IntroductionNative mass spectrometry (MS) has emerged as a powerful technique for studying protein-ligand interactions and elucidating the structure of macromolecular assemblies, including both soluble and membrane protein complexes. Native MS relies on maintaining a protein’s natural folded state and associated non-covalent interactions for analysis.

A few years ago, we introduced an Orbitrap™-based mass spectrometer called the Thermo Scientific™ Exactive™ Plus EMR mass spectrometer specifically designed for native MS. The instrument provided high mass resolution (>140,000), extended mass range up to 20,000, and the ability to perform all-ion fragmentation. The limited mass range up to 20,000 and low sensitivity at high m/z made it impossible to tackle large molecules such as whole viruses or protein-nucleic acid complexes. Also, this instrument was limited in terms of performing native top-down analysis due to the poor fragmentation of protein complexes into subunits in the front end of the mass spectrometer and due to the lack of a quadrupole to isolate specific subunit ions for further fragmentation in the HCD collision cell.

This technical note examines the technology innovations that enable the Thermo Scientific™ Q Exactive™ UHMR (Ultra High Mass Range) Hybrid Quadrupole-Orbitrap™ mass spectrometer to go beyond the capabilities of previous MS instruments in allowing interrogation of heteromeric protein assemblies using top-down pseudo-MS3 approaches. Experiments were

Authors Eugen Damoc, Kyle Fort, Maria Reinhardt-Szyba, Mikhail Belov, Alexander Makarov; Thermo Fisher Scientific, Bremen, Germany

Rosa Viner; Thermo Fisher Scientific, San Jose, CA

Albert Konijnenberg; Thermo Fisher Scientific, Eindhoven, Netherlands

Keywords Q Exactive UHMR mass spectrometer, native top-down MS, native MS, protein complexes, Orbitrap

Advancing native top-down MS analysis of non-covalent protein complexes: The Thermo Scientific Q Exactive UHMR mass spectrometer

TECHNICAL NOTE 65379

2

performed using E. coli GroEL, rabbit 20S proteasome, LmrP membrane protein, and E. coli AmtB membrane protein complex as model systems.

Optimized Q Exactive Hybrid Quadrupole-Orbitrap technology solves native MS challengesAs analytes have gotten larger and more complex, mass spectrometry capabilities have needed to advance rapidly to keep pace. A combination of enhancements to hybrid quadrupole-Orbitrap technology have come together in the Q Exactive UHMR mass spectrometer to allow direct detection of large, intact proteins and protein complexes with unprecedented resolution and sensitivity.

The Q Exactive UHMR mass spectrometer3,4 provides the ability to perform pseudo-MS3 for native top-down analysis and transmission of very high m/z ions through the implementation of both hardware and software enhancements (Figure 1). The most important of these enhancements were 1) pulsed trapping of ions in the injection flatapole region, a process called “in-source trapping” (Figure 1 inset) and 2) the reduction of the frequency of RF voltages applied to the injection and bent flatapoles, quadrupole, transfer multipole, C-trap, and HCD cell.

Figure 1. Schematic of the Q Exactive UHMR Hybrid Quadrupole-Orbitrap mass spectrometer with in-source trapping (inset)

+60 V

2, 3 Shown in the Figure 1 inset, the injection flatapole is pulsed down to a negative voltage to improve desolvation of large protein complexes, while the inter-flatapole lens is maintained at a high positive potential to prevent ions from eluting out. Trapping is followed by restoration of the voltage levels, allowing low-energy elution of ions into the bent flatapole (Advanced Active Beam Guide).

4 The bent flatapole guides and focuses ions using an axial DC field and a focusing RF field, enhancing sensitivity.

5 The high-mass quadrupole operates at a three-fold lower frequency than its predecessor and can be used to select ions with an m/z up to 25,000.

2, 4–8 The RF frequencies of all ion routing multipoles—the injection and bent flatapoles, quadrupole, transfer multipole, and HCD cell—are reduced to improve ion transmission.

9 High mass ions are efficiently injected into the Orbitrap mass analyzer by adjusting the slew rate of the high-voltage pulse that captures ions in the analyzer.

3

In-source trapping allows controllable, highly efficient desolvation of large ions and the optional application of energy to dissociate these ions into subunits in the injection flatapole region prior to their release for subsequent mass analysis, including pseudo-MS3 analysis.

The reduced RF voltages increase the efficiency of high m/z ion transfer and utilization, delivering orders of magnitude higher sensitivity, and thus much lower sample volume requirements. Multiply charged ions are detected with such efficiency that individual ions can be detected within milliseconds, providing the ultimate analytical sensitivity needed for minute sample volumes.

Other developments include gas pressure control, an increase in the maximum HCD energy from 200 to 300 V, and an adjustment of the voltage ramp rate of the Orbitrap mass analyzer, which facilitates transmission of very high m/z ions from the C-trap into the Orbitrap mass analyzer.

Pseudo-MS3 analysisThe improvements in performance provided by the modifications described earlier can be demonstrated in pseudo-MS3 experiments applied to non-covalent protein complexes. A pseudo-MS3 analysis begins by using moderate desolvation energy and transferring the intact protein complex through the mass spectrometer without fragmentation to produce its MS1 spectrum. Next, dissociation of the protein complex into its subunits in the injection flatapole region yields the MS2 spectrum. Following that, quadrupole selection and subsequent fragmentation of individual subunits in the HCD cell provide the MS3 spectrum that can be used for sequence analysis.

Native MS and native top-down analysis of the GroEL protein complex The GroEL protein complex is a molecular chaperone required for the proper folding of many proteins in cells. Using in-source trapping, efficient desolvation and fragmentation of the GroEL 14-mer complex into its monomer and stripped complexes (13-mer and 12-mer) was achieved (Figure 2). When followed by quadrupole selection of the monomer and subsequent fragmentation in the HCD cell, a total of 112 b and y ions were identified, which represents 21% of the residue cleavages.

Thermo Scientific Q Exactive UHMR system

4

Figure 2. The native MS and native top-down analysis of the GroEL protein complex enabled identification of a total of 112 b and y ions, representing 21% of the residue cleavages.

5000 10000 15000 20000 25000

m/z

0

20

40

60

80

1005361.9769

2696.3860

13970.33329051.7368

15069.929612460.5581 18327.5454 21500.2037 25558.2598

500 1000 1500 2000 2500 3000 3500 4000

m/z

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

Rel

ativ

e A

bund

ance

1406.7993z=3 2682.5039

z=21046.5243z=1

1789.0018z=3 2753.4817

R=200,000 @ m/z 400

z=22109.6936z=2

2604.7175z=2

3019.5930z=3

2330.5786z=6

885.3068z=1 3801.2073

z=2

MS3

5000 10000 15000 20000 25000 30000 35000 40000m/z

0

20

40

60

80

100

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

Rel

ativ

e A

bund

ance

Rel

ativ

e In

tens

ity

Rel

ativ

e A

bund

ance

0

20

40

60

80

100

2724.5415

MS2

MS114-mer

14-mer

13-mer

12-merMonomer

Quadrupole selection

0

20

40

60

80

100800746.38

400454.25

500000 1000000M ass

ReSpect

deconvolution

(in-source trapping)

. . .

MS3 (deconvoluted)

5

Figure 3. Native MS and MS2 (HCD) analysis of the rabbit 20S proteasome complex with MS-level deconvolution results (top inset). Sequentially increasing the MS2 HCD energy revealed the loss of the alpha-6, alpha-2, and alpha-5 subunits.

Native MS and native top-down analysis of rabbit 20S proteasomeThe rabbit 20S proteasome is a heterogeneous protein complex composed of 28 subunits arranged into four stacked heptameric rings. Its two outer rings contain seven non-identical alpha subunits and its two inner rings contain seven non-identical beta subunits.

The native MS analysis of the rabbit 20S proteasome produced a baseline resolved charge envelope for the

intact, 28-subunit complex with a measured molecular mass of 717,023.187 Da after deconvolution using the ReSpect™ algorithm (Positive Probability Ltd.) (Figure 3 inset).5 The deviation from expected mass was <0.082%. Native MS2 analysis of the intact 20S proteasome complex using sequentially increased HCD energy revealed the dissociation of the alpha-6, alpha-2, and alpha-5 subunits (Figure 3).

5000 10000 15000 20000 25000 30000 35000 40000m/z

0

20

40

60

80

1000

20

40

60

80

1000

20

40

60

80

100

Relat

ive A

bund

ance

0

20

40

60

80

10011753.0811564.85

11950.97

11377.08 12156.19

12364.7411204.72

10907.21 12596.14

11948.55

12153.0811561.65

12358.951681811377.61

11212.251707.83

1821.70 17674.7716416.8919144.571951.71 16033.59

28778.451986.3226492.64

12069.762151.7331506.942347.21 11681.79

2868.71 33098.0011173.28

1821.68

18621 89..17233.2912069.88 30080.09

11680.0133078.4211171.74

- α6,- α2

- α6

717023.187

200000 400000600000 800000 1000000

Mass

ReSpect deconvolution

Deviation from expected mass: < 0.082%

MS2 (HCD, 160 eV)

MS2 (HCD, 200 eV)

MS2 (HCD, 220 eV)

MS1

- α6, - α2, - α5

818

17674.77

19144.5728778.45

26492.64

31506.94

33098.00

621 89

600000 800000 100

Mass

ReSpectdeconvolution

17235.09

6

Figure 4. Native MS2 (in-source-trapping) analysis of the rabbit 20S proteasome complex showing deconvoluted monoisotopic masses and mass accuracies. Six of the seven subunits were confidently identified (bottom).

Native MS2 analysis of the intact 20S proteasome complex using in-source trapping showed dissociation of six out of seven alpha subunits. The Q Exactive UHMR MS isotopically resolved all six dissociated subunits at a resolution setting of 200,000 and provided accurate monoisotopic masses (Figure 4).

Native top-down pseudo-MS3 experiments allowed unambiguous identification of the alpha-6, alpha-2, and alpha-5 subunits. To obtain the data shown in Figure 5, the quadrupole was used to isolate the 16+ charged ion at m/z 1707, and then HCD fragmentation was performed in the HCD cell. The pseudo-MS3 HCD spectrum acquired at a resolution setting of 200,000 at m/z 400 allowed unambiguous identification of the alpha-6 subunit.

1200 1300 1400 1500 1600 1700 1800 1900 2000m/z

0

20

40

60

80

100

Rela

tive

Abun

danc

e

1607.4614R=70704

z=17

1707.8005R=70404

z=16

1518.2137R=67104

z=18 1821.6521R=71704

z=151295.5345R=120306

z=1

1343.4070R=117906

z=11475.4489R=112506

z=1 1931.5190R=95906

z=2

1739.5308R=102206

z=11844.4469R=70104

z=14

1643.5110R=105006

z=1

1568.7012R=108406

z=1

1289.6433R=119906

z=1

21000 22000 23000 24000 25000 26000 27000 28000 29000 30000 31000m/z

0

10

20

30

40

50

60

70

80

90

Rela

tive

Abun

danc

e

27292.69226

25793.19448 27746.47770

26420.12217 28392.81042 29375.9862121356.5729431006.88560

-1.8 ppm

-3.7 ppm -5.6 ppm

-3.6 ppm

MS2 (deconvoluted)

-3.2 ppm

MS2 (In-source trapping)

α2

α4

α6

α3+P

α5

α7

7

Accurate intact analysis of membrane proteinsMembrane protein complexes are generally hydrophobic, requiring highly heterogeneous micelle assemblies for solubilization. This makes them extremely challenging for many biophysical analytical methods, including native MS. With in-source trapping, the Q Exactive UHMR mass spectrometer can release intact membrane-protein assemblies directly in the injection flatapole region from a variety of detergent micelles and membrane mimetics. By varying the desolvation energy, protein subunits can be released for top-down sequencing or, with very gentle

Figure 5. Native top-down pseudo-MS3 analysis of the rabbit 20S proteasome alpha-6 subunit

activation, membrane proteins bound to multiple ligands can be retained for whole complex MS analysis.

Shown in Figure 6, in-source trapping allowed efficient removal of the detergent micelles for accurate intact mass determination of the AmtB trimer. Native top-down analysis of this membrane protein complex with low ppm mass accuracy resulted in high-confidence identification of AmtB protein, with 75 b and y matched ions representing 19% of the residue cleavages.

C:\Users\...\20180420_20sprote /02/40emosa 18 15:34:56

5000 10000 15000 20000 25000 30000 35000 40000m/z

0

20

40

60

80

100

Rel

ativ

e Ab

unda

nce

1707.82R=1502

28359.35R=200

30855.03R=200

24461.35R=200

33067.69R=100

20643.31R=300

400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200m/z

0

20

40

60

80

100

Rel

ativ

e Ab

unda

nce

1739.53R=101406

z=1

1665.50R=103506

z=11522.48R=100406

z=31288.02

R=108706z=3

1396.99R=102306

z=4

1931.02R=91006

z=2

773.41R=148706

z=1

1094.59R=122206

z=1

1197.03R=97506

z=5

907.39R=132706

z=1

1742.54R=93702

z=1674.35

R=156606z=1

1862.31R=89006

z=3

423.50R=208000

z=?

603.31R=119200

z=?

2126.62R=68404

z=2

20180420_20sproteasome_XT_00001_M__180420171839 #2 RT: 2.00 AV: 1 NL: 1.40E2T: FTMS + p NSI sid=100.00 Full ms2 1707.90@hcd75.00 [350.00-10000.00]

2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000 24000 26000m/z

0

20

40

60

80

100

Rel

ativ

e Ab

unda

nce

1738.52

3859.035166.68 5978.12

27292.743048.61 13285.71 21712.699196.66 17210.4815537.766471.31

1707.0 1707.5 1708.0 1708.5m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e Ab

unda

nce

1707.7987R=69904

z=16 1707.9871R=67900

z=161707.6743R=69700

z=16 1708.0480R=69200

z=161708.1117R=68600

z=161707.5486R=67700

z=161708.2360R=70600

z=161707.4833R=64800

z=161708.3612R=70100

z=161707.3010R=74502

z=161706.8594R=63606

z=161708.7953R=66802

z=16

MS3 (HCD)

Quadrupole selection

MS3 (deconvoluted)

a6

a6

8

5000 6000 7000 8000 9000 10000 11000m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bund

ance

Rel

ativ

e In

tens

ity

7925.46

8453.76

7459.36

9057.547044.85

17+

16+

15+

18+ 14+

0

10

20

30

40

50

60

70

80

90

100126790.516

126864.992

126500 127000 127500Mass

+2.9 ppm

1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000m/z

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

4051.0059z=7

3079.0283z=5

4982.4927z=2 7926.6577

z=162432.2810z=3

1299.6334z=2

7459.8511z=17

5759.0264z=3

8454.66895z=157043.8652

z=18

5000 10000 15000 20000 25000m/z

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

15381.0053

2614.2637 17264.0994

9956.95611 22444.8745

7474.870728331.0084

6650.5527

14907.96247 25621.7640011056.70154 20675.85380

MS2 (HCD)

MS2 (deconvolued)

12000

MS1

MS1 (deconvolued)

Figure 6. Native MS and native top-down analysis of the AmtB membrane protein complex. Low-ppm mass accuracy allowed confident identification of AmtB protein. Seventy-five b and y ions were identified, representing 19% residue cleavages.

ReSpect™ Deconvolution

Scores

PCS: 711.17

P-Score: 9.30E-63

% Fragments Explained: 24%

% Residue Cleavages: 19%

Matching Fragments (Count 77)

NameIon

TypeIon

NumberTheoretical

MassObserved

Mass

Mass Difference

(Da)

Mass Difference

(ppm)

B230 B 230 24,349.81 24,349.80 -0.009 -0.38

B231 B 231 24,496.88 24,496.83 -0.043 -1.74

B239 B 239 25,252.29 25,252.36 0.062 2.45

B244 B 244 25,677.56 25,677.52 -0.040 -1.54

B270 B 270 28,330.98 28,331.01 0.031 1.10

B271 B 271 28,388.00 28,387.97 -0.026 -0.91

9

Another example of efficient removal of detergent micelles and subsequent top-down analysis of a membrane protein is presented in Figure 7. Native MS and native top-down analysis of the multidrug transporter membrane protein LmrP from L. lactis using the Q Exactive UHMR mass spectrometer allowed identification of a total of 104 b and y ions, representing 33% residue cleavages.

ConclusionThe results of the native MS and native top-down experiments presented here demonstrate the unprecedented performance of the Q Exactive UHMR mass spectrometer when applied to the structural characterization of homomeric and heteromeric protein assemblies, including membrane protein complexes. With a unique combination of high mass resolution and accuracy, high sensitivity, and MS2 and pseudo-MS3 capabilities, the Q Exactive UHMR mass spectrometer pushes beyond the limits of today’s native MS experiments.

Figure 7. Native MS and native top-down analysis of the multidrug transporter membrane protein LmrP from L. lactis enabled identification of a total of 104 b and y ions, representing 33% residue cleavages

©2018 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific and its subsidiaries unless otherwise specified. ReSpect is a trademark of Positive Probability Ltd. This information is presented as an example of the capabilities of Thermo Fisher Scientific products. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representatives for details. TN65379-EN 1218S

Find out more at thermofisher.com

References1. Belov, M.E.; Damoc, E.; Denisov, E.; Compton, P.D.; Horning, S.; Makarov, A.A.;

Kelleher, N.L. Anal. Chem. 2013 Dec 3, 85(23), 11163–73.

2. Ben-Nissan, G.; Belov, M.E.; Morgenstern, D.; Levin, Y.; Dym, O.; Arkind, G.; Lipson, C.; Makarov, A.A.; Sharon, M. Anal. Chem. 2017 Apr 18, 89(8), 4708–4715.

3. van de Waterbeemd, M.; Fort, K.L.; Boll, D.; Reinhardt-Szyba, M.; Routh, A.; Makarov, A.; Heck, A.J. Nat. Methods. 2017 Mar, 14(3), 283–286.

4. Fort, K.L.; van de Waterbeemd, M.; Boll, D.; Reinhardt-Szyba, M.; Belov, M.E.; Sasaki, E.; Zschoche, R.; Hilvert, D.; Makarov, A.A.; Heck, A.J.R. Analyst 2017 Dec 18, 143(1), 100–105.

5. Viner, R.; Horn, D.M.; Damoc, E.; Konijnenberg, A. Integrative Structural Proteomics Analysis of the 20s proteasome complex. IMSC 2018, Florence, Italy, WPS-S01.

AcknowledgmentThe authors thank Dr. A. Laganowsky from Texas A&M University for supplying the AmtB sample and Dr. C. Govaerts from the Université Libre de Bruxelles for providing the LmrP protein.

For Research Use Only. Not for use in diagnostic procedures.