Profiling and Imaging Mass SpectrometryProfiling and Imaging Mass Spectrometry

Compilation of Work from the Caprioli Lab

James Mobley, Ph.D.Director of Urologic ResearchAssociate Director of Mass SpectrometryUniversity of Alabama at Birminghammobleyja@uab.edu

Primary Focus of TodayPrimary Focus of Today’’s Lecture?s Lecture?

• Brief Overview of Biomarker Discovery for Clinical Applications, Why do we do it, Why do we Often use MALDI-ToF?

• Understanding Advances in MALDI-ToF Driven Profiling of Tissue Sections for BMD.

• How to produce a Mass Image from a Series of Profiles.• What to Do with All that Data.• How to ID those Peaks, are they really Proteins?

First offFirst off………………....Why Do Biomarker Discovery?Why Do Biomarker Discovery?

• To find associations between biological components (i.e. SM, FA’s, Proteins) and any clinical endpoint quickly, non-invasively, affordably.

• To non-invasively determine…..– Pathologic Changes (i.e. early detection of cancer)– Aggressiveness/ Stage of Disease– Predicting Rx Response– Drug Target Discovery– Mechanistic Studies (Systems Biology)

• The Potential Clinical Impact is Tremendous!!

Can 2D PAGE Get Us There?Can 2D PAGE Get Us There?

O. Golaz etal. (1993) Electrophoresis 14 1223-1231. Acidic region IPG, pH 3.5-10

Serum Proteome Anderson, PNAS, 1977

Direct Analysis ~1000 Protein Spots

Pieper, Proteome, 2003Affinity, SCX, Size Exclusion

Yielded 74 Fractions - 2D3700 Proteins Spots327 Distinct Proteins

25yrs later25yrs later

Primarily HMWPrimarily HMWProteins!Proteins!

MALDI-TofCrude Protein

2D PAGE, ID by MALDI-Tof

What About MALDIWhat About MALDI--ToFToF for Biomarker Discovery?for Biomarker Discovery?

Primarily LMW Proteins!Primarily LMW Proteins!

H+

H+

H+

Laser; High Energy Monochromatic Light Source

Detector

Time of Flight TubeAccelerator

DE

---

---

Rel

Inte

nsity

Time (ms) – m/z

A Closer Look? A Closer Look? Matrix Assisted Laser Desorption Ionization (MALDI)Matrix Assisted Laser Desorption Ionization (MALDI)--

Time of Flight (Time of Flight (TofTof) Mass Spectrometry) Mass Spectrometry

Surface Enhanced Probe (SELDI-Tof)

Imaging Mass Spectrometry of Thin Tissue Sections

1. To obtain information on the local protein composition at any coordinate on the section

2. To reconstruct 2-dimensional density maps (or images) for all of the signals detected

3. To extract molecular information from grouped profiles and images indicative of health status

4. To be used in “directed” and “non-directed” discovery; to answer fundamental questions relevant to protein expression in normal and diseased tissues

Acquire mass spectra

Slice frozen tissue on cryostat (~12 μm thick)

Thaw slice onto MALDI plate, allow to dry

Apply matrix

Droplets Spray coating

m/z 18 388

Non-DirectedDiscovery“Imaging”

Directed Discovery

“Profiling”

Laser

+ + +

Laser

+ + +

Protein profiles Protein images

Human breast tumor needle biopsy

% In

tens

ity

17000 27600 38200 48800 59400 700000

2

4

6

8

10 28085

17253

17381

3101144085

43255

41854

6675322265

~50500

70000 116000 162000 208000 254000 300000Mass (m/z)

0

0.6

1.2

1.8

2.4

3 81934

132990

148196

x5

162805

78494

94266 97455

199988

215149

2000 5000 8000 11000 14000 170000

8

16

24

32

4015127α+

15869β+

7564α2+ 7934

β2+

6434

8810

140375360 10091

6633

1164913760

12344

9442

8686

1 mm

Protein expression profiling by MALDI-MS

Human glioma biopsy

Schwartz et Al. Clin. Cancer Res., 10, 981-987 (2004).

15000

Tumor (A)

Tumor (B)

Non Tumor (C)

Non Tumor (D)

m/z4000 8000

Tumor (A)

Tumor (B)

Non Tumor (C)

Non Tumor (D)

X 5

A B

C

D

3 mm

Non-Tumor

Tumor

Brain glioma section stained with Methylene blue

200x

2 mm

2000 5600 9200 12800 16400 200000

10

20

30

40

50

% In

tens

ity

20000 30000 40000 50000 60000 70000Mass (m/z)

0

2

4

6

8

10%

Inte

nsity

6700 7340 7980 8620 9260 9900

Principle of MALDI MS Imaging

Frozen section

Matrixapplication

MALDI Mass Spectrum

4000 6000 8000 10000m/z

MS Images

Raster of section

Spray deposition of matrix on tissue sectionsSpray nebulizer for TLC plates

NitrogenMatrix solution Tissue section

3000 8400 13800 19200 24600 30000Mass (m/z)

0

20

40

60

80

100%

Inte

nsity

5445

6228

95694965 6575

9906

11208 140051223926838

18310

16773

20746

3000 8400 13800 19200 24600 30000Mass (m/z)

0

20

40

60

80

100

% In

tens

ity

Spotted

Sprayed

5445

6228

9569

4965

6575

9906

11208

1400512239

268351677320746

200 μm

Comparing matrix coatings: “Applied Biosystems DE-STR”

50um Laser spot

2000 8600 15200 21800 28400 35000Mass (m/z)

0

20

40

60

80

100

% In

tens

ity

1130

8, f)

1498

3

1213

4

7539

, d)

9620

1378

2

3458

6720

, c)

8565

1561

9

9978

, e)

2217

3

1841

2, g

)

5708

4965

5445

1412

7

~299

00024

779

2464

2

9978

8565

9620

9327

9293

8546

8982

9912

9079

8946

1026

0

7934

x30

m/z 6720

m/z 9978

m/z 11 308 m/z 18 412

m/z 7539

12

3

3 mm

0 100%

Mouse Brain Imaging resolution:50 μm

Chaurand et Al. Anal Chem, 76, 86A-93A (2004).

4965

8565

4747

5444

5708

7810

, β2+

6573

5171

7537 9910

6719

6251

2000 3800 5600 7400 9200 11000

7492

, α2+

6545

1025

9

3347

9193

1213

4

1994

0

1164

1

1237

2

11000 13800 16600 19400 22200 25000Mass (m/z)

1679

0

2074

3

NT)

T)

2217

3

9975

1498

2, α

+

1561

8, β

+ x25

NT)

T)

Rel

ativ

e in

tens

ities

Tumor

Tumor Non tumor

2 mm

Glioma mouse model - intracranial injection of GL261 cancer cells

m/z 12 134Cytochrome C

m/z 22 173

m/z 7539

m/z 14 039m/z 11 307 and 11 348Histone H4

m/z 13 804 and ~15 350Histones H2B1 and H3

m/z 18 420m/z 14 982 and 15 617α and β hemoglobins

m/z 6924

m/z 24 807

m/z 9910 Acyl CoA-binding protein2 mm

0 100%

Glioma mouse model. Imaging resolution: 100 μm

Chaurand et Al. Anal Chem, 76, 86A-93A (2004).

Tissue Profiling – Which Approach?

Matrix Deposition Variables:

Time Repr. Inherent Spot Res. LaserDemand BG Dependent

Spot Size

Manual Spotting seconds variable N >1mm YRobotic Spotting minutes good N ~200µ YLaser Capture hours good Y ~7µ to 100 µ N

Current Laser Spot Size for Old STR: 25µ x 50µ

The RapidSpotter

Substrate image capturing

Drop imaging

MALDI target with sample

Acoustic drop ejector with reservoir

Computer controlled motorized stage

Strobe LED for drop illumination

Acoustic Drop Ejection technology provides a nozzle-free means of generating microdroplets on-demand

yx “Ejection of microdroplets”Sample plate

Matrix reservoir

Coupling reservoir

Coupling membrane

Transducer

RF generator

Waveform generator

Current performances:- Spot size: ~180-200 µm

- Drop ejection rate: 10 Hz

- Drops per pixel: 60-80

Tissue Sectioning for Protein Profiling

100 µ

Hand Spotting Vs. Pico Spotting Vs. LCM

PicoSpot ~200µ

10x

4xHandSpot ~2mm

H & E Overlay at 4x magnification

H & E Overlay’s are Used to Determine Regions of Interest

Seamless High ResolutionImaging of Histological Slides

Whole Mouse Mammary Gland

Histology Directed Matrix Deposition

Whole Lung Sections

Lung Mets

A2 Not used A3

A4 A5 B1

B2B3 B4

C4 C3 C2 C1

Automating – “Whole Plate Profiling”

MALDI-Tof Plate

6X15dps was found to be optimal for normal glands and tumors of the breast. Spot Sets can be made to coincide with specific histologic features for each section over the entire plate.

. . .

. . .. . .

. . . .

A 12 point calibration can correct for plate placement errors from the PicoSpotter to the STR.

Rapid Spotting Over Entire Plate

Results from Mass Profiling Tissue SectionsLung Mets Vs. Late Carcinoma

M.G. – Late BCa

2 kDa 20 kDa

Calgranulin A

Lung Mets

A)

B)

Manual spotting vsThe RapidSpotter

Mouse brain,Analysis of the corpus callosum

Matrix printing

RapidSpotter Imaging of a Mouse Brain Section

Create custom plate file

1000 μm

2014 15579 29144 42708 56273 69838Mass (m/z)

Inte

nsity

m/z 17885 m/z 11839 m/z 20688

m/z 11790 m/z 6759 m/z 7338

0 %0 % 100 100 %%

Lung TumorHomogenate

Add Post-Translational Modification

MD HPLC

Fractions MwMatches

MwMatches

Database Search Protein ID

5000 12500 20000m/z

*8750 16250

MALDI MS

10000 12500 15000m/z

*Tryptic

DigestionMS/MS

100 300 500 700 900 1100 1300m/z

Schematic Representation of Protein Marker Identification

m/z 22535

m/z 8958

Combined Image

Combined Imagem/z 7930

Normal human colon biopsy

1000 µm

mucosa submucosa

muscle

Rat kidney sagittal section

m/z 9943

m/z 9979

Overlay9950 10000 10050

100

200

300

400

500

600

Inte

nsity

Mass (m/z)

9943

9979

2.5 mm

m/z 17,514

m/z 10,935

Overlay

m/z 9943

m/z 9979

OverlayMass ( )

m/z 17,514

m/z 10,935

Overlay9900

0

cortexmedula

1 cm

Histological tumor margin

Tumor

Non-tumor

Skeletal muscle

Vessel

Fibrous stroma

Molecular Determination of Tumor Margins

Robert Caldwell,

15 x 100 spots250 µm resolution

A

B

C

D

E

A

B

C D

E

Histological vs. molecular assessment of the tumor margin:

1

Laser1; Matrix droplets are regionally deposited on the sample. Each spot area is analyzed using MALDI-MS

N N N N2; Computer algorithms are used to smooth and correct baseline of each spectrum

2 3

T T T T T T T3; Hierarchical cluster analysis of human brain cancer and normal brain spectra

Brain Cancer protein expression profiling by MALDI-MS

Schwartz SA et Al, Cancer Research, 65(17), 7674-7681 (2005)

Non-Tumor

Grade 2 Tumor

Grade 3 Tumor

Grade 4 Tumor

8000 10000Mass (m/z)

Abs

olut

e In

tens

ity

Human glioma protein expression profiling by MALDI-MS

Month post surgery

P-value < 0.000119 proteins were significant

% S

urvi

ving

Pat

ient

Pop

ulat

ion

0

20

40

60

80

100

0 10 20 30 40 50 60

Long-term survival group (n = 46)

Short-term survival group (n = 45)

Kaplan-Meier survival curves for groups with poor and good prognostic. Classification according to the expression pattern of 19 distinct MS peaks

Grade 4 Glioma

N = 25

N = 26

0 10 20 30 40 50

0

0.2

0.4

0.6

0.8

1.0

P-value < 0.0058 proteins where significant

Schwartz SA et Al, Cancer Research, 65(17), 7674-7681 (2005)

HG

50 µm

LG

10650 10740 10830 10920 11010 11100Mass (m/z)

0

20

40

60

80

100

Rel

ativ

e In

tens

ities

10836

11068

10738

HG

LG

010

0% m/z 10836

HG LG

Human brain glioma – Low grade vs. high grade

Immunohistochemistry

m/z 10836, S100 Beta

Proteins are OK……

But What About Drug Distribution/ Metabolism?

Dose animal♦ orally♦ i.v.

Laser

Remove tissue

♦Cut frozenslice (12 μm)

♦Apply matrix

♦Analyze by:MALDI MS andMALDI MS/MS

Reyzer ML et al, J Mass Spectrom, 38, 1081-1092 (2003)Reyzer ML et al, Cancer Res 64, 9093-9100 (2004)

Days post-treatment0 5 10 15 20 25 30 35

0

200

400

600

800

1000

Tum

or V

olum

e (m

m3 )

Captisol 200 µl (vector)OSI 774 100 mg/kg

MMTV/HER2 transgenic mouse mammary tumors

Contributed by M. Sliwkowski (Genentech, Inc.)

● MMTV/HER2 cells transplanted in FVB female mice.

● Tumor grown to a size of ~200 mm3.

● OSI 774 is an intracellular tyrosine kinase EGF receptor inhibitor.

● Administered orally for 1 week

MS/MS analysis of OSI-774 in tumor tissueTumors removed after a single 100 mg/kg dose of OSI-774

1

65

432

4 hr

♦ Monitored CAD transition m/z 394 → 278

Spot #3

240 260 280 300 320 m/z

330

Inte

nsity

278

N

N

NH

CHC

O

O

O

O

m/z 278.1+H

m/z 336.2+H++H

N

N

NH

CHC

O

O

O

O

m/z 278.1+H

m/z 336.2+H++H

MW = 393.4

100 200 300 400 500

278.1

336.2

[OSI774 + H]+

m/z

Inte

nsity

100 200 300 400 500

278.1

336.2

[OSI774 + H]+

m/z

Inte

nsity

Dose dependence of protein alteration (20 hr after dose)

9 control mice,Av of 54 spectra

3x3 dosed mice,Av of 9 spectra per dose

Rel

ativ

e in

tens

ity

thymosin β 4

4700 4780 4860 4940 5020 5100

8300 8400 8500 8600 8700 8800Mass (m/z)

Control10 mg/kg30 mg/kg100 mg/kg

1900 7520 13140 18760 24380 30000Mass (m/z)

Average Ion Spectrum for Whole Rat Sagittal Section

1 cm

Kidney Cortex m/z 4643Cecum wall m/z 5155Thymus m/z 6893Heart m/z 7921Liver m/z 14321Caudal region m/z 14400Brain m/z 21952Muscle m/z 11836

Whole Rat Imaging Sheerin Khatib-Shahid

1 cm

100%100%50%50%00%%

ssiiggnnaall

iinntteennssiittyy

Optical image Brain

m/z 20736

White matter

m/z 10271

Corpus Callosum

m/z 5978

Eye socket

m/z 6858

Rat Head Subsection

Whole Rat Sagittal Section Imaging

Work by Sheerin Khatib-Shahid

Liver

Kidney

Spleen

Cecum

Stomach

Heart

Lung

1 cm

MALDI MS/MS Image of OLZ (m/z 256) in stomach 2 hours post-dose

Optical image of orally dosed (8 mg/mL) rat abdomen

Drug Imaging

MS/MS Image of OLZ, 10x signal intensity with stomach signals masked

N

NH S CH3

N

N

CH3

282

256

213

Olanzapine(MW 313)

Inte

nsity

240 250 260 270 280 290Mass (m/z)

282.1

256.1

kidney liver lung spinal cord brain

testis bladder ? thoracic cavity

thymus

2h Optical Image

Olanzapine m/z 313→256

N-desmethyl metabolite m/z 299→256

2-hydroxymethyl metabolite m/z 329→272

fur

Areas of Further Development

- Increase number of proteins analyzed to >>1000- Develop sample protocols to investigate peptides - Achieve high sensitivity above 50 kDa- Integrate detergents into tissue protocol for membrane

associated proteins

- Pixel-to-pixel cycle time of about 100 ms needed

- Integrate bioinformatic tools, image processing

- Develop sample protocols targeted at specific tissues/diseases

- Improve protein identification protocols

Conclusions

- Excellent discovery tool

- MW annotated patterns keyed to tissue location

- Augments (not replaces) current molecular technologies

- Significant clinical potential (disease diagnosis, state and progression, prognosis, risk assessment (?)

- Provides temporal proteomic profile that, with genomic profile, will be vital to personalized medicine

Tissue Profiling/Imaging MS

FundingNIH (NCI, GMS, NIDA)Vanderbilt University

Acknowledgments

OthersPer Andren, Uppsala UniversityRobert Weil, NIH Walter Korfmacher, Schering-PloughMarkus Stoeckli, Novartis

Vanderbilt CollaboratorsCarlos Arteaga (Breat SPORE Dir.)David Carbone (Lung SPORE Dir.)Robert Coffey Jr. (GI SPORE Dir.)Marie-Claire Orgebin-CristAriel DeutchDennis HallahanPat Levitt Robert MatusikJason Moore Hal MosesJennifer PietenpolNed Porter Yu ShyrRobert WhiteheadKen Schriver

MSRCRichard Caprioli (Director)Shannon CornettJames MobleyMichelle ReyzerJeremy NorrisRobert CaldwellErin SeeleyStacey OppenheimerSheerin Khatib-ShahidPierre ChaurrandKristin BurnumKristen HerringSaturday PuolitaivalHans Reudi AerniGreg Bowersock (IT)Ken Shriver