High Recovery FASP Applied to the Proteomic ... - ACS Publications

ARTICLE pubs.acs.org/jpr

High Recovery FASP Applied to the Proteomic Analysis of Microdissected Formalin Fixed Paraffin Embedded Cancer Tissues Retrieves Known Colon Cancer Markers Jacek R. Wisniewski,*,† Pawel Ostasiewicz,‡ and Matthias Mann*,† † ‡

Department of Proteomics and Signal Transduction, Max-Planck-Institute of Biochemistry, D-82152 Martinsried, Germany Department of Pathology, Wroczaw Medical University, PL-50368 Wroczaw, Poland

bS Supporting Information ABSTRACT: Proteomic analysis of samples isolated by laser capture microdissection from clinical specimens requires sample preparation and fractionation methods suitable for small amounts of protein. Here we describe a streamlined filter-aided sample preparation (FASP) workflow that allows efficient analysis of lysates from low numbers of cells. Addition of carrier substances such as polyethylene glycol or dextran to the processed samples improves the peptide yields in the low to submicrogram range. In a single LCMS/MS run, analyses of 500, 1000, and 3000 cells allowed identification of 905, 1536, and 2055 proteins, respectively. Incorporation of an additional SAX fractionation step at somewhat higher amounts enabled the analysis of formalin fixed and paraffin embedded human tissues prepared by LCM to a depth of 36004400 proteins per single experiment. We applied this workflow to compare archival neoplastic and matched normal colonic mucosa cancer specimens for three patients. Label-free quantification of more than 6000 proteins verified this technology through the differential expression of 30 known colon cancer markers. These included Carcino-Embryonic Antigen (CEA), the most widely used colon cancer marker, complement decay accelerating factor (DAF, CD55) and Metastasis-associated in colon cancer protein 1 (MACC1). Concordant with literature knowledge, mucin 1 was overexpressed and mucin 2 underexpressed in all three patients. These results show that FASP is suitable for the low level analysis of microdissected tissue and that it has the potential for exploration of clinical samples for biomarker and drug target discovery. KEYWORDS: FFPE, FASP, laser capture microdissection, colon cancer, high resolution mass spectrometry

’ INTRODUCTION Proteomic analysis of biological or clinical samples is generally limited by their availability. Typically, the small numbers of cells that can be isolated are not sufficient to characterize the proteome in depth. Several recent publications have focused on this technical problem and proposed approaches for the proteomic analysis of low numbers of cells. Wang et al. described protein identification from 500 to 5000 MCF7 cells.1 The cells were lysed with NP-40 and the proteins were precipitated with acetone. After tryptic digestion the samples were analyzed by LC-MS/MS using a Quadrupole Time-of-Flight mass spectrometer and 167 and 619 proteins were identified from 500 and 5000 cells, respectively. In another study, a proteomic reactor was applied to study differentiation of human embryonic stem cells.2 Material from 50 000 cells was applied to the reactor, peptides were separated into 13 fractions by an SCX monolithic column followed online by LCMS/MS using an Orbitrap XL mass spectrometer, leading to the identification of 2281 proteins. Waanders et al. have demonstrated the analysis of single islets of Langerhans from mice, which consist of 20004000 cells. Employing a high sensitivity chromatographic system that measures nanogram protein mixtures by splitting r 2011 American Chemical Society

gradient effluents into a capture capillary, they identified 2400 proteins in single analyses.3 Laser-capture microdissection (LCM) is a popular method for the analysis of mRNA of cancer tissues, and many proteomics studies have also focused on this method. Cha et al.4 performed a comparative analysis of normal and invasive malignant breast epithelial samples. Cryosections were fixed with ethanol, stained with hematoxylin, and after air-drying subjected to LCM. The microdissected material containing 60 000 cells per sample was lysed with lithium dodecyl sulfate and the extracted proteins were separated by SDS-PAGE. Gel lanes were cut into three sections, “in-gel” digested, and the released peptides were measured by LCMS/MS. Analysis of a total of 18 samples resulted in identification of 12 970 peptides corresponding to 1623 proteins. Proteins extracted from formalin fixed paraffin embedded (FFPE) tissue have been analyzed in many papers, but usually large amounts of LCM tissue were used (reviewed in refs 5 and 6). The ability to combine LCM and in-depth proteomic analysis of FFPE tissues was demonstrated in recent publications in Received: January 10, 2011 Published: April 28, 2011 3040

dx.doi.org/10.1021/pr200019m | J. Proteome Res. 2011, 10, 3040–3049

Journal of Proteome Research which lung adenocarcinoma7 and squamous cell carcinoma of the head and neck8 were investigated. From tissue samples corresponding to 20 00030 000 cells, in total 650900 proteins were identified. Another study used microdissected cells from tongue cancer and adjacent normal epithelia.9 From 20 patientmatched samples containing approximately 10 000 cells 25 018 MS peaks were quantified between runs. Among these mass peaks, 11 corresponded to peptides that appeared to differ significantly between cancer and normal tissues; however, 6 of these corresponded to keratins. Recently, on the basis of our observation that high levels of SDS solubilized material could readily be analyzed by LCMS/ MS,10 we have developed the filter aided sample preparation (FASP) method.11 The technique features efficient exchange of SDS for urea in a centrifugal ultrafiltration unit, followed by protein digestion and isolation of peptides, which are then eluted in high purity for LCMS/MS. The usefulness of FASP for proteomic analysis, including identification of phosphorylation and N-glycosylation sites, has already been demonstrated in several publications.1218 In a response to a letter questioning applicability of FASP at low protein levels,19 we have already demonstrated suitability for samples containing submicrogram protein amounts.20 Here we present additional evidence that the standard FASP protocol combined with a single LCMS/MSrun analysis enable efficient proteomic profiling of samples originating from 100010 000 cells using an Orbitrap Velos instrument. Furthermore, we show that for low protein amounts the performance of FASP can be improved by addition of high molecular weight carrier substances to the lysate (such as polyethylene glycol or dextran) by increasing the peptide yields at submicrogram protein amounts. We apply this slightly modified FASP procedure to the analysis of LCM microdissected FFPE cancer and normal material. A simple six fraction SAX procedure12 led to the identification of more than 4000 proteins, whereas the analysis of three patient-matched colon cancer samples pairs allowed identification of more than 6000 proteins and quantification of 30 previously described colon cancer markers.

’ EXPERIMENTAL PROCEDURES Formalin Fixed Paraffin Embedded Human Tissue

Neoplastic tissue specimens with corresponding surgical margins (serving as a proxy for healthy tissue) were fixed with 4% buffered formalin. They were dehydrated via an ethanol/ xylene series and embedded in paraffin. Samples were selected based on the abundance of target cell type and suitability for microdissection without stromal contamination. All clinical samples were obtained from the Department of Pathology of Wroczaw Medical University. Mouse liver was fixed as described previously.16 Microdissection

Membrane Slides 1.0 PEN (Zeiss, G€ottingen, Germany) were irradiated with UV-light for 45 min before use. For microdissection FFPE-samples were sliced in a microtome (7 μm sections), mounted on the membrane slides and dried at 37 °C for 2 h. Then they were deparaffinized with xylene (2.5 min followed by 1.5 min incubation), and hydrated in absolute ethanol, 70% ethanol and water, each for 1 min. Sections were stained with Mayer’s hematoxylin for 20 s, rinsed with water for 1 min, and air-dried. Tissue was dissected with the Laser Pressure

ARTICLE

Catapulting (LPC) PALM Instrument (Zeiss, G€ottingen, Germany). Tissue areas to be subjected to laser microdissection were marked using a 10 objective and dissected in Robo-LPC mode. Dissected tissue was collected in Adhesive Caps (Zeiss, G€ottingen, Germany). Protein Extraction from FFPE

Microdissected tissue was lysed in a buffer consisting of 0.1 M Tris-HCl, pH 8.0, 0.1 M DTT, 0.5% (w/v) polyethylene glycol 20 000 and 4% SDS at 99 °C in a heating block with agitation (600 rpm) for 1 h. Approximately 1 μL of buffer was used for 10 nL of the microdissected tissue. The crude extract was then clarified by centrifugation at 16 000 g at 18 °C for 10 min and stored frozen. Cell Culture

After determination of the density of the harvested HeLa and RKO cells, aliquots containing 100010000 cells were collected in individual tubes. Cell pellets were flash frozen in liquid nitrogen and stored at 80 °C. Cells were lysed in a buffer consisting of 0.1 M Tris-HCl, pH 8.0, 0.1 M DTT, and 4% SDS at 99 °C for 5 min. In some experiments the buffer was supplemented by polyethylene glycol 20 000 or Dextran T70 as described. Protein Digestion by FASP

Detergent was removed from the lysates and the proteins were digested with trypsin using the FASP protocol11 using spin ultrafiltration units of nominal molecular weight cut of 30 000. Briefly, to YM-30 microcon filter units (Cat No. MRCF0R030, Millipore) containing protein concentrates, 200 μL of 8 M urea in 0.1 M Tris/HCl, pH 8.5 (UA), was added and samples were centrifuged at 14 000 g at 20 °C for 15 min. This step was performed twice. Then 50 μL of 0.05 M iodoacetamide in 8 M urea was added to the filters and the samples were incubated in darkness for 20 min. Filters were washed twice with 100 μL of 8 M UA followed by two washes with 100 μL of 40 mM NH4HCO3. Finally, trypsin (Promega, Madison, WI) was added in 40 μL of 40 mM NH4HCO3 to each filter. The protein to enzyme ratio was 100:1. Samples were incubated overnight at 37 °C and released peptides were collected by centrifugation. Total Protein and Peptide Determination

Protein content was determined using a Cary Eclipse Fluorescence Spectrometer (Varian, Palo Alto, CA) as described previously.21 Tissue lysates were assayed by adding 12 μL of sample or tryptophan standard to 2 mL of 8 M urea in 10 mM Tris-HCl pH 8.0. The peptides resulting from FASP digests were analyzed in 0.2 mL of 40 mM NH4HCO3 in 5 5 mm quartz cells. Fluorescence was measured at 295 nm for excitation and 350 nm for emission. The slits were set to 5 and 20 nm for excitation and emission, respectively. In-gel digestion

Aliquots of SDS HeLa lysates were separated by SDS-PAGE, using NuPAGE Novex Bis-Tris gels (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. The gel was stained with Colloidal Blue Staining Kit (Invitrogen). Each lane was cut into 10 slices and in-gel digestion was performed basically as described in22 but without protein carbamidomethylation. Gel plugs were washed with 50 mM ammonium bicarbonate, 50% ethanol and incubated with 10 mM DTT in 50 mM ammonium bicarbonate for 1 h at 56 °C for protein reduction. Gel pieces were washed twice with a 50 mM ammonium bicarbonate, 50% CH3CN solution, dehydrated with 100% ethanol and dried in a 3041

dx.doi.org/10.1021/pr200019m |J. Proteome Res. 2011, 10, 3040–3049

Journal of Proteome Research

ARTICLE

vacuum concentrator. Gel pieces were rehydrated in 250 μL 50 mM NH4HCO3 and digested with trypsin (protein to enzyme 1:50) at 37 °C overnight. Supernatants were transferred to new tubes, and the peptides remaining in the gel were extracted by incubating gel pieces two times with 30% CH3CN, followed by dehydration with 100% CH3CN. The extracts were combined, vacuum concentrated, and assayed for peptide content after passing them through 0.22 μm filters. Peptide Fractionation

Peptides were fractionated according to the previously described pipet tip protocol (FASP-SAX)12 with minor modifications. Briefly, peptides were loaded into tip-columns made by stacking six layers of a 3 M Empore Anion Exchange disk (12145012 Varian, Palo Alto, CA) into a 200 μL micropipet tip. For column equilibration and elution of fractions, we used Britton & Robinson universal buffer (BRUB) composed of 20 mM acetic acid, 20 mM phosphoric acid, and 20 mM boric acid titrated with NaOH to the desired pH. Peptides were loaded at pH 11 and fractions were subsequently eluted with buffer solutions of pH 8, 6, 5, 4, and 2 in turn. Entire eluted fractions of peptides were analyzed by LCMS/MS. LCMS/MS Analysis

LC was performed on a Proxeon Easy-nLC System (Proxeon Biosystems, Odense, Denmark; now Thermo Fisher Scientific). Peptides were separated on a 15 cm fused silica emitter (Proxeon Biosystems) packed in-house with the reverse phase material ReproSil-Pur C18AQ, 3 μm resin (Dr. Maisch, AmmerbuchEntringen, Germany) with a 230 min gradient from 2 to 80% of 80% (v/v) CH3CN, 0.5% (v/v) acetic acid. For the investigation of unfixed cells an LTQ-Orbitrap XL (Thermo Fisher Scientific, Bremen, Germany) was used whereas FFPE samples were analyzed with an LTQ-Orbitrap Velos mass spectrometer. In the LTQ-Orbitrap XL full mass range scans at a resolution of 60 000 at m/z 400 were acquired while up to 10 MS/MS spectra were acquired at low resolution in the linear ion trap (“top10” method). On the LTQ-Orbitrap Velos, MS scans were combined either with tandem mass spectrometry by CID in the linear ion trap or in a “highhigh” strategy with Higher Energy Collisional Dissociation (HCD)23 with high accuracy analysis of the fragment ions in the Orbitrap analyzer. For HCD analysis, transients of 100 ms were acquired, corresponding to a resolution of 7500. Because HCD does not operate in parallel with the acquisition of the MS spectrum, transients were limited to 0.5 s for the MS spectra in the Velos instrument (30 000 resolution). The high sequencing speed of the Velos instrument allowed employing a “top20” method for CID (target value for MS/MS 5000) and a “top10” method for HCD (target values for MS/MS 50 000). Data Analysis

The MS data were analyzed using the software environment MaxQuant24 version 1.1.1.25. Proteins were identified by searching MS and MS/MS data of peptides against both a regular and a decoy version of the International Protein Index (IPI) database for mouse (v. 3.68) and human (v. 3.68). Carbamidomethylation of cysteines was set as fixed modification. The minimum peptide length was specified to be 6 amino acids. The initial maximal mass tolerance in MS mode was set to 7 ppm, whereas fragment mass tolerance was set to 0.5 Th for CID data and 20 ppm for HCD data. The maximum false peptide discovery rate was specified as 0.01. Label free quantification was carried out in MaxQuant as previously described.25 Briefly, the peptide identifications corresponding to the isotope

Figure 1. Optimization of sample digestion conditions. (A) Comparison of the peptide yield using the FASP and the in-gel digestion methods. Equal amounts of HeLa lysates were loaded onto the ultrafiltration filter and processed by the FASP protocol or loaded onto a gel, followed by electrophoretic separation and digestion according to the ingel protocol. Experiments were performed in duplicate (1 and 2). The peptide yields were measured by fluorescence using 295 and 350 nm for excitation and emission, respectively. (B) Effect of 0.5% PEG 20 000 on the peptide yield. HeLa lysates were processed by the FASP procedure in the absence (control) and presence of 0.5% PEG 20 000. The obtained peptides were quantified as in A. (C and D) Mass spectrometric analysis of HeLa lysate aliquots containing 0.5 μg and 2.5 μg total protein. Peptides were prepared by the FASP method in the absence or presence of PEG 20,000 or Dextran T-70. (C) Number of identified peptides per single run. (D) Number of identified proteins per single run. Experiments shown in panels BD were performed in triplicate.

patterns were matched to each other across runs using the very high mass accuracy and nonlinearly remapped retention time. Total peptide signals within each run were normalized to enable comparison between experiments performed at different times.

’ RESULTS AND DISCUSSION FASP Optimization for Analysis of Low Sample Amounts

In proteomic experiments, efficient identification and reliable quantification strongly depends on the quality of protein 3042



ARTICLE

Figure 2. Samples contain only trace amounts of PEG chains. (A) Total ion currents (TICs) in LCMS of a sample processed in the presence of 0.5% PEG. (B) MS spectrum of PEG chains.

solubilization and digestion. Therefore before we investigated the analysis of very low levels of protein, we first compared the FASP method to the standard “in gel” digestion procedure.26 We selected the in gel technique because this approach has been reported to be superior to several other fractionation and digestion methods.2729 Moreover, the gel digestion has been used in a recent analysis of microdissected tissue.4 For the method comparison, the same amounts of total cell lysate were processed either with the FASP or with the in gel procedure. The FASP protocol specifies use of ultrafiltration devices with nominal molecular weight cutoff of 30 000, which have best performance for detergent removal, peptide yield and filtration rates.30 We found that the yields obtained with FASP were always above 50% of the loaded protein amount, whereas the in gel procedure yielded on average less than 20% (Figure 1A). This result is in agreement with observations made by Katayama et al.31 and indicates that the gel based approach may not be optimal for very low protein amounts. The recently reported advantages of the in gel strategy over several other methods28,29 presumably solely reflect its robustness and convenience of sample handling of this method. Addition of High Molecular Weight Carrier Substances Improves Peptide Yields

Unspecific adsorption of analytes to surfaces always accompanies biochemical procedures and reduction of the extent of this loss is often essential in high sensitivity work. Addition of carrier substances should reduce adsorption and improve yields provided they do not interfere with subsequent analysis. Since in FASP high molecular weight nonproteinaceous substances remain in the filter after protein digestion we tested if common carrier substances that are detrimental to MS analysis could nevertheless be used in FASP. When adding PEG 20 000 and Dextran T-70 to the filters we found that they did not negatively affect the LCMS/MS procedure. For samples containing submicrogram to a few microgram of total protein the peptide

yields improved by up to 30% due to the addition of high molecular weight PEG (Figure 1B). Above 10 μg of total protein we did not observe any significant increase of the peptide yield upon addition of the carriers. Peptide and protein identification improved in samples processed by FASP in the presence of the carriers (Figure 1C and D). As this effect was slightly larger for PEG 20 000 compared to Dextran all experiments described below used PEG. Trace amounts of low molecular mass polyethylene chains originating from spontaneous degradation of long PEG chains were observed to elute during the washing of the LC-column but they did not affect the peptide identification process (Figure 2). When peptides were fractionated by SAX the contamination was observed only in the flow-through fraction. Analysis of Low Numbers of Cultured Cells

To test the performance of FASP with samples containing only a few thousands cells we analyzed lysates prepared from 500, 1000, 3300, or 10 000 HeLa or RKO cells. Single analysis of the digest of 1000 HeLa cells using 2 h LCMS/MS runs allowed identification of 5,300 sequence unique peptides. For RKO cells 3900 peptides were identified (Figure 3A, Supplementary Table 1, Supporting Information). A 10-fold higher number of cells resulted in about 6000 HeLa and 5500 RKO peptides. These peptides correspond to 1100 HeLa and 900 RKO proteins (1000 cells) and 1400 HeLa and 1200 RKO proteins (10 000 cells), respectively (Figure 3B). The number of identifications per single analysis can substantially be increased by extending the analysis time to 4 h LC-gradients. These longer gradients achieved on average 1500 protein identifications from 1000 HeLa cells and 2500 proteins from 10 000 HeLa cells (Figure 3C). To put these results into perspective, we compared them to those from the recently published study by Wang et al.1 For an equivalent number of analyzed cells, we here identified 510fold more proteins using the FASP-based workflow. Likewise, the identification of more than 2000 proteins in a single 4 h 3043



ARTICLE

Figure 3. Identification of peptides and proteins using 50010 000 cultured HeLa or RKO cells. (A and B) Identification of peptides (A) and proteins (B) using an LTQ Orbitrap XL instrument and a 2 h LC gradient. (C) Identification of proteins from HeLa cells using an LTQ Orbitrap Velos instrument and a 4 h LC gradient. Measurements shown in panels A and B were performed in triplicate with the exception of the 1000 HeLa cell sample which was measured in duplicate. Data shown in panel C are from 4 analyses.

Table 1. Summary of Single Analyses of the Laser-Capture Microdissected FFPE Samples colon

normal

ductal in situ

breast

cancer

colon

carcinoma

adenosis

Total identification in 6 SAX fractions Proteins

4419

4410

4398

3645

Peptides

24 323

22 862

23 135

17 065

Membrane

39.5

38.8

35.8

36.5

Integral to

19.6

19.3

16.6

15.5

Plasma membrane 17.4

18.3

16.3

18.1

Nucleus

36.6

32.5

36.7

33.9

Mitochondrion

19.8

19.1

18.1

18.3

GeneOntology annotation (%)

membrane

Analysis of Microdissected FFPE Tissue

Figure 4. Tissues microdissection. (A) Pieces of FFPE mouse liver were LPC microdissected and processed by the FFPE-FASP procedure. Total peptide amount was determined by fluorescence measurements. (B) Microdissection of normal colonic mucosa (ac) and colon cancer samples (df).

LCMS/MS run from 3300 cells (Figure 3C, Supplementary Table 2, Supporting Information) compares favorably with identification of 2200 proteins from 50 000 hESCs in the recently described “rare cell proteomic reactor” coupled to two-dimensional LCMS/MS.2

As shown above, in single runs only a moderate depth of proteome coverage—about 20003000 proteins—was achieved. Therefore, for a more comprehensive analysis additional fractionation before LCMS/MS was applied. To enable the indepth analysis of microdissected clinical material we first tested the peptide yields from the FFPE by the FFPE-FASP procedure using mouse liver. We obtained a linear relationship between the volume of microdissected tissue and peptide yield showing that for 1 μg peptide about 30 nL tissue were required (Figure 4A). Using human samples, we found that 175 nL of microdissected tissue yielded about 57 μg of peptide material (not shown). We estimate that we dissected the equivalent of 20 000 cells per single experiment but such estimates are imprecise because the shapes and sizes of cells in tissues are highly heterogeneous To test the performance of the FFPE-FASP-SAX protocol we next analyzed microdissected material from human breast and colon cancer samples. We isolated cells of breast adenosis, which is a noncancerous proliferation of glands, carrying no increased 3044


Ribosome biogenesis protein

Caspase-7

Core-binding factor subunit beta

Complement decay-accelerating factor

Carcinoembryonic antigen

Chromogranin-A

Calcium-activated chloride channel regulator 1

Cytosolic nonspecific dipeptidase

FABP1 protein

IgGFc-binding protein

High mobility group protein B1

Laminin subunit gamma-2

Galectin-4

LIM domain and actin binding protein 1

Metastasis-associated in colon cancer protein 1

BOP1

CASP7

CBFB

CD55; DAF

CEA

CHGA

CLCA1

CNDP2

FABP1

FCGBP

HMGB1

LAMC2

LGALS4

LIMA1

MACC1

3045

Alpha-1,6-mannosylglycoprotein

MGAT5

Metalloendopeptidase OMA1, mitochondrial

Mucin-1

Mucin-2

Nucleophosmin

Olfactomedin-4

Phospholipase A2, membrane associated Phosphoserine aminotransferase

MPRP1

MUC1

MUC2

NPM1

OLFM4

PLA2G2A PSAT1

6-beta-N-acetylglucosaminy-ltransferase A

Mitotic spindle assembly checkpoint protein MAD2A

MAD2L1

BOP1

Anterior gradient protein 2 homologue

protein name

AGR2

gene names

2 14

27

14

60

6

5

4

3

2

22

16

6

13

98

12

22

47

17

8

4

4

7

9

11

peptides

17.4 35.2

48.2

44.6

14.1

6

10.7

7.3

16.1

2.3

34.1

43.3

7.1

46

33.2

79.9

56

42.6

39.4

17.2

6.4

17.6

25.3

15.4

53.8

sequence coverage [%]

0.17 6.7

26

1.3

0.12

2.62

0.36

3.6

4.6

22

8.1

0.26

6.3

2.5

0.12

0.36

0.70

0.02

0.11

0.85

1.5

1.6

3.7

2.6

1.4

1

b

0.06 >100

0.06

6.1

0.02

>100

0.53

15

2.5

6.4

0.32

0.56

9.2

3.1

0.01

0.59

0.17

0.01

0.02

>100

1.2

1.60

0.12

0.15 63

>100

2.5

0.03

9.5

4.48

4.8

0.53

>100

0.28

0.24

39

3.2

0.01

0.01

0.83

0.00

0.00

96

13

2.9

0.11

5.8

0.10

0.27a 6.0

3

2

sample pair

intensity ratio of neoplastic to normal mucosa

Table 2. Examples of Known Proteins Deregulated in Colon Cancer which were Found in Our Data Set

Downregulated in CRC44 Upregulated in CRC61

confirmed with IHC60

Protein overexpressed in CRC, discovered with use of MS and

Protein involved in carcinogenesis, overexpressed in CRC58,59

with proliferation markers and with poor prognosis42,57

Marker of differentiated intestinal cells, its reduced expression correlates

Frequently overexpressed in CRC, marker of poor prognosis42

Well known protein, frequently overexpressed in human malignancies including CRC56

potential55

Glycoprotein involved in oncogenesis, increasing tumor metastatic

Upregulated in CRC54

Frequently overexpressed in CRC, marker of poor prognosis53

Downregulated in CRC44


expression correlates with poor prognosis51 Upregulated in CRC52

Protein frequently overexpressed in human malignancies, in CRC its



Overexpressed in CRC50

Regulator of calcium channels, frequently downregulated in CRC49

exception of neuroendocrine tumors48

in CRC40 Protein downregulated in all human malignancies including CRC with

Widely used serum marker of CRC, frequently overexpressed


independent prognostic factor41

Protein frequently overexpressed in CRC, recently proved to be




comment (reference)

Journal of Proteome Research ARTICLE


Upregulated in CRC66 19 48 0.31

Median values are in bold case; CRC, colorectal cancer. Ratios of 0.00 and >100 reflect presence/absence values or are beyond the ratios that we can confidently assign.

a

Plasminogen activator inhibitor 2

Serpin B5

Transcription factor SOX-9

Thioredoxin

Uracil-DNA glycosylase

SERPINB2

SERPINB5

SOX9

TXN

UNG

gene names

Table 2. Continued

protein name

b

7

31.9

including CRC65

Protein frequently overexpressed in many malignancies, 0.81 0.68 7

51.4

1.3

Protein frequently overexpressed in many malignancies, including CRC64 53 0.71 6

12

>100

Upregulated in CRC63 8.2 0.05 8.4 13

43.5

including CRC62

Protein frequently overexpressed in many malignancies, 1.0 0.53 2

4.3

0.34

3 1

2

ARTICLE

sequence coverage [%] peptides

sample pair

mucosa

intensity ratio of neoplastic to normal

comment (reference)


Figure 5. HCD MS/MS spectra of peptides matching Carcinoembryonic antigen (A) and Complement decay-accelerating factor (DAF;CD55) (B). Δm, mass difference between the observed and the calculated value in ppm.

risk of carcinoma, as well as of breast Ductal In Situ Carcinoma (DCIS) tissue, a preinvasive lesion requiring oncological treatment. We also dissected normal and neoplastic colonic mucosa. Examples illustrating the microdissection process and collection of the tissues are shown in Figure 4B. The collected tissue (175 nL) was lysed in the presence of 0.5% PEG 20 000 for 1 h and the lysate was processed using the FASP procedure (Experimental Procedures). The peptides were fractionated on anion exchangers in the StageTip format and were analyzed by LCMS/MS using Orbitrap Velos instruments. Per single sample we identified 17 00024 000 peptides corresponding to 36004400 proteins (Table 1, Supplementary Table 3 and 5, Supporting Information). Gene Ontology analysis revealed a high content of membrane proteins (3640%), and especially plasma membrane proteins (1618%) as expected from the suitability of FASP for the analysis of this class of proteins.32 Comparative analysis of patient matched non-neoplastic and neoplastic lesions of the breast reveals significant overexpression of proteins known to be involved in cell cycle progression like Ki67 (a widely used marker of proliferation), topoisomerases I, II alpha, II beta,33 cyclin dependent kinases (cdk 2,4, 7, 9, 11)34 and also 3046


Journal of Proteome Research some other markers of breast cancer like bub3,35,36 fortilin,37 and NUCKS.38 Comparative Analysis of Patient Matched Normal Colon and Colon Cancer Samples

The above results demonstrate that the described proteomic platform is useful for analyzing archival clinical FFPE samples and suggested that it has the potential for biomarker discovery. To investigate this further, we analyzed in total three patient matchedpair samples of normal and neoplastic colonic tissue. Raw data obtained from the analysis of these six proteomes were analyzed together in MaxQuant, using label-free protein quantification between cancer and adjacent tissue. In colon cancer and corresponding normal mucosa we identified per single sample 39 953 ( 1060 and 38 677 ( 1305 peptides representing 5985 ( 54 and 5868 ( 110 different proteins, respectively (Supplementary Table 4 and 5, Supporting Information). Analysis of the data revealed several proteins differentially expressed in normal vs neoplastic mucosa. Interestingly many of them were previously known established or potential colorectal cancer markers (Table 2). Each of these markers was identified and quantified on the basis of multiple peptides. The abundance of 26 of the 30 proteins listed in Table 2 changed in the cancer specimens accordingly to the previously observed over- or under-expression (Table 2). The fact that not all markers are consistently deregulated in all individuals in our data may reflect the known heterogeneity of this cancer.39 Carcinoembryonic antigen (CEA) is the most commonly used colorectal cancer marker and is overexpressed in more than 90% of primary CRC but not in all of them.40 Here we found about hundred-fold overexpression of CEA in two of the three patients. Among the markers listed in Table 3 there are proteins with diverse functions from all cell compartments: membrane (DAF, CD55), nucleus (Nucleophosmin), cytoplasm (Caspase 7), secretory vesicles/extracellular matrix (Mucin-2) and mitochondria (Thioredoxin-2). Since expression level can roughly be estimated from label free quantification—and usually much more accurately than by immunohistochemistry—the acquired data may also provide therapeutic clues. For example, overexpression of MRP1 indicates resistance to some therapeutics whereas Cbfb,41 Mucin-1 and -242 offer prognostic information. We also note that the certainty of identification provided by the above workflow is very high. Illustrating this, Figure 5 shows the high resolution HCD spectra identified peptides matching CEA and CD55.

’ CONCLUSIONS The analysis of small amounts of sample for the identification of biomarkers and drug targets is a long-standing challenge in proteomics. In particular, there are many reports dealing with the proteomic analysis of colorectal cancer (reviewed by Jimenez et al.43). Often, these investigations have suffered from the application of immature technology limiting the analysis to the identification of proteins that are very abundant and readily soluble. As a result, well-known colorectal cancer markers such as CEA or mucins, which are frequently used in histopathological analysis,39 were usually not detected in previous proteomic studies. In this work, we describe a workflow allowing efficient analysis of minute sample amounts. In comparison to the standard FASP approach, we improved the peptide yields at low amounts by adding PEG 20 000 as a carrier substance. We show that this modified FASP method allows analysis of small numbers of cells

ARTICLE

with excellent peptide yields. The depth of analysis at very low amounts of sample now appears to be constrained as much by the mass spectrometric analysis than by the sample preparation. Currently, the improved FFPE-FASP-SAX approach allows quantitative comparisons of healthy and neoplastic tissues to a depth of more than 6000 proteins per single sample obtained from 175 nL microdissected FFPE tissue. This enables comparison of the levels of know cancer markers and opens up new possibilities in clinical samples analysis. An important benefit of the approach described here is that it is not restricted to biobanks of frozen specimens but that it can be used for fixed archival material. While the current work was not aimed at the identification of novel biomarkers, it clearly shows the potential to do so. The retrieval of known, clinically used markers is already a very important step in demonstrating the potential role of proteomics in analyzing different cancer types. To our knowledge, there are very few studies that independently “rediscover” these proteins as they are usually of too low abundance or not quantified accurately enough. Therefore, the ability of our proteomics workflow to correctly retrieve these markers and to successfully analyze two different cancer types bodes well for the potential application of proteomics in the clinic. In the future, we plan to extend these results by analyzing larger numbers of clinical samples, which will potentially enable statistically sound identification of colon cancer related proteins.

’ ASSOCIATED CONTENT

bS

Supporting Information Supplemental tables. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]; [email protected]. de. Department of Proteomics and Signal Transduction, MaxPlanck Institute of Biochemistry, Am Klopferspitz 18, D-82152 Martinsried, Germany.

’ ACKNOWLEDGMENT We thank Dr. Piotr Ziozkowski (Wroclaw Medical University) for providing FFPE material, Dr. Nagarjuna Nagaraj and Korbinian Mayr for assistance in mass spectrometric analysis and Gabrielle Sowa for technical assistance. This work was supported by the Max-Planck Society for the Advancement of Science, by the European Commission’s seventh Framework Program (grant agreement HEALTH-F4-2008-201648/PROSPECTS) and the Munich Center for Integrated Protein Science (CIPSM). ’ REFERENCES (1) Wang, N.; Xu, M.; Wang, P.; Li, L. Development of mass spectrometry-based shotgun method for proteome analysis of 500 to 5000 cancer cells. Anal. Chem. 2010, 82 (6), 2262–71. (2) Tian, R.; Wang, S.; Elisma, F.; Li, L.; Zhou, H.; Wang, L.; Figeys, D. Rare cell proteomic reactor applied to SILAC based quantitative proteomic study of human embryonic stem cell differentiation. Mol. Cell. Proteomics 2011, 10, M110.000679. (3) Waanders, L. F.; Chwalek, K.; Monetti, M.; Kumar, C.; Lammert, E.; Mann, M. Quantitative proteomic analysis of single pancreatic islets. Proc. Natl. Acad. Sci. U.S.A. 2009, 106 (45), 18902–7. 3047


Journal of Proteome Research (4) Cha, S.; Imielinski, M. B.; Rejtar, T.; Richardson, E. A.; Thakur, D.; Sgroi, D. C.; Karger, B. L. In Situ proteomic analysis of human breast cancer epithelial cells using laser capture microdissection (LCM)-LC/ MS: Annotation by protein set enrichment analysis (PSEA) and gene ontology (GO). Mol. Cell. Proteomics 2010, 9, 2529–44. (5) Matsuda, K. M.; Chung, J. Y.; Hewitt, S. M. Histo-proteomic profiling of formalin-fixed, paraffin-embedded tissue. Expert Rev. Proteomics 2010, 7 (2), 227–37. (6) Reimel, B. A.; Pan, S.; May, D. H.; Shaffer, S. A.; Goodlett, D. R.; McIntosh, M. W.; Yerian, L. M.; Bronner, M. P.; Chen, R.; Brentnall, T. A. Proteomics on Fixed Tissue Specimens - A Review. Curr. Proteomics 2009, 6 (1), 63–9. (7) Kawamura, T.; Nomura, M.; Tojo, H.; Fujii, K.; Hamasaki, H.; Mikami, S.; Bando, Y.; Kato, H.; Nishimura, T. Proteomic analysis of laser-microdissected paraffin-embedded tissues: (1) Stage-related protein candidates upon non-metastatic lung adenocarcinoma. J. Proteomics 2010, 73 (6), 1089–99. (8) Patel, V.; Hood, B. L.; Molinolo, A. A.; Lee, N. H.; Conrads, T. P.; Braisted, J. C.; Krizman, D. B.; Veenstra, T. D.; Gutkind, J. S. Proteomic analysis of laser-captured paraffin-embedded tissues: a molecular portrait of head and neck cancer progression. Clin. Cancer Res. 2008, 14 (4), 1002–14. (9) Negishi, A.; Masuda, M.; Ono, M.; Honda, K.; Shitashige, M.; Satow, R.; Sakuma, T.; Kuwabara, H.; Nakanishi, Y.; Kanai, Y.; Omura, K.; Hirohashi, S.; Yamada, T. Quantitative proteomics using formalinfixed paraffin-embedded tissues of oral squamous cell carcinoma. Cancer Sci. 2009, 100 (9), 1605–11. (10) Nagaraj, N.; Lu, A.; Mann, M.; Wisniewski, J. R. Detergentbased but gel-free method allows identification of several hundred membrane proteins in single LCMS runs. J. Proteome Res. 2008, 7 (11), 5028–32. (11) Wisniewski, J. R.; Zougman, A.; Nagaraj, N.; Mann, M. Universal sample preparation method for proteome analysis. Nat. Methods 2009, 6 (5), 359–62. (12) Wisniewski, J. R.; Zougman, A.; Mann, M. Combination of FASP and StageTip-based Fractionation Allows In-Depth Analysis of the Hippocampal Membrane Proteome. J. Proteome Res. 2009, 8 (12), 5674–8. (13) Zielinska, D. F.; Gnad, F.; Jedrusik-Bode, M.; Wisniewski, J. R.; Mann, M. Caenorhabditis elegans has a phosphoproteome atypical for metazoans that is enriched in developmental and sex determination proteins. J. Proteome Res. 2009, 8 (8), 4039–49. (14) Zielinska, D. F.; Gnad, F.; Wisniewski, J. R.; Mann, M. Precision mapping of an in vivo N-glycoproteome reveals rigid topological and sequence constraints. Cell 2010, 141 (5), 897–907. (15) Wisniewski, J. R.; Nagaraj, N.; Zougman, A.; Gnad, F.; Mann, M. Brain phosphoproteome obtained by a FASP-based method reveals plasma membrane protein topology. J. Proteome Res. 2010, 9 (6), 3280–9. (16) Ostasiewicz, P.; Zielinska, D. F.; Mann, M.; Wisniewski, J. R. Proteome, phosphoproteome, and N-glycoproteome are quantitatively preserved in formalin-fixed paraffin-embedded tissue and analyzable by high-resolution mass spectrometry. J. Proteome Res. 2010, 9 (7), 3688–700. (17) Weekes, M. P.; Antrobus, R.; Lill, J. R.; Duncan, L. M.; Hor, S.; Lehner, P. J. Comparative analysis of techniques to purify plasma membrane proteins. J. Biomol. Tech. 2010, 21 (3), 108–15. (18) Geiger, T.; Cox, J.; Ostasiewicz, P.; Wisniewski, J. R.; Mann, M. Super-SILAC mix for quantitative proteomics of human tumor tissue. Nat. Methods 2010, 7 (5), 383–5. (19) Liebler, D. C.; Ham, A. J. Spin filter-based sample preparation for shotgun proteomics. Nat. Methods 2009, 6 (11), 785;author reply 7856. (20) Wisniewski, J. R.; Mann, M. Spin filter-based sample preparation for shotgun proteomics Reply. Nat. Methods 2009, 6 (11), 785–786. (21) Nielsen, P. A.; Olsen, J. V.; Podtelejnikov, A. V.; Andersen, J. R.; Mann, M.; Wisniewski, J. R. Proteomic mapping of brain plasma membrane proteins. Mol. Cell. Proteomics 2005, 4 (4), 402–8.

ARTICLE

(22) Shevchenko, A.; Wilm, M.; Vorm, O.; Jensen, O. N.; Podtelejnikov, A. V.; Neubauer, G.; Mortensen, P.; Mann, M. A strategy for identifying gel-separated proteins in sequence databases by MS alone. Biochem. Soc. Trans. 1996, 24 (3), 893–6. (23) Olsen, J. V.; Macek, B.; Lange, O.; Makarov, A.; Horning, S.; Mann, M. Higher-energy C-trap dissociation for peptide modification analysis. Nat. Methods 2007, 4 (9), 709–12. (24) Cox, J.; Mann, M. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat. Biotechnol. 2008, 26 (12), 1367–72. (25) Luber, C. A.; Cox, J.; Lauterbach, H.; Fancke, B.; Selbach, M.; Tschopp, J.; Akira, S.; Wiegand, M.; Hochrein, H.; O’Keeffe, M.; Mann, M. Quantitative proteomics reveals subset-specific viral recognition in dendritic cells. Immunity 2010, 32 (2), 279–89. (26) Shevchenko, A.; Wilm, M.; Vorm, O.; Mann, M. Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal. Chem. 1996, 68 (5), 850–8. (27) Cottingham, K. 1DE proves its worth again. J. Proteome Res. 2010, 9 (4), 1636. (28) Fang, Y.; Robinson, D. P.; Foster, L. J. Quantitative analysis of proteome coverage and recovery rates for upstream fractionation methods in proteomics. J. Proteome Res. 2010, 9 (4), 1902–12. (29) Piersma, S. R.; Fiedler, U.; Span, S.; Lingnau, A.; Pham, T. V.; Hoffmann, S.; Kubbutat, M. H.; Jimenez, C. R. Workflow comparison for label-free, quantitative secretome proteomics for cancer biomarker discovery: method evaluation, differential analysis, and verification in serum. J. Proteome Res. 2010, 9 (4), 1913–22. (30) Wisniewski, J. R.; Zielinska, D. F.; Mann, M. Comparison of Ultrafiltration Units for Proteomic and N-Glycoproteomic Analysis by the FASP Method. Anal. Biochem. 2011, 410, 307–309. (31) Katayama, H.; Tabata, T.; Ishihama, Y.; Sato, T.; Oda, Y.; Nagasu, T. Efficient in-gel digestion procedure using 5-cyclohexyl-1pentyl-beta-D-maltoside as an additive for gel-based membrane proteomics. Rapid Commun. Mass Spectrom. 2004, 18 (20), 2388–94. (32) Wisniewski, J. R. Tools for phospho- and glycoproteomics of plasma membranes. Amino Acids 2011 . (33) Nitiss, J. L. DNA topoisomerase II and its growing repertoire of biological functions. Nat. Rev. Cancer 2009, 9 (5), 327–37. (34) Malumbres, M.; Barbacid, M. Cell cycle, CDKs and cancer: a changing paradigm. Nat. Rev. Cancer 2009, 9 (3), 153–66. (35) Yuan, B.; Xu, Y.; Woo, J. H.; Wang, Y.; Bae, Y. K.; Yoon, D. S.; Wersto, R. P.; Tully, E.; Wilsbach, K.; Gabrielson, E. Increased expression of mitotic checkpoint genes in breast cancer cells with chromosomal instability. Clin. Cancer Res. 2006, 12 (2), 405–10. (36) Cahill, D. P.; Lengauer, C.; Yu, J.; Riggins, G. J.; Willson, J. K.; Markowitz, S. D.; Kinzler, K. W.; Vogelstein, B. Mutations of mitotic checkpoint genes in human cancers. Nature 1998, 392 (6673), 300–3. (37) Li, F.; Zhang, D.; Fujise, K. Characterization of fortilin, a novel antiapoptotic protein. J. Biol. Chem. 2001, 276 (50), 47542–9. (38) Ziolkowski, P.; Gamian, E.; Osiecka, B.; Zougman, A.; Wisniewski, J. R. Immunohistochemical and proteomic evaluation of nuclear ubiquitous casein and cyclin-dependent kinases substrate in invasive ductal carcinoma of the breast. J. Biomed. Biotechnol. 2009, 2009, 919645. (39) Goldstein, N. S.; Bosler, D. S., Immunochemistry of the gastrointestimal tract, pancreas, bile ducts, gallbadder and liver. In Diagnostic Immunochemistry; Dabs, D. J., Ed.; Elsevier: New York, 2006. (40) Goldstein, M. J.; Mitchell, E. P. Carcinoembryonic antigen in the staging and follow-up of patients with colorectal cancer. Cancer Invest. 2005, 23 (4), 338–51. (41) Andersen, C. L.; Christensen, L. L.; Thorsen, K.; Schepeler, T.; Sorensen, F. B.; Verspaget, H. W.; Simon, R.; Kruhoffer, M.; Aaltonen, L. A.; Laurberg, S.; Orntoft, T. F. Dysregulation of the transcription factors SOX4, CBFB and SMARCC1 correlates with outcome of colorectal cancer. Br. J. Cancer 2009, 100 (3), 511–23. (42) Li, A.; Goto, M.; Horinouchi, M.; Tanaka, S.; Imai, K.; Kim, Y. S.; Sato, E.; Yonezawa, S. Expression of MUC1 and MUC2 mucins 3048


Journal of Proteome Research and relationship with cell proliferative activity in human colorectal neoplasia. Pathol. Int. 2001, 51 (11), 853–60. (43) Jimenez, C. R.; Knol, J. C.; Meijer, G. A.; Fijneman, R. J. Proteomics of colorectal cancer: overview of discovery studies and identification of commonly identified cancer-associated proteins and candidate CRC serum markers. J. Proteomics 2010, 73 (10), 1873–95. (44) Lee, S.; Bang, S.; Song, K.; Lee, I. Differential expression in normal-adenoma-carcinoma sequence suggests complex molecular carcinogenesis in colon. Oncol. Rep. 2006, 16 (4), 747–54. (45) Killian, A.; Sarafan-Vasseur, N.; Sesboue, R.; Le Pessot, F.; Blanchard, F.; Lamy, A.; Laurent, M.; Flaman, J. M.; Frebourg, T. Contribution of the BOP1 gene, located on 8q24, to colorectal tumorigenesis. Genes Chromosomes Cancer 2006, 45 (9), 874–81. (46) Palmerini, F.; Devilard, E.; Jarry, A.; Birg, F.; Xerri, L. Caspase 7 downregulation as an immunohistochemical marker of colonic carcinoma. Hum. Pathol. 2001, 32 (5), 461–7. (47) Mikesch, J. H.; Buerger, H.; Simon, R.; Brandt, B. Decayaccelerating factor (CD55): a versatile acting molecule in human malignancies. Biochim. Biophys. Acta 2006, 1766 (1), 42–52. (48) Helman, L. J.; Gazdar, A. F.; Park, J. G.; Cohen, P. S.; Cotelingam, J. D.; Israel, M. A. Chromogranin A expression in normal and malignant human tissues. J. Clin. Invest. 1988, 82 (2), 686–90. (49) Bustin, S. A.; Li, S. R.; Dorudi, S. Expression of the Ca2þactivated chloride channel genes CLCA1 and CLCA2 is downregulated in human colorectal cancer. DNA Cell Biol. 2001, 20 (6), 331–8. (50) Toiyama, Y.; Inoue, Y.; Yasuda, H.; Saigusa, S.; Yokoe, T.; Okugawa, Y.; Tanaka, K.; Miki, C.; Kusunoki, M. DPEP1, expressed in the early stages of colon carcinogenesis, affects cancer cell invasiveness. J. Gastroenterol. 2011, 46, 153–63. (51) Yao, X.; Zhao, G.; Yang, H.; Hong, X.; Bie, L.; Liu, G. Overexpression of high-mobility group box 1 correlates with tumor progression and poor prognosis in human colorectal carcinoma. J. Cancer Res. Clin. Oncol. 2010, 136 (5), 677–84. (52) Pyke, C.; Salo, S.; Ralfkiaer, E.; Romer, J.; Dano, K.; Tryggvason, K. Laminin-5 is a marker of invading cancer cells in some human carcinomas and is coexpressed with the receptor for urokinase plasminogen activator in budding cancer cells in colon adenocarcinomas. Cancer Res. 1995, 55 (18), 4132–9. (53) Stein, U.; Walther, W.; Arlt, F.; Schwabe, H.; Smith, J.; Fichtner, I.; Birchmeier, W.; Schlag, P. M. MACC1, a newly identified key regulator of HGF-MET signaling, predicts colon cancer metastasis. Nat. Med. 2009, 15 (1), 59–67. (54) Li, G. Q.; Zhang, H. F. Mad2 and p27 expression profiles in colorectal cancer and its clinical significance. World J. Gastroenterol. 2004, 10 (21), 3218–20. (55) Lau, K. S.; Dennis, J. W. N-Glycans in cancer progression. Glycobiology 2008, 18 (10), 750–60. (56) Hinoshita, E.; Uchiumi, T.; Taguchi, K.; Kinukawa, N.; Tsuneyoshi, M.; Maehara, Y.; Sugimachi, K.; Kuwano, M. Increased expression of an ATP-binding cassette superfamily transporter, multidrug resistance protein 2, in human colorectal carcinomas. Clin. Cancer Res. 2000, 6 (6), 2401–7. (57) Ogata, S.; Uehara, H.; Chen, A.; Itzkowitz, S. H. Mucin gene expression in colonic tissues and cell lines. Cancer Res. 1992, 52 (21), 5971–8. (58) Yung, B. Y. Oncogenic role of nucleophosmin/B23. Chang Gung Med. J. 2007, 30 (4), 285–93. (59) Nozawa, Y.; Van Belzen, N.; Van der Made, A. C.; Dinjens, W. N.; Bosman, F. T. Expression of nucleophosmin/B23 in normal and neoplastic colorectal mucosa. J. Pathol. 1996, 178 (1), 48–52. (60) Conrotto, P.; Roesli, C.; Rybak, J.; Kischel, P.; Waltregny, D.; Neri, D.; Castronovo, V. Identification of new accessible tumor antigens in human colon cancer by ex vivo protein biotinylation and comparative mass spectrometry analysis. Int. J. Cancer 2008, 123 (12), 2856–64. (61) Vie, N.; Copois, V.; Bascoul-Mollevi, C.; Denis, V.; Bec, N.; Robert, B.; Fraslon, C.; Conseiller, E.; Molina, F.; Larroque, C.; Martineau, P.; Del Rio, M.; Gongora, C. Overexpression of phosphoserine aminotransferase PSAT1 stimulates cell growth and increases chemoresistance of colon cancer cells. Mol. Cancer 2008, 7, 14.

ARTICLE

(62) Bergeron, S.; Lemieux, E.; Durand, V.; Cagnol, S.; Carrier, J. C.; Lussier, J. G.; Boucher, M. J.; Rivard, N. The serine protease inhibitor serpinE2 is a novel target of ERK signaling involved in human colorectal tumorigenesis. Mol. Cancer 2010, 9, 271. (63) Zheng, H.; Tsuneyama, K.; Cheng, C.; Takahashi, H.; Cui, Z.; Murai, Y.; Nomoto, K.; Takano, Y. Maspin expression was involved in colorectal adenoma-adenocarcinoma sequence and liver metastasis of tumors. Anticancer Res. 2007, 27 (1A), 259–65. (64) Lu, B.; Fang, Y.; Xu, J.; Wang, L.; Xu, F.; Xu, E.; Huang, Q.; Lai, M. Analysis of SOX9 expression in colorectal cancer. Am. J. Clin. Pathol. 2008, 130 (6), 897–904. (65) Powis, G.; Mustacich, D.; Coon, A. The role of the redox protein thioredoxin in cell growth and cancer. Free Radic. Biol. Med. 2000, 29 (34), 312–22. (66) Dusseau, C.; Murray, G. I.; Keenan, R. A.; O’Kelly, T.; Krokan, H. E.; McLeod, H. L. Analysis of uracil DNA glycosylase in human colorectal cancer. Int. J. Oncol. 2001, 18 (2), 393–9.

3049


High Recovery FASP Applied to the Proteomic ... - ACS Publications

Recommend Documents