6/28/2013

1

Mass Spectrometry Imaging of Protein and Lipid Distribution in Ex Vivo

Human Skin

Malcolm R Clench

MALDI-MSI workflow diagram. (McDonnell et al 2010).

MALDI-MSI Microprobe ModeWhat do MALDI Images Represent?

• Each image pixel correlates to the corresponding region of the original sample

• Images are produced for intensity of a selected ion or ions.

• Ion intensity is shown as a change in ‘brightness/hue" for each pixel

• Images of different ion distributions can be overlaid for more complex analysis

6/28/2013

2

Sections

• Some Comments on Xenobiotic Imaging by

MALDI-MSI

• Lipid Imaging in Skin

• Protein Imaging in Skin

• Other ProjectsPositive Ion MALDI Mass Spectrum of Imipramine RMM 280

Showing An Intense [M+H]+ ion at m/z 281

Sometimes Xenobiotic Imaging is "Easy"………..

Positive Ion MALDI Mass Spectra Recorded from Control (a) and treated (b)

Straticell-RHE-EPI/001. The Imipramine [M+H]+ ion is clearly visible (*) along with signals arising from endogenous species.

*

m/z 184 PC Head Group

m/z 264 Ceramide Specific Ion

m/z 281 Imipramine [M+H]+

6/28/2013

3

For compounds that do not “fly” by MALDI-MS derivatisation can be used. Here

a carbonyl compound has been derivatised wiith DMNTH (4-dimethylamino-6-(-4-methoxy-1-napthyl)-1,3,4-triazine-2-hydrazine ) (Buttaro et al 2007) to yield

a derivative RMM 424 with good MALDI properties..

………and sometimes the mass spectrometry is difficult

MALDI-MS Images of the Distribution of A Carbonyl Compound as its DMNTH Derivative

in ex-vivo Human Skin . The 30 µm MALDI-MSI data is shown overlaid on H&E stained sections and indicates that the compound does not penetrate into the dermis.

…….and when it does all work quantification may be possible………

Optical image showing an

embedded homogenate

standard array.

MALDI-IMS-MS image

showing the

distribution of

Tiotropium ([M]+) at

m/z 392.

Calibration curve showing the average

intensity of each region plotted against

concentration

Ex-Vivo Human Skin Analysis

6/28/2013

4

Human Skin Analysis

• Intact Protein Analysis

– Sections washed to remove salts/lipids coated with sinapinic acid matrix for

direct MALDI analysis

• On-tissue digestion for identification of proteins

– Sample Preparation

• Overnight on tissue tryptic digestion (sample was sprayed with trypsin) for peptide analysis

• Ethanol washes, 70% and 90% followed by a brief chloroform wash

• α-CHCA matrix application using SunCollect autosprayer (KR Analytical)

– Instrumentation

• Applied Biosystems Voyager-DE STR (modified with Nd:YAG laser)

• Applied Biosytems, MALDI-Q-TOF “Q-Star Pulsar-I “(modified with Nd:YnO4 laser)

• Waters, MALDI-HDMS Synapt G2

Intact protein spectrum generated from MALDI-MS of untreated human skin, using a

Voyager De ProTM (Applied Biosystems) Modified with an Elforlight UV-FQ Nd:YAG laser.

Intact Protein Analysis

From: skop.genetics.wisc.edu/AhnaMassSpecMethodsTheory.ppt

Bottom-up “Shotgun” Proteomics

lyse cells Mixture of 1000’s

of peptidesdigest with

trypsin

2-D LC-

MS/MS RPLC-

MS/MS LC-IMS-

MS/MS

LC

m/z

LC

m/z

drift

Database searching - matching MS/MS data with peptide sequence

SCX

steps

On- Tissue Bottom-up “Shotgun” Proteomics

Mixture of 1000’s of

peptidesdigestion

with trypsin

2-D LC-

MS/MS

MALDI-MS/MS MALDI-IMS-MS/MS

m/zm/z

drift

Database searching - matching MS/MS data with peptide sequence

Tissue Section Sprayed or

Printed with TrypsinMALDI-PMF

6/28/2013

5

4.6 high reso mode MS with LM

m/z700 750 800 850 900 950 1000 1050 1100 1150 1200 1250 1300 1350 1400 1450 1500 1550 1600 1650 1700 1750 1800 1850

%

0

100

SHU_LC_100426_04 36 (0.767) Sb (1,40.00 ); Sm (Mn, 2x3.00); Cm (1:37) TOF MS LD+ 1.62e31048.5586

725.5553

734.5670

850.4199

756.5531

782.5654

786.5955

808.5814

861.06681032.5936

877.0448949.4908

944.5308

934.4691950.4830

966.5105

1027.5192

1274.7274

1161.64431049.5587

1070.5421

1071.5383

1086.5151

1162.6447

1198.7021

1199.7108

1213.5782

1247.5940

1416.65161275.7284

1276.7299

1297.7075

1312.6603

1313.6602

1325.6089

1417.65731529.7327

1438.6281

1439.6333

1570.6787

1842.8623

1841.8564

1578.7518

1819.87391593.7184

1594.7239

1599.73011740.8114

1615.7021 1719.83281741.8167

1843.8588

1857.8306

1864.8409

Typical Current "on-tissue" PMFOn-Tissue MALDI PMF from Human Skin Acquired with ABI QStar Pulsar-I

From skin all peptide MS/MS spectra we acquire without ion-mobility

show the presence of lipid peaks – multiple species

Waters: SYNAPT G2 HDMS

6/28/2013

6

SHU_PJH_Negative_MS01.raw : 3

SHU_PJH_Negative_MS01.raw : 3

Incorporate IMS Separation on Synapt-G2

SHU_PJH_Negative_MS01.raw : 3

SHU_PJH_Negative_MS01.raw : 3

Incorporate IMS Separation on Synapt-G2

SHU_PJH_Negative_MS01.raw:3

SHU_PJH_Negative_MS01.raw : 3

Incorporate IMS Separation on Synapt-G2 Mobilogram shows at least 3 species at this m/z

6/28/2013

7

Negative control skin MSMS 1118.5112

m/z100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 1050 1100

%

0

100

SHU_PJH_100427_05_dt_01 133 (7.161) Cm (103:178) TOF MSMS 0.00LD+ 3.38e3110.0672

1118.5402

272.1873

195.0934175.1220

167.0923

255.1596

225.1108

309.1449 505.2491

378.1700

368.1800

351.1628

396.1789

487.2383406.1600

452.2068

961.5081

533.2418

753.2737

550.2682658.4055591.2358

607.2797

661.3052

745.4527710.3260

858.5458819.3763

800.3512

904.5416

862.4475 915.4832

960.5314

1117.6039

1056.6737

971.62841057.5698

1058.5803

1118.6134

1118.6865

1119.5354

1119.6819

1119.8136

Acquire MS/MS Data with mobility separation no evidence of lipid peaks

Localisation of Peptide Signals

MALDI-MS image of collagen

1 alpha (III) peptide at m/z

1138.56

MALDI-MS image overlay of

keratin 1 peptide 1118.5112

(in orange) and Collagen 1

alpha (III) (in green)

A MALDI-MS image of

keratin 1 peptide 1118.5112

Proteins Identified• Colllagen

• Decorin

• Keratin

• Haemoglobin

• Serum Albumin

• Lumican

• In reality a very small list : suggests complementary

techniques (conventional proteomics) needed for

identification and only use MALDI-MSI for imaging.

6/28/2013

8

Analysis of Treated Human Skin

in Multiple Sample Experiments

Experimental Setup

� MALDI-MS Image,

acquired at a spatial

resolution of 150 µm x 150

µm, from untreated

human skin.

� Data displayed using

Waters HD Imaging

software

MALDI Imaging of Multiple Samples:

Upregulation in Treated Skin

Ai ii

iii iv v vi

Bi ii

iii iv v vi

A MALDI image of a peptide species present at m/z (A) and m/z (B), both of which

are thought to belong to a single protein. The image shows difference in levels of expression between: (i) human skin that was treated with the acetone:olive oil vehicle, (ii) sodium lauryl sulphate, (iii) untreated, (iv) treated with glycerol, (v) DNCB

and (vi) sulfamethoxazole.

Lipid Imaging in Ex Vivo Human

Skin

6/28/2013

9

High mass resolution, positive ion MALDI Mass Spectrum of normal human skin, using

a-CHCA/ANI as a matrix, with an enlarged inset showing the peak resolution achieved (35,000- 40,000 FHWM) (Hart et al., 2011).

MALDI mass spectra taken from regions of treated skin sections

Princlpal Component Analysis Scores and Loading s Plots for a Series of MALDI Mass Spectra Taken from Different Layers of Ex-Vivo

Human Skin

negative_050711.raw : 2

negative_050711.raw : 2

IMS Separation of Lipid Species from Human Skin - Synapt G2

6/28/2013

10

negative_050711.raw : 2

negative_050711.raw : 2

(A) Positive Ion MALDI Product Ion Mass Spectrum of the m/z species 703, identified to

be SM(18:1/16:0) [M+H]+, (B) Positive Ion MALDI Product Ion Mass Spectrum of the lithium adduct of SM(18:1/16:0) ([M+Li]+) m/z 709.5, displaying the corresponding molecular structure (Hart et al., 2011).

Calculated

Monoisotopic

m/z

Experimental

m/z

m/z error in

ppm (1 dp)Possible Lipid Species Ion Species

MS/MS

confirmation of

identity

Isomer with Side chains

identified in MS/MS

544.3379 544.3385 1.1 LPC 18:1 [M+Na]+ √√√√ LPC(18:1(9Z)/0:0)

575.5016 575.5021 -0.8 DG 32:0 or DG 32:1 [M+Na]+ ---- ----

666.4839 666.4839 0.0 CerP(d18:1/18:1(9z)) [M+Na]+ ---- ----

702.5432 702.5418 -2.0 PnE(16:0/18:1(9Z)) [M+H]+ ---- ----

703.5749 703.5751 0.3 SM (d18:1/16:0) [M+H]+ √√√√ SM(d17:1/17:0)

725.5574 725.5579 0.7 SM 34:1 [M+Na]+ ---- ----

734.5695 734.5699 0.5 PC 32:0 [M+H]+ √√√√ PC(16:0/16:0)

756.5520 756.5521 -0.1 PC 32:0, or PE 35:0 [M+Na]+ ---- ----

758.5695 758.5697 -0.2 PC 34:2, or PE 37:2 [M+H]+ ---- ----

760.5851 760. 5854 0.4 PC 34:1 [M+H]+ √√√√ PC(16:0/18:1)

769.4803 769.4803 0.0 PA 42:10 [M+H]+ ---- ----

780.5520 780.5521 -0.1 PC 34:2, or PE 37:2 [M+Na]+ ---- ----

782.5676 782.5681 -0.6 PC 34:1, or PE 37:1 [M+Na]+ ---- ----

786.6008 786.6017 -1.1 PC 36:2, or PE 39:2 [M+H]+ ---- ----

808.5833 808.5835 -0.2 PC 36:2, or PE 39:2 [M+Na]+ ---- ----

810.5989 810.5989 0.0 PC 36:1, or PE 39:1 [M+Na]+ ---- ----

832.5833 832.5826 0.8 PC 38:4, or PE 41:4 [M+Na]+ ---- ----

907.7731 907.7731 0.0 TG 54:3 [M+Na]+ ---- ----

Instrument is also a 40,000 FWHM Resolution Instrument Capable

of 1ppm Accurate Mass Measurement

MALDI-MS Images of the Distribution of Selected Lipid Species in Human Skin (150 µm)

6/28/2013

11

Images at 30 µm Spatial Resolution

MALDI images of a species at m/z 417. The image shows difference in levels of

expression between: (A) human skin treated with hydroquinone, (B) sulfamethoxazole, (C) SLS, (D) the acetone: olive oil vehicle, (E) DNCB, (F) cinnamaldehyde and G)

human skin left, untreated.

Other Projects

Response to Treatment in LSE

Physiogel 24

hours

Oilatum 24 hours Physiogel 6

hoursOilatum 4 hours Physiogel 4

hours

Oilatum 6 hours

6/28/2013

12

Compound formulations studied

Physiogel A.I CreamAcitive ingredient: Paltimitoylethanolamide (PEA) [299.2824 m/z]

Composed of purified water, olea europeaea, glycerol 92.0473 m/z,

pentylene glycol, palm glycerides, olus, hydrogenated lecithin, squalane 410.3913 m/z, betaine 117.0790 m/z, palmitoylethanolamine (0.3%) [active

ingredient], sarcosine 89.0477 m/z, acetamide MEA, hydroxyethylcellulose, sodium carbomer, carbomer and Xanthan Gum.

Oilatum Junior Cream (control)

Oilatum is composed of Active ingredients; light paraffin 6.0% and whit

soft paraffin 15%. Other ingredients include; Macrogol 1000 monostearate, cetostearyl alcohol, glycerol, potassium sorbate, benzyl

alcohol, citric acid, povidone and purified water.

(All m/z values are given in the neutral ion).

759.8682 m/z

Lipid Markers of LSE Layers Observed

Image of LSE sections (across 3 different treatment groups). The samples

were incubated for 24 hours after the treatment, (Control group untreated). The image was acquired at a 25 um x 25 um resolution; normalised to the total ion count (Data produced 15th/4/13).

Ctrl Grp Oilatum Grp Physiogel Grp

SC

SC

SC

758.4265 m/z

Ctrl Grp Oilatum Grp Physiogel Grp

628.03 m/z

SC

SC

SC

Image of LSE sections (across 3 different treatment groups). The samples

were incubated for 24 hours after the treatment, (Control group untreated). The image was acquired at a 25 um x 25 um resolution; normalised to the total ion count (Data produced 15th/4/13).

6/28/2013

13

Fingermark Analysis by MALDI MSI(Dr Simona Francese)

MALDI MS/MS of m/z

195.1

B

MALDI MS

Conclusions

� The use of IMS with MALDI images aids specificity.

�Statistical analysis of the large data sets obtained is essential.

�Using complementary technique provides a means to identify targets which

can be related to the MALDI imaging data set.

Current/Future Work

�MS/MS and TLC/MS/MS using SYNAPT G2 of LSE for identification of species detected.

�Knock down LSE models which mimic disease state.

�HDMSe simultaneous MS – MS/MS of each peak during a MALDI-MS

acquisition using preset ramping collision energies

�Statistical analysis using Matlab (refining of methodology)

Acknowledgements

Co-Workers and Collaborators

•Joan Hague, Dr Anna Crecelius, Dr Josephine Bunch, Dr Karen Warburton, Dr Simona Francese, Dr Brendan Prideaux, Alex Mullen, Dr Sally Atkinson, Dr

Caroline Earnshaw, Nidhi Bindhal, Dr David Anderson, Marie-Claude Didja, Paul Trim, Laura Cole, Philippa Hart Sheffield Hallam University

•Dr Emmanuelle Claude, Dr Marten Snel, Waters

•Dr Julie Wingate, Dr Ron Bonner ABI/MDS•Dr Alan Barnes, Shimadzu

•Sue Kennerley, KR Analytical•Keith Oakes Elforlight Ltd

•Dr Klaus Dreisewerd, Univ. Muenster•Prof Michel Salzet, Dr Isobelle Fournier, Univ of Lille

•Prof Stefano Turilazzi, Prof Gloriano Moneti, CISM, Univ of Florence•Dr Don Richards, Pfizer Global R&D

•Prof Jane Thomas-Oates, Dr Sarah Robinson, Univ of York

•Dr Ruth Pendlington , Dr Maja Aleksic, Dr Raniero Zazzeroni Unilever•Dr Susan Crosland, Dr Kate Sharples, Dr Philp Green; Syngenta

•Dr Paul Loadman, Dr Chris Sutton; Univ. of Bradford. •Dr John White, Dr Jackie Morton, Dr Duncan Rimmer;

Health and Safety Laboratory•Prof Mike Burrell; Univ. of Sheffield.

•Dr Peter Marshall, Dr Josie Morell, Dr Lisa Renshaw GSK

•Dr Andrew McEwen, Jennie Avery, Quotient Bioanalytical•Mr Peter Scriven;, Prof Gillian Tozer Surgical Oncology Unit, U of Sheffield

Funding

•Pfizer Global R&D, BMSS, HSL, SHU, EPSRC/RSC, BBSRC/Syngenta, BBSRC/GSK, BMSS, EPARC/CRUK, COLIPA

Thank you!