug synthetic protein standard digest in 5 µL;2-45% B in 120 minutes, A: 0.1 % Formic Acid +/- 10 mM Ammonium formate, B: 0.1% Formic acid in Acetonitrile Conclusions and Future Directions

Hi h th h t RP HPLC ti d t d t i th tilit f ddi i f tColumns of HALO Peptide ES‐C18 were produced at Advanced Materials Technology Inc. (Wilmington, DE). HPLC separations used the quaternary Agilent 1100, binary 1200 SL or capillary 1100 LC systems controlled with ChemStation software. The capillary LC was 

Gradient: 20 mM Ammonium Formate/0.1% Formic Acid : 24% to 27% B in 20 min.;0.1% Formic Acid : 20% to 24% B in 20 min.

Column overloaded to show impuritiesFA

60

70

80

90

Abu

ndan

c

63.0

TIC F: ITMS + c ESI Full ms [300.00-2000.00] MS

• High throughput RP HPLC separations were conducted to examine the utility of adding ammonium formate salt to standard formic acid eluents.

• Ammonium formate addition to formic acid eluent was beneficial for reversed phase separations, improving p y y p yconnected to the ThermoFisher LTQ ion‐trap mass spectrometer via the Michrom Bioresource Advance spray source. Samples from the autoinjector were captured on the EXP Stem Trap (2 6 µL) cartridge packed with Halo Peptide ES C18 (Optimize

1 µg of β-amyloid (1-38) injected 10 µg of β-amyloid (1-38) injected

30

40

50

60

Rel

ativ

e A

51.9 58.981.371.270.9

77.156.0 68.8 71.539.038.929 57 9 50 139 429 48 445 95

p p , p gpeak shape, band width and column load tolerance for peptides and mixtures of protein digests. 

• Trace impurity detection was improved with the use of ammonium formate modified formic acid mobile phase We are coupling this with online MS as a standard approach with synthetic peptides used in our labsEXP Stem Trap (2.6 µL) cartridge packed with Halo Peptide ES‐C18 (Optimize 

Technologies), using the LTQ automated valve.Data collection for the narrow peaks obtained in this and an associated study (See Poster 

10 20 30 40 50 60 70 80 90 100 110 1200

10

20

30R 38 929.57.9 50.139.429.48.49.0 36.1 45.735.926.022.510.1 11.5 82.917.7 111.184.8 91.2 111.692.7 97.0 103.9 112.935

45

215 nm 20 mM Ammonium

Formate/0.1% Formic Acid

75

215 nm 20 mM Ammonium

Formate/0.1% Formic Acid

phase. We are coupling this with online MS as a standard approach with synthetic peptides used in our labs.• Capillary columns packed with these Fused‐core particles can be operated at very high linear velocities to permit rapid and efficient proteomic analyses using standard LC/MS instruments.

MP 226.) required optimization of the DDA and Minimum Signal setting for the LTQ. Data was searched with Mascot, and ProteoIQ (Nusep, Bogart, GA) was used for data comparisons FA/AF

10 20 30 40 50 60 70 80 90 100 110 120Time (min)RT: 5.00 - 120.00

80

90

100

danc

e 77.39

61 64

NL: 9.78E6TIC F: ITMS + c ESI Full ms [300.00-2000.00] MS

25

nce @ 2

55nce @ 2 • The ammonium formate additive is fully compatible with ion trap LC‐MS operation.

• In a proteomic workflow higher Peptide IDs are obtained with the ammonium formate additive for a complex digest sample (22 5% increase) Ongoing work will reveal whether potential problems could arise withcomparisons.

Synthetic peptides were obtained from AnaSpec (Freemont, CA) or from ThermoFisher, in the case of the Retention Standard Mix (Mant and Hodges), the S1‐S5 sequences are:

FA/AF

50

60

70

80

ve A

bund 61.64

52.32 63.0257.915

15

Absorba

n

0.1% Formic Acid

15

35

Absorba

n

0.1%Formic Acid

digest sample (22.5% increase). Ongoing work will reveal whether potential problems could arise with identifying post translational modifications. 

S1     RGAGGLGLGK‐Am S2 Ac‐RGGGGLGLGK‐AmS3 Ac‐RGAGGLGLGK‐Am

20

30

40

Rel

ativ

79.35 83.1372.9864.7541.14 53.7831.90 43.6739.886.25 7.73 85.6545.8729.747.90 28.33 87.3719.658.10 96 5427 90

‐5

5A

‐5

15A AcknowledgementsSupport of this project by the NIH (GM077688, Kirkland) and (P41RR018502; AI077538, Orlando) is gratefully 

S3 Ac‐RGAGGLGLGK‐Am S4 Ac‐RGVGGLGLGK‐AmS5 Ac‐RGVVGLGLGK‐Am

10 20 30 40 50 60 70 80 90 100 110 120Time (min)

0

108.10 96.5427.9010.56 97.08

117.04107.060 10 20Time (min)

0 10 20Time (min)

acknowledged.