Digestion of Proteins: A Rapid On

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Plasmonic-thermal decomposition/digestion of proteins: a rapid onsurface protein digestion technique for mass spectrometry imaging Rong Zhou, and Franco Basile Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b00430 • Publication Date (Web): 20 Jul 2017 Downloaded from http://pubs.acs.org on July 21, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6

Analytical Chemistry

Plasmonic-thermal decomposition/digestion of proteins: a rapid on-surface protein digestion technique for mass spectrometry imaging Rong Zhou, and Franco Basile* Department of Chemistry, University of Wyoming, 1000 University Ave., Laramie, WY 82071, USA

ABSTRACT

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

A method based on plasmon surface resonance absorption and heating was developed to perform a rapid on-surface protein thermal decomposition and digestion suitable for imaging MS and/or profiling. This photo-thermal process or plasmonic thermal decomposition/digestion (plasmonic-TDD) method incorporates a continuous wave (CW) laser excitation and gold nanoparticles (Au-NPs) to induce known thermal decomposition reactions that cleave peptides and proteins specifically at the C-terminus of aspartic acid and at the N-terminus of cysteine. These thermal decomposition reactions are induced by heating a solid protein sample to temperatures between 200 °C to 270 °C for a short period of time (10 s to 50 s per 200 m segment) and are reagentless and solventless, and thus are devoid of sample product delocalization. In the plasmonic-TDD setup the sample is coated with Au-NPs and irradiated with 532 nm laser radiation to induce thermoplasmonic heating and bring about site specific thermal decomposition on solid peptide/protein samples. In this manner the Au-NPs act as nano-heaters that result in a highly localized thermal decomposition and digestion of the protein sample that is independent of the absorption properties of the protein, making the method universally applicable to all types of proteinaceous samples (e.g., tissues or protein arrays). Several experimental variables were optimized to maximize product yield and they include heating time, laser intensity, size of Au-NPs and surface coverage of Au-NPs. Using optimized parameters, proof-of-principle experiments confirmed the ability of the plasmonic-TDD method to induce both C-cleavage and D-cleavage on several peptide standards and the protein lysozyme by detecting their thermal decomposition products with MALDI-MS. The high spatial specificity of the plasmonic-TDD method was demonstrated by using a mask to digest designated sections of the sample surface with the heating laser and MALDI-MS imaging to map the resulting products. The solventless nature of the plasmonic-TDD method enabled the non-enzymatic on-surface digestion of proteins to proceed with undetectable delocalization of the resulting products from their precursor protein location. The advantages of this novel plasmonic-TDD method include short reaction times (< 30 s/200 m), compatibility with MALDI, universal sample compatibility, high spatial specificity and localization of the digestion products. These advantages point to potential applications of this method for on-tissue protein digestion and MS-imaging/profiling for the identification of proteins, high fidelity MS imaging of high molecular weight (>30 kDa) proteins and the rapid analysis of formalin-fixed paraffin-embedded (FFPE) tissue samples.

35

*corresponding author: [email protected]

36 37 38 39

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 22

1 2

INTRODUCTION

3

Matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) has been widely

4

applied for the analysis of large biological molecules, i.e. peptides, proteins and lipids, since its

5

inception.1–3 The range of applications of this soft ionization method was expanded after Caprioli’s group

6

used MALDI-MS to map the spatial distribution of small molecular weight dye molecules. 4 This work

7

paved the way for the development of MALDI-MS imaging (MSI) for the direct analysis of biomolecules

8

from tissue samples. Because the mapping of biomolecules directly from tissue has obvious benefits in

9

guiding biological research and clinical diagnosis, MALDI-MSI has been embraced by many laboratories

10

around the world. This has led to a several improvements in analysis sensitivity and signal-to-noise (S/N)

11

enhancements brought about mostly by developing optimized protocols for the preparation of tissue

12

samples prior to MALDI-MSI measurements. For example, Setou and coworkers determined that thinner

13

tissue samples enhanced signal intensity for on-tissue MALDI-MS detection,5 while Caprioli’s group

14

found an optimum tissue thickness in the range 10-20 µm.6 This range was determined by reaching a

15

balance between the signal intensity and technical issues with the preparation and manipulation of very

16

thin tissue sections. An optimized protocol was published for direct on-tissue protein imaging that

17

includes washing steps needed to remove molecules leading to signal suppression in MSI.7,8 Different

18

methods were investigated to deposit the MALDI matrix onto the sample surface, including automatic

19

spotting droplet, spraying and sublimation.9,10 These methodologies were optimized to yield reproducible

20

and sensitive signals, while at the same time providing high spatial resolution with minimal amount of

21

analyte delocalization. Today the MALDI-MSI technique can provide on-tissue imaging of small to

22

medium size molecules, for example, metabolites, peptides and proteins.11–15

23

Despite these advances, the application of MALDI-MSI to routinely detect proteins with MW > 30 kDa

24

remains a challenge. This shortcoming has been attributed to factors such as low ionization efficiency

25

(and detection) of high molecular weight molecules,16 and several investigators have made efforts to

26

extend the upper mass range of analyzed biomolecules by MALID-MSI. Instrumental hardware, for

27

example the CovaIX high mass HM1 detector, was designed to detect high mass ions and its

28

implementation allowed the imaging-MS of proteins up to 70 kDa from tissue samples.17 Various

29

matrices have been tested to replace the traditional sinapinic acid matrix for high MW proteins in

30

MALDI-MSI.15 For example, the ferulic acid matrix was found to enhance signal intensity in the m/z

31

range of 20,000 to 150,000 and enabled the imaging of biomolecules with MW’s above 70 kDa. However,

32

these methods either required the use of harsh solvents like hexafluoroisopropanol (HFIP) or were shown

33

to have a low shot-to-shot mass spectral reproducibility.15,18 An approach more accessible to most 2 ACS Paragon Plus Environment

Page 3 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

laboratories and utilizing existing instrumentation is the implementation of an enzymatic protein digestion

2

step (e.g., using trypsin) directly on the tissue sample followed by detection of the digestion products.19–21

3

By mapping the resulting peptide products, it was possible to obtain the spatial distribution of precursor

4

proteins with molecular weights up to 25 kDa.16 On-tissue digestion also enabled the identification of the

5

precursor protein by incorporating a MS/MS analysis of the resulting peptide(s).22 Although on-tissue

6

digestion is a successful and accessible approach in MALDI-MSI, several issues remain a challenge: i)

7

digestion reaction times ranging from 2 hours to overnight, ii) the need for stringent reaction conditions

8

(temperature, pH), and iii) the need to conduct the digestion reaction under solution conditions. With

9

regards to analysis time, the trypsin digestion process for a tissue sample with dimensions 15 x 10 mm

10

was reported to take about 4 hours,20,23 while other investigators have reported overnight incubations,24

11

limiting the use of this technology for high throughput applications. Regarding the need to conduct the

12

digestion reaction in solution (i.e., high hydration conditions), delocalization of the tryptic peptide

13

products during the digestion step leads to a decreased fidelity of the resulting image, that is, a loss of

14

resolution by distortion of the spatial integrity of the sample itself. This is a direct result of the

15

requirement that enzymes must be in solution during the digestion step in order to be effective and is

16

particularly severe when implementing multiple droplet applications onto a single location. For example,

17

using an automatic spotter (for the enzyme digestion steps) the resolution has been reported to be in the

18

order of 250 to 300 µm, mainly due to the spacing between enzyme droplet applications (~250 m);

19

however, image fidelity is probably lower due to sample delocalization within each droplet.20,25 For a

20

trypsin spraying setup spatial resolutions improves to 50-100 µm,23,24 but still requires digestion times

21

ranging from 4 hours to overnight in controlled environments (37 oC in either 100% humidity or CO2

22

atmosphere). Therefore, the ability to rapidly digest high MW proteins on-tissue in a matter that avoids

23

the delocalization of the peptide products would extend the application of MSI to the high resolution

24

imaging and identification of proteins directly from tissue samples, regardless of their MW. To date, all

25

on-tissue protein digestion protocols involve the deposition of enzyme solutions, and the resulting

26

analyses thus suffer from spatial and/or temporal constraints.

27

Our laboratory reported the site-specific cleavages at the N-terminus of cysteine (C) and the C-terminus

28

of aspartic acid (D) induced by heating solid peptides and proteins sample to temperatures between 200-

29

250 °C for a short period of time (10-30 s).26,27 The resulting peptide products from this thermal

30

decomposition/digestion (TDD) process have been detected by ESI, DESI and MALDI-MS and

31

confirmed by MS/MS analyses. The digestion step is performed without reagents (i.e., solventless) and is

32

extremely rapid, performed in seconds, when compared to any digestion step performed with enzymes (4

33

h to overnight). These attributes make the TDD method a prime candidate to perform on-tissue digestion



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 22

1

for MSI applications, as it would address issues related with long digestion times and product

2

delocalization.

3

Initial development of the TDD process was performed in a tube-pyrolyzer setup, and in this format it

4

is not amenable for MSI applications. Accordingly, we recently reported the implementation of the TDD

5

process using a convective heating technique, where hot air was used to heat protein-covered surfaces to

6

250 °C, thus successfully inducing site-specific cleavages at C and D in model peptides and proteins.28 In

7

the present work we demonstrate an on-surface TDD method compatible with MSI (imaging or profiling

8

mode) based on using gold nanoparticles (Au-NP’s) and 532 nm laser radiation to produce nano-heaters

9

via a plasmonic heating process (plasmon resonance absorption and heating). Using these plasmonic

10

nano-heaters, the rapid (seconds) site-specific cleavages at the amino acids C and D is demonstrated on

11

standard peptides and the protein lysozyme, which were evenly coated on surfaces. MALDI-MS images

12

of surfaces treated with the plasmonic-TDD process are compared directly with those treated with on-

13

surface trypsin digestions, and results highlight the attributes of this novel non-enzymatic digestion

14

method for MSI applications.

15

EXPERIMENTAL

16

Chemicals and Materials. Peptides anti-oxidant peptide A (sequence PHCKRM) and angiotensin II

17

(sequence DRVYIHPF) were purchased from AnaSpec (San Jose, CA). The protein lysozyme (chicken

18

egg white, MW 14.3 kDa), trypsin (porcine pancreas), matrices α-cyano-4-hydroxycinnamic acid (CHCA)

19

and sinapinic acid (SA) were from Sigma Aldrich (St. Louis, MO). Water, methanol and acetonitrile

20

solvents used in all sample preparation for MS measurement were HPLC grade (Burdick and Jackson,

21

Muskegon, MI). Spherical gold nanoparticles with diameter 31, 50, 70, 92 nm (all with citrating cap)

22

were used in experiments as described (Nanopartz, Loveland, CO). All chemicals were used without

23

further purification. Indium-tin-oxide (ITO) coated glass slides with 70-100 Ohms resistivity were used as

24

the substrate for all samples (SPI Supplies, West Chester, PA).

25

Sample Preparation. All solutions, including peptides, protein lysozyme, matrix and Au-NPs, were

26

evenly coated onto ITO glass slides using by spraying with an in-house nebulizer spray setup.29 ITO

27

plates were washed with HPLC grade methanol and air dried prior to use. All sprayed sample solutions

28

were made in acetonitrile/water 50/50 (v/v). To achieve a homogeneous film, the concentration of

29

peptides solutions was set at ~1 mg/mL, while the protein lysozyme was set at 2.5 mg/mL, for a final

30

peptide/protein surface density of 15 g/cm2. The final solvent of the Au-NPs suspension was set to

31

acetonitrile/water 50/50 (v/v) by centrifuging the original Au-NPs suspension, removing half of the

32

original solvent and adding the same volume of acetonitrile. The final concentration of Au-NPs used for 4 ACS Paragon Plus Environment

Page 5 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

spraying was estimated spectrophotometrically at 4.93×1010 NPs/mL. MALDI matrices CHCA and SA

2

were used for reflectron mode and linear mode, respectively, and were prepared at a concentration of 10

3

mg/mL in 50/50 acetonitrile/water solution with 0.1 % trifluoroacetic acid (TFA) final concentration.

4

Plasmonic-TDD Setup. A simplified sequence of the steps involved for conducting on-surface

5

plasmonic-TDD is shown in Figure 1. The Au-NPs solutions (acetonitrile/water, 50/50) were measured

6

with a UV-Vis spectrophotometer (Varian Cary 50, Agilent Technologies, CA) to obtain their absorption

7

spectra in the wavelength range of 300 – 700 nm using a quartz cuvette with a 1 cm path length. In the

8

plasmonic-TDD setup, the sprayed sample layer with Au-NP’s was irradiated (and heated) by a focused

9

continuous wave (CW) green laser (wavelength 532 nm, 150 mW, Wickedlasers, Hong Kong, China) at a

10

fixed distance away from the focusing lens (focal distance of 3.8 cm) as described in Figure S-1 in the

11

supplemental material. The surface coverage of the Au-NPs was adjusted by changing the volume of Au-

12

NPs solution sprayed on the sample surface, and the theoretical surface coverage was calculated by its

13

concentration and sprayed area. During the heating process, the sample holder was attached and moved

14

by a syringe pump (PHD 4400, Harvard Apparatus, Holliston, MA) at a constant rate. The heating time

15

was adjusted by changing the rate at which the sample holder was moved and heating times varied from

16

10 s per 200 m segment (for imaging experiments) to 50 s per 200 m segment for digestion of peptide

17

and protein standards. The size of the laser beam after focusing was related to the distance away from the

18

focusing lens so that the light intensity in the heating process was adjusted by placing the sample at

19

different distances to achieve different temperatures. The photon power density (mW/cm2) was calculated

20

by dividing laser power by the beam area at the selected distance. Products generated were covered by

21

with the sprayed MALDI matrix layer and detected by MALDI-ToF-MS (Voyager DE-STR, Sciex,

22

Foster City, CA).

23

Mass spectrometry and imaging. All MS measurements were performed either with a MALDI-ToF-

24

MS (Voyager-STRTM, AB-Sciex, Foster City, CA) or a MALDI-ToF/ToF-MS instrument (5800TM, Sciex,

25

Foster City, CA). Using the MALDI-ToF-MS instrument, data were collected using MMSIT (Novartis,

26

Switzerland) and analyzed with BioMap (Novartis, Switzerland). Using the MALDI-ToF/ToF-MS

27

instrument, data were collected with the ToF/ToF Imaging Acquisition SoftwareTM (Sciex, Foster City,

28

CA) and analyzed with TissueViewTM (Sciex, Foster City, CA).

29

Plasmonic-TDD experiment and MALDI-MS imaging. A glass mask (9 mm × 9 mm) with an

30

opaque institution logo (Bucking horse and rider; Figure 3) was placed in front of the prepared protein

31

glass slide covered with a layer of Au-NPs. The heating laser was then scanned in parallel paths (500 µm

32

spacing; e.g., see Figure 3) to achieve an average heating time of 10 s per 200 m segment (a MALDI-

33

MS imaging “pixel”). As such, the laser could go through the transparent part of the mask to perform the 5 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 22

1

TDD on the solid sample, but was completely blocked by the opaque part of the mask (i.e., no TDD

2

reaction). The plasmonic-TDD treated sample was imaged by MALDI-ToF-MS (Voyager DE-STR, Sciex,

3

Foster City, CA) with a spatial resolution of 200 µm between laser spots and 200 µm between rows.

4

Digestion products delocalization experiments. Experiments were conducted to compare the extent

5

of sample delocalization between the plasmonic-TDD process and on-surface trypsin digestion, the latter

6

performed with parameters commonly used for on-tissue digestion for MSI.20,30 Both plasmonic TDD and

7

trypsin digestions were performed on protein samples evenly deposited on the surface of an ITO plate.

8

Only half of the surface of the ITO plate was coated with the protein lysozyme, with a sharp edge

9

delineating the coated and uncoated areas. The digestion using either method was performed at this sharp

10

edge of protein sample on the plate. In the plasmonic-TDD experiment, Au-NP’s were evenly sprayed

11

onto both areas (Figure S-5 a), followed by a scan of the heating laser in parallel paths over the entire

12

surface. For the on-surface trypsin digestion experiments, the ITO cover slides with half surface covered

13

by lysozyme were placed in a humidity chamber (100% relative humidity, 37 °C) and 30 iterations of

14

buffered trypsin solution (83 ng/µL, 0.2 µL per iteration, 8 min time intervals between iterations) were

15

delivered manually with a pipet to the same spot at the edge of the two areas, as shown in Figure S-5b.20,30

16

To measure the extent of product delocalization for both digestion methods, the entire surface was imaged

17

after digestion by MALDI-ToF/ToF-MS. All images collected for this experiment were set to a spatial

18

resolution of 100 µm and 200 laser shots for each spot.

19

RESULTS AND DISCUSSION

20

In developing a TDD methodology compatible with MALDI-MSI several attributes were considered.

21

First, the new methodology would preserve the reagentless and solventless characteristics of the TDD

22

process in order to avoid sample delocalization (convection) during the digestion process. Second, the

23

new methodology would heat the sample via a non-contact mechanism in order to avoid sample

24

contamination. Lastly, the approach needs to be universal, that is, applicable to all forms of tissue and/or

25

biological samples. To this end, the development of a photo-thermal decomposition method for the on-

26

surface heating and digestion fulfills all of the above requirements. For a photo-thermal methodology to

27

be universal it would require that any sample absorb light and achieve a temperature high enough for

28

thermal decomposition (200-250 °C), an essential condition for site specific cleavages at C and D.26–28

29

However, not all biomolecules are capable to absorb light of the same wavelength with similar efficiency,

30

that is, they do not have the same molar absorptivity. In order to make this new methodology suitable for

31

universal use with any peptide and/or protein sample, Au-NP’s are used as surrogate light absorbing and

32

heating elements. Most spherical Au-NP’s absorb light in the visible region of the electromagnetic


Page 7 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

spectrum and can simultaneously heat the surrounding sample via surface plasmon resonance absorption

2

and heat generation/dissipation.31,32 This approach is widely used in photothermal cancer therapy, where

3

targeted cancer cells (via Au-NP’s conjugated with antibodies) undergo laser photothermolysis,33 albeit at

4

a lower temperature than those achieved in this study. Gold is inert with most biological samples and it is

5

known not to interfere with the MALDI process, and in fact Au-NP’s have been used as MALDI matrix

6

in several applications.34,35 For this work, the absorption maxima of Au-NP’s of different diameters (31,

7

50, 70, and 92 nm) were measured (Figure S-2c) and particles with a diameter of 50 nm were selected to

8

perform the photo/plasmonic TDD because of its large absorption at 532 nm. The laser wavelength at 532

9

nm was chosen as low cost laser pointers with sufficient power are now commercially available, but in

10

practice any Au-NP size-shape and light wavelength combination that yields a plasmon resonance

11

absorption can potentially be used for this application. Several factors were optimized to achieve a high

12

yield peptide bond cleavage and include heating time, light intensity and Au-NPs surface coverage, and a

13

detailed discussion of this optimization process is presented in the Supplemental Information section.

14

Using optimized conditions (photon power density 52 W/cm2; heating time of 50 s per 200 m segment;

15

Au-NPs coverage 4.93×105 Particles/mm2), several peptides and the protein lysozyme were used to test

16

the performance of the plasmonic-TDD in inducing site specific cleavages at C and/or D. The peptide

17

standard angiotensin II (sequence DRVYIHPF), evenly coated (sprayed) onto a glass plate, was subjected

18

to plasmonic-TDD and analyzed by MALDI-ToF-MS. Figure 2a shows the MALDI-mass spectrum of

19

angiotensin II before and after plasmonic-TDD. After plasmonic-TDD the protonated intact peptide ion

20

[M+H]+ was observed at m/z 1046.54 as well as its sodium adduct ion [M+Na]+ at m/z 1068.54. Signals

21

corresponding to the thermally induced water loss, [M-H2O+H]+, and ammonia loss, [M-NH3+H]+, from

22

the intact peptide were detected at m/z 1028.51 and m/z 1029.51, respectively. The peptide product from

23

the cleavage at the C-terminus of D was observed at m/z 931.52 (with sequence RVYIHPF) and it is

24

attributed to hydrolysis of the C-terminus peptide bond. These products are consistent with previous

25

results obtained using the tube-furnace pyrolyzer setup to induce TDD.26 Similarly, the peptide anti-

26

oxidant peptide A (sequence PHCKRM) was also tested to demonstrate the ability of the plasmonic-TDD

27

process to cleave peptide bonds at the N-terminus of the amino acid C (note: this sample also showed a

28

signal for the peptide containing an oxidized methionine or PHCKRMox before being treated). The

29

MALDI-mass spectrum after plasmonic-TDD (Figure 2b) shows the protonated intact peptide detected at

30

m/z 771.40 as well as the product resulting from cleavage at the N-terminus of C, with a mass change of -

31

33 Da at the C residue, observed at m/z 504.28 and corresponds to the protonated ion [-33CKRM+H]+.

32

This thermal decomposition product was identical to the product detected using a convective-TDD setup

33

(using hot air) and the same peptide.28 Previous measurements using either the tube furnace or convective



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 22

1

heating (hot air) established that a temperature in the range of 200-250 °C was required to induce site-

2

specific thermal decomposition cleavages at D and C. The cleavage product of the peptide containing the

3

oxidized methionine (Mox) was also observed at m/z 520.3. As such, it can be inferred that with the

4

photon power density (~50 W/cm2) and the Au-NP surface density used here, temperatures above 200 °C

5

in the environment surrounding the Au-NP’s were achieved. However, at photon power density above

6

~320 W/cm2 none of the expected cleavage products were detected as pyrolysis is presumed to be taking

7

place (see Figures S-2). Using mass spectra in Figures 2a and 2b, an approximate estimate of the

8

efficiency of the TDD process in cleaving at D and C can be made. For the cleavage at D, and assuming

9

equal ionization efficiencies for the precursor and product peptides, the efficiency of the process is about

10

30%.

11

measurements in our laboratory performed with phosphorylated peptides,36 where the digestion leading to

12

site-specific cleavage processes was measured to be about 50-75% efficient, with non-specific cleavages

13

and dehydration/deamination reactions being the limiting factors.

For the cleavage at C, the TDD process is about 50% efficient. This agrees with earlier

14

The plasmonic-TDD for the protein lysozyme (sequence in Figure S-3; MW 14,305 g/mol) yielded

15

products that were attributed to both cleavages at C and D. For example, the signal at m/z 605.4

16

corresponds to the protonated molecule [KVFGR-1+H]+ resulting from cleavage at N-terminus of C (at C-

17

6 of the lysozyme sequence) and showed the expected mass change of -1 Da resulting from the amidation

18

of the R residue (Figure 2c). Also, the signal at m/z 828.5 corresponds to the protonated molecule

19

[VQAWIRG-1+H]+ from thermal cleavages at D-119 and C-127 in the lysozyme sequence. In addition,

20

the peptides detected at m/z 1255.7 and m/z 1299.8 corresponded to the protonated molecules [-

21

33

22

33

23

observed when lysozyme was heated using a convective heating TDD setup and their sequences have

24

been verified by ESI-MS/MS measurements.28 Similar to previous TDD reactions, fragments due to

25

missed cleavages were also found in the plasmonic-TDD process, such as the peptide

26

33

27

process has been successfully tested with other proteins including bovine serum albumin (BSA)28 and

28

RNase A.37 (new ref added: Zhang et al. / J. Anal. Appl. Pyrolysis 80 (2007) 353–359)

CNDGRTPGSRNL-1+H]+ (cleavage at C-64 and C-76 in the lysozyme sequence) and [CKGTDVQAWIRG-1+H]+ (cleavage at C-115 and C-127), respectively. These cleavage products were

-

CKGTDVQAWIRG-1 detected at m/z 1299.8, as well as water and/or ammonia loss products. The TDD

29

In terms of protein digestion efficiency, the precursor protein was undetected after the TDD process, as

30

protein levels were below the detection limit of the MALDI-MS measurement (vide infra Figures 4a and

31

4b). Assuming about a 1 mm2 area of protein digested (with a surface coverage of 15 g/cm2) a total of

32

150 ng of protein were completely converted into peptides in 20 min by the plasmonic-TDD process

33

(using a heating time of 50 s heating/200 m segment), albeit not all peptides produced derive from site8 ACS Paragon Plus Environment

Page 9 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

specific cleavages. A better indicator of the performance of the plasmonic-TDD process as a protein

2

digestion technique would be the sequence coverage obtained. For the protein lysozyme and using

3

plasmonic-TDD (Figure 2) an overall sequence coverage of 18.6% is obtained. This value can be

4

considered an underestimation of the true sequence coverage as the detected mass range was narrower

5

than the expected products and no chromatographic step was implemented (i.e., ionization suppression

6

effects). However, using other measurements employing convective heating (hot air) TDD a sequence

7

coverage of 58.9% for the same protein was obtained.28 Further LC-ESI-MS/MS are planned as they are

8

expected to yield a more accurate estimate of the TDD sequence coverage for a series of proteins of

9

different sizes.

10

Plasmonic-TDD for on-surface digestion and MALDI-MS imaging. The compatibility of the

11

plasmonic-TDD technique with MALDI-MS imaging is demonstrated next, and in particular, its ability to

12

perform the digestion process within a predefined area and without product delocalization. It is worth

13

noting that the plasmonic-TDD parameters were optimized in this study to increase product yield, and

14

these values may not be optimal to achieve the highest spatial resolution for digestion, and the possibility

15

of a trade-off between digestion spatial resolution and product yield has not yet been explored. However,

16

it is reasonable to conclude that the main factor in providing a high spatial resolution digestion with the

17

plasmonic-TDD method would be the diameter of the heating laser beam, which is a current hardware

18

limitation in the present study. With the current experimental setup, using a low-cost 532 nm laser pointer

19

with a beam diameter of approximately 600 m (and an average heating time of 10 s per ~200 m

20

segment), a digestion product distribution across the heating laser was found to have a full width at half

21

maximum (FWHM) of 185 ± 50 µm (see Figure S-4). Better control of the heating laser beam shape

22

could, in principle, increase the spatial resolution of the digestion for profile-based on-tissue MALDI-MS

23

applications.

24

To demonstrate the applicability of the plasmonic-TDD process to MALDI-MS imaging, a square mask

25

(9 mm × 9 mm) with the appointed opaque area (i.e., the cowboy logo) was placed in front of an ITO

26

plate coated with a uniform layer of the protein lysozyme followed by a layer of Au-NPs. The heating

27

laser was scanned across this surface in a parallel paths 500 µm apart (see Figure 3) to achieve an average

28

heating time of 10 s per 200 m segment (to decrease digestion time), and only the surface under the

29

transparent area of the mask was exposed to the laser light to induce the TDD reaction (i.e., plasmon

30

resonance heating). The entire area (exposed and unexposed) was later imaged by MALDI-MS in the

31

reflectron mode with a spatial resolution of 200 µm. The resulting images indicating the spatial

32

distribution of several of the digestion products are shown in Figures 3a-d. As expected in the area

33

exposed to the heating laser, homogenous signals of the TDD fragments were observed at m/z 605.4, m/z 9 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 22

1

828.5, m/z 1255.7 and m/z 1299.8 (see Figure 2c for mass spectrum), which were generated from site-

2

specific cleavages at the C and D amino acids. Conversely, no TDD products were detected in the area

3

not exposed to the laser light (i.e., under the opaque part of the mask). These results indicate that the

4

plasmonic-TDD process is able to induce highly reproducible protein digestions (i.e., constant digestion

5

efficiency) throughout the exposed area of the sample as indicated by the uniform signal intensity

6

distribution of the products across the imaged area. In addition, these results also point to the spatial

7

specificity of the process as only the exposed areas of the samples are digested, a trait that is expected to

8

prove useful when performing profile-based on-tissue MALDI-MS measurements. Peptides due to non-

9

specific cleavages and/or of unassigned sequence were also present in this sample and are spatially

10

correlated to the plasmonic-TDD process (see Figure S-6).

11

In the on-tissue trypsin digestion using spotting devices, delocalization (i.e., diffusion and convection)

12

of the generated peptides from their original position (precursor protein) is often cited as one of the main

13

factors limiting the spatial resolution of direct on-tissue analysis of protein via enzymatic digestion in

14

MSI.30,38 On the other hand, the solventless nature of the plasmonic-TDD process is expected to

15

considerably alleviate this product delocalization, and thus increase the fidelity of the MS imaging

16

process after on-surface digestion. To test this hypothesis, experiments were conducted that directly

17

compare the extent of sample delocalization between the on-tissue trypsin digestion and plasmonic-TDD,

18

and a detailed diagram describing this experimental setup is shown in Figure S-5. In Figure 4a and Figure

19

4i MALDI-MS images of the lysozyme sample before digestion by either method show an evenly

20

distributed signal and a clear interface between the sprayed protein sample and blank areas. Using the

21

plasmonic-TDD method across this interface (shown with a yellow doted box in Figure 4b), the MALDI-

22

MS image of the intact lysozyme protein shows that the protein is completely consumed (or undetected).

23

Meanwhile, using the same TDD digestion condition, the MALDI-MS images of the expected TDD

24

products (Figures 4c-e) yielded images showing the same well-defined boundary between the sample and

25

the blank area. No TDD products are detected beyond the original boundary, clearly demonstrating that,

26

at the MALDI-MS resolution level used (200 m), no product delocalization takes place during the

27

plasmonic-TDD process. As a comparison, solution-based on-surface trypsin digestion was performed by

28

manually pipetting the trypsin solution directly on the boundary of the sample/blank interface as shown in

29

Figure 4g and Figure 4h. After digestion, an area of the lysozyme protein corresponding to the deposited

30

trypsin droplet was consumed (Figure 4j). The trypsin digestion products were also detected by MALDI-

31

MS imaging and they were present in an area corresponding to the original trypsin droplet dimensions.

32

That is, peptide products were detected beyond the sharp boundary delineating the position of the

33

precursor protein, as shown in Figure 4k. Peptide products detected in the blank zone were attributed to


Page 11 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

the delocalization of peptide digestion products within the dimensions of the trypsin droplet. This

2

delocalization is mainly attributed to capillary flows within the droplet that are driven by differential

3

evaporation rates across the drop, a phenomenon commonly known as the coffee ring effect.39 It is

4

evident that the fidelity of the resulting MALDI-MS image is preserved when using the plasmonic-TDD

5

process, while solution-based on-surface digestion tends to distort or lower the fidelity of the resulting

6

image. However, it is worth noting that our imaging resolution (i.e., MALDI-MS laser characteristics and

7

settings) in this study was 200 m, thus the degree of delocalization of the plasmonic-TDD process can

8

only be determined to be at least 200 m.

9 10

CONCLUSION

11

A thermo-plasmonic heating process was developed to perform on-surface protein thermal

12

degradation/digestion or TDD of standard protein samples coated on surfaces. Although not yet tested

13

with actual biological tissue samples, proof-of-principle MALDI-MS imaging experiments demonstrated

14

the spatial selectivity and specificity of the plasmonic-TDD process. Results showed the potential of the

15

TDD process to perform on-surface digestions for either profile-mode or image-mode MALDI-MS tissue

16

measurements. Conclusive evidence was obtained in the form of MALDI-MS images indicating that the

17

plasmonic-TDD process is uniform across a surface. Most importantly, in the plasmonic-TDD process

18

little to no product delocalization is observed when compared to the traditional (i.e., spotting) enzymatic

19

on-surface digestion, thus preserving the fidelity of the resulting product spatial distribution after on-

20

surface digestion. Even though some experimental plasmonic-TDD parameters were optimized, the

21

current study is limited in terms of spatial resolution and digestion time by the hardware used, mainly in

22

the quality of the heating laser used. For example, by increasing the heating laser power it is conceivable

23

that the TDD reactions can be completed in 0.1 s or less (instead of 10 s per 200 m segment as in this

24

work), thus decreasing the overall digestion time. That is, a 1.0 cm2 sample area could in principle be

25

digested in just under 2 min, instead of 3 h as in the current study. It is also conceivable to irradiate the

26

entire surface area with a high intensity, high surface area light source to achieve plasmonic-TDD on the

27

entire sample in a single step (e.g., a 1 cm2 surface area could be digested in 10-30 s). Spatial distribution

28

of the TDD digestion products (i.e., resolution of the digestion process) can be improved by optimizing

29

the beam quality of the heating laser and other parameters like the heating time, laser intensity and Au-NP

30

surface coverage. Overall, the advantages of the plasmonic-TDD method include short reaction times (10-

31

30 s/200 m), compatibility with MS detection, site-specific cleavages at D and C in peptides and

32

proteins, high spatial specificity and no product delocalization (solventless digestion). This last advantage 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 22

1

has direct implications in increasing the fidelity of the resulting image of samples treated with on-surface

2

(or on-tissue) digestion and analyzed by MALDI-MS. On the other hand, the TDD process has side

3

reactions that produce dehydration and/or deamination products. While these products may increase the

4

complexity of the peptide mixture generated from tissues, we have shown that they retain sequence

5

information, thus allowing their MS/MS analysis.

6

complexity may be countered by the reagentless nature of the TDD process, as no exogenous proteins

7

(e.g.., trypsin) are added to the tissue, which may cause signal suppression effects. Thus, in TDD the

8

sample complexity is not augmented by the addition of an enzyme. Furthermore, the number of peptides

9

formed by either trypsin or TDD is comparable since trypsin cleaves proteins at two locations (R and K),

10

which are very abundant in proteins. On the other hand, in TDD the number of peptide products due to

11

cleavages as well as dehydration/deamination is defined by two factors: 1) in TDD peptides are formed by

12

cleavage at the amino acids D and C, which are less abundant than R and K, and 2) although not

13

quantified, we have observed that the TDD method tends to miss cleavages more often than trypsin

14

digestion, thus not all of the expected peptides are formed. This last factor leads to the formation of

15

larger peptides that can be detected with an enhanced S/N at a higher, less noisy m/z range.

16

disadvantage of this methodology, however, is the susceptibility of post-translational modifications like

17

phosphorylation to heating.36 Work is currently in progress in our laboratory to expand the capabilities of

18

on-surface TDD MS-imaging/profiling of large MW proteins directly from tissue samples, the analysis of

19

FFPE tissue samples and to develop new radiative heating methods for the rapid digestion of the entire

20

tissue section in a single step.

21

Acknowledgement

22

This work was supported in part by a generous grant from the NSF Chemical Measurement and Imaging

23

directorate (NSF-CMI 1611538). The acquisition of the MALDI-ToF/ToF-MS instrument (Sciex 5800TM)

24

was made possible by a generous grant from the NSF Major Research Instrumentation program (NSF-

25

MRI 1429615).

26

Supporting Information. The Supporting Information section contains experimental details pertaining to:

27

1) laser/sample configuration schematics, 2) optimization of parameters for Plasmonic-TDD for increased

28

product yield, 3) spatial resolution of the Plasmonic-TDD process, 4) experimental design for the

29

comparison of on-surface trypsin digestion versus Plasmonic-TDD, and 5) spatial correlation of TDD

30

products not identified.

In addition, this perceived increased sample

A clear

31


Page 13 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47


REFERENCES (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

(18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30)

Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60 (20), 2299–2301. Beavis, R. C.; Chait, B. T.; Standing, K. G. Rapid Commun. Mass Spectrom. 1989, 3 (7), 233–237. Fuchs, B.; Schiller, J. Eur. J. Lipid Sci. Technol. 2009, 111 (1), 83–98. Caprioli, R. M.; Farmer, T. B.; Gile, J. Anal. Chem. 1997, 69 (23), 4751–4760. Sugiura, Y.; Shimma, S.; Setou, M. J. Mass Spectrom. Soc. Jpn. 2006, 54 (2), 45–48. Schwartz, S. A.; Reyzer, M. L.; Caprioli, R. M. J. Mass Spectrom. 2003, 38 (7), 699–708. Seeley, E. H.; Oppenheimer, S. R.; Mi, D.; Chaurand, P.; Caprioli, R. M. J. Am. Soc. Mass Spectrom. 2008, 19 (8), 1069–1077. Yang, J.; Caprioli, R. M. Anal. Chem. 2011, 83 (14), 5728–5734. Franck, J.; Arafah, K.; Barnes, A.; Wisztorski, M.; Salzet, M.; Fournier, I. Anal. Chem. 2009, 81 (19), 8193–8202. Bouschen, W.; Schulz, O.; Eikel, D.; Spengler, B. Rapid Commun. Mass Spectrom. 2010, 24 (3), 355–364. Verhaert, P.; Uttenweiler-Joseph, S.; de Vries, M.; Loboda, A.; Ens, W.; Standing, K. G. PROTEOMICS 2001, 1 (1), 118–131. DeKeyser, S. S.; Kutz-Naber, K. K.; Schmidt, J. J.; Barrett-Wilt, G. A.; Li, L. J. Proteome Res. 2007, 6 (5), 1782–1791. Stoeckli, M.; Staab, D.; Schweitzer, A. Int. J. Mass Spectrom. 2007, 260 (2–3), 195–202. Lietz, C. B.; Gemperline, E.; Li, L. Adv. Drug Deliv. Rev. 2013, 65 (8), 1074–1085. Mainini, V.; Bovo, G.; Chinello, C.; Gianazza, E.; Grasso, M.; Cattoretti, G.; Magni, F. Mol. Biosyst. 2013, 9 (6), 1101–1107. Cillero-Pastor, B.; Heeren, R. M. A. J. Proteome Res. 2014, 13 (2), 325–335. Remoortere, A. van; Zeijl, R. J. M. van; Oever, N. van den; Franck, J.; Longuespée, R.; Wisztorski, M.; Salzet, M.; Deelder, A. M.; Fournier, I.; McDonnell, L. A. J. Am. Soc. Mass Spectrom. 2010, 21 (11), 1922–1929. Williams, T. L.; Andrzejewski, D.; Lay, J. O.; Musser, S. M. J. Am. Soc. Mass Spectrom. 2003, 14 (4), 342–351. Cole, L. M.; Djidja, M.-C.; Bluff, J.; Claude, E.; Carolan, V. A.; Paley, M.; Tozer, G. M.; Clench, M. R. Methods 2011, 54 (4), 442–453. Groseclose, M. R.; Andersson, M.; Hardesty, W. M.; Caprioli, R. M. J. Mass Spectrom. 2007, 42 (2), 254–262. Heijs, B.; Carreira, R. J.; Tolner, E. A.; de Ru, A. H.; van den Maagdenberg, A. M. J. M.; van Veelen, P. A.; McDonnell, L. A. Anal. Chem. 2015, 87 (3), 1867–1875. Minerva, L.; Clerens, S.; Baggerman, G.; Arckens, L. PROTEOMICS 2008, 8 (18), 3763–3774. Patel, E.; Clench, M. R.; West, A.; Marshall, P. S.; Marshall, N.; Francese, S. J. Am. Soc. Mass Spectrom. 2015, 26 (6), 862–872. Schober, Y.; Guenther, S.; Spengler, B.; Römpp, A. Rapid Commun. Mass Spectrom. 2012, 26 (9), 1141–1146. Casadonte, R.; Caprioli, R. M. Nat. Protoc. 2011, 6 (11), 1695–1709. Zhang, S.; Basile, F. J. Proteome Res. 2007, 6 (5), 1700–1704. Basile, F.; Zhang, S.; Kandar, S. K.; Lu, L. J. Am. Soc. Mass Spectrom. 2011, 22 (11), 1926–1940. Zhou, R.; Basile, F. J Anal Appl Pyrolysis 2017, in press. Basile, F.; Kassalainen, G. E.; Ratanathanawongs Williams, S. K. Anal. Chem. 2005, 77 (9), 3008– 3012. Schober, Y.; Schramm, T.; Spengler, B.; Römpp, A. Rapid Commun. Mass Spectrom. 2011, 25 (17), 2475–2483. 13 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11

(31) (32) (33) (34) (35) (36) (37) (38) (39)

Page 14 of 22

Govorov, A. O.; Richardson, H. H. Nano Today 2007, 2 (1), 30–38. Baffou, G.; Quidant, R.; Girard, C. Appl. Phys. Lett. 2009, 94 (15), 153109. Qin, Z.; Bischof, J. C. Chem. Soc. Rev. 2012, 41 (3), 1191. McLean, J. A.; Stumpo, K. A.; Russell, D. H. J. Am. Chem. Soc. 2005, 127 (15), 5304–5305. Amendola, V.; Litti, L.; Meneghetti, M. Anal. Chem. 2013, 85 (24), 11747–11754. Lu, L.; Basile, F. J. Anal. Appl. Pyrolysis 2013, 104, 412–417. Zhang, S.; Shin, Y.-S.; Mayer, R.; Basile, F. J. Anal. Appl. Pyrolysis 2007, 80 (2), 353–359. Römpp, A.; Spengler, B. Histochem. Cell Biol. 2013, 139 (6), 759–783. Robert, D. D.; Olgica, B.; Todd, F. D.; Greb, H.; Sidney, R. N.; Thomas, A. W. Nature 1997, 389 (6653), 827–829.

12


Page 15 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1 2

Captions to figures

3 4

Figure 1. Sequence of steps involved in plasmonic-Thermal Decomposition/Digestion (TDD). After plasmonic heating, the sample is coated with MALDI matrix (not shown) and analyzed by MALDI-MS.

5 6 7 8 9 10 11 12

Figure 2. Peptides and protein standards undergoing cleavages at the amino acids C and D via plasmonicTDD and analyzed by MALDI-MS. Samples were evenly coated onto a surface (ITO plate) before being treated by plasmonic-TDD: a) MALDI-mass spectrum of products from angiotensin II; b) MALDI-mass spectrum of products from anti-oxidant peptide A; c) MALDI-mass spectrum of products from the protein lysozyme. Cleavage at the C-terminus of D undergoes without any modification (i.e., peptide bond hydrolysis). Cleavage at the N-terminus of C undergoes with a net -33 Da modification on the C-terminus product and a -1 Da modification at the N-terminus product. Average heating time at each 200 m segment is 50 s. Mox = oxidized methionine (methionine sulfoxide residue)

13 14 15 16 17 18 19 20

Figure 3. Proof-of-concept plasmonic-TDD MALDI-MS imaging. The protein lysozyme was uniformly coated onto a ITO plate: Figure (a) - (d) represent the spatial distributions of the thermal decomposition products of lysozyme with signals at m/z 605, 828, 1225 and 1299, respectively. Experimental parameters: photon power density 31 W/cm2, average heating time at each 200 m segment is 10 s, Au-NP coverage ~4.9×105 NP/mm2 (Log10 (N) = 5.69), laser scanning heating time for the entire imaged area of 9.0 x 9.0 mm2 was about 3 hours. The distance between parallel paths of the heating laser was of 500 µm. MALDIMS measurements in the image were acquired every 200 µm. A total of 50 laser shots were collected and averaged for each mass spectrum.

21 22 23 24 25 26 27 28 29 30 31 32 33

Figure 4. Comparison of on-surface digestion product delocalization between plasmonic-TDD and trypsin digestion: (a) MALDI-MS image of the lysozyme-coated surface before plasmonic-TDD; (b) MALDI-MS image of lysozyme-coated surface after plasmonic-TDD (TDD-treated area delineated by a yellow dash square); (c)-(e) MALDI-MS images of fragments in the plasmonic-TDD digested area (at the same scale); (f) optical images of the lysozyme-coated surface before trypsin digestion (yellow dash line represents the boundary between sample droplet and blank areas); (g) and (h) optical images of the lysozyme-coated surface after trypsin digestion for MALDI-MS analyses in linear mode and reflectron mode, respectively; (i) MALDI-MS image of the lysozyme-coated surface before trypsin digestion; (j) MALDI-MS image of the lysozyme-coated surface after trypsin digestion; (k) MALDI-MS image of trypsin digestion product. Scale bar in all images represents 1 mm. On-surface plasmonic-TDD of lysozyme (experimental parameters: photon power density 31 W/cm2, actual heating time for each 200 m segment was 10 s, Au-NP coverage 4.9×105 Particles/mm2 (Log10 (N) = 5.69)). Fragments and intact protein were mapped by MALDI-MS in reflectron and linear mode, respectively.

34 35 36 37 38 39 40



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 22

1 2

Figure 1.

3 4

AuNP layer Step 1: Spray AuNP onto protein sample

Protein/peptide layer (or tissue)

Step 2: Expose sample with 532nm laser light for 10 s Laser radiation (532 nm) Thermally degraded/ digested proteins

Plasmon resonance absorption and heating

5 6 7 8 9 10 11 12 13 14 15 16 17 18 16 ACS Paragon Plus Environment

Page 17 of 22

1 2

Figure 2a

25000

a

After TDD

15000

10000

[M+H]+ 1046.54

[M+H]+ 1046.54

20000

Intensity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Before TDD

[M-NH3+H]+ 1029.51

[M-CO-NH3+H]+ 1001.51

800

1000

1200

1400

m/z

[M+Na]+ 1068.54

[RVYIHPF+H]+ 931.52 5000

Fragmentation: Plasmonic -TDD 0 800

3

1000

1200

1400

1600

1800

2000

m/z

4 5 6 7 8 9 10 11 12 13 17 ACS Paragon Plus Environment


1 2

Figure 2b

3

50000

771.40 [M+H]+

b 40000

Intensity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 22

[-33CKRM+H]+ 504.28

Before TDD

After TDD

[PHCKRMox+H]+

30000

600

800

1000

1200

m/z

771.36 [M+H]+ 20000

[-33CKRMox+H]+ 520.27

10000

[PHCKRMox+H]+ 787.36

0 500

4

600

700

800

900

1000

m/z

5 6 7 8 9 10 11 12


Page 19 of 22

1 2

Figure 2c

3

60000

50000

c

656.10

a: KVFGR 605.43

568.15

matrix signals

Before TDD 40000

Intensity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


After TDD 30000

600 800 1000 1200 1400 1600 1800 2000 m/z

c: CNDGRTPGSRNL 1255.73

20000

d: CKGTDVQAWIRG 1299.76

b:VQAWIRG 828.54

10000

0 600

4

800

1000

1200

1400

1600

1800

2000

m/z



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 22

1 2

Figure 3.

3 4 transparent area

Mask

a: m/z 605

Signal Intensity a

Heating laser path

b

c

d

High

Low 500 µm

MALDI laser path

Image: 9 mm × 9 mm

opaque area

b: m/z 828

c: m/z 1255

3 mm

d: m/z 1299

200 µm

5 6 7 8 9 10 11 12 13 14 15 16


Page 21 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1 2

Figure 4

3

Plasmonic-TDD m/z 14428

m/z 605.2

m/z 14428

c blank

m/z

protein sample

high

1255.5

d m/z 1299.5

a

1 mm

b

low

e

Trypsin blank

protein sample

trypsin drop

f

g

1 mm

m/z 14428

trypsin drop

Optical images

h m/z 874.3

m/z 14428

MALDI-MS images 1 mm

i

j

k

4 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 22

1 2

TOC graphic high fidelity spatial distribution of P-TDD products a) m/z 605.2

protein

No protein Laser radiation (532 nm)

digested protein

Plasmonic thermal decomposition and digestion (P-TDD)

Au-NP’s

After P-TDD (30 sec) m/z 14,313

imaging MALDI-MS

3

b) m/z 1255.5

m/z 14,313

c) m/z 1299.5

imaging MALDI-MS

Protein cleaved at Asp and Cys

imaging MALDI-MS


Digestion of Proteins: A Rapid On

Recommend Documents