Quantitative Analysis of N-Linked Glycoproteins in Tear Fluid of

Feb 2, 2009 - We have sampled the tears, the extracellular fluid of the epithelial cells covering the surface of the eye, of patients with climatic dr...
0 downloads 0 Views 3MB Size
Download PDF
Quantitative Analysis of N-Linked Glycoproteins in Tear Fluid of Climatic Droplet Keratopathy by Glycopeptide Capture and iTRAQ† Lei Zhou,‡,§ Roger W. Beuerman,*,‡,§ Ai Ping Chew,| Siew Kwan Koh,‡ Thamara A. Cafaro,⊥ Enrique A. Urrets-Zavalia,#,¶ Julio A. Urrets-Zavalia,#,¶ Sam F. Y. Li,| and Horacio M. Serra⊥ Singapore Eye Research Institute, Singapore, Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Department of Chemistry, National University of Singapore, Singapore, CIBICI, Faculty of Chemistry, National University of Co´rdoba, Argentina, Department of Ophthalmology, Clı´nica Universitaria Reina Fabiola, Argentina, Ophthalmology, Catholic University of Cordoba, Argentina Received November 6, 2008

Glycoproteins are potentially important biomarkers of disease and therapeutic targets. In particular, the N-linked glycoproteins are a focus of interest as they can be found in the extracellular environment and body fluids. In this study, we have sampled the tears, the extracellular fluid of the epithelial cells covering the surface of the eye, of patients with climatic droplet keratopathy (CDK) using tears of unaffected normal patients for comparison. Prefractionation of the tear sample used a hydrazide-resin capture method, and the previously N-glycosylated peptides were then subjected to two-dimensional nano-LC-nano-ESI-MS/MS analysis to obtain peptide fragmentation patterns for identification through protein database searches. We have identified a total of 43 unique N-glycoproteins, 19 of which have not previously been reported in tear fluid. In addition, we have quantitatively compared N-glycoprotein profiles in tear fluid of patients with CDK to tears of nondiseased controls using glycopeptide capture, iTRAQ labeling and 2D nano-LC-nano-ESI-MS/MS analysis. In tears of CDK patients, increased levels of four N-glycosylated proteins including haptoglobin (at sites N207, N211 and N241), polymeric immunoglobulin receptor (at sites N83, N90, N135, N186, N421, and N469), immunoglobulin J chain (at site N49) and an uncharacterized protein DKFZp686M08189 (at site N470), as well as a decrease in the N-glycosylation level of one N-glycosylated protein, lacritin (at site N119) were observed. However, the overall levels of these five proteins showed no appreciable changes between control and CDK samples. The findings could be clinically significant in terms of disease etiology and biomarkers. Keywords: glycoproteomics • glycosylation • tear proteomics • iTRAQ • quantitative proteomics • climatic droplet keratopathy

Introduction The tears covering the surface of the eye are an important component of the extracellular environment of the surface epithelial cell layer. Secretory and plasma contributions result in a fluid containing lipids, carbohydrates, proteins, and electrolytes which can affect the activity of the cells of the ocular surface.1 The composition is complex, both in the variety of proteins present as well as the dynamic concentration range which can span up to 10 orders of magnitude.1,2 Analysis of † Originally submitted and accepted as part of the “Glycoproteomics” special section, published in the February 2009 issue of J. Proteome Res. (Vol. 8, No. 2). * To whom correspondence should be addressed. Roger W. Beuerman, Singapore Eye Research Institute, 11 Third Hospital Avenue, Singapore 168751. E-mail: [email protected]. ‡ Singapore Eye Research Institute. § Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore. | Department of Chemistry, National University of Singapore. ⊥ National University of Co´rdoba. # Clı´nica Universitaria Reina Fabiola. ¶ Catholic University of Cordoba.

1992 Journal of Proteome Research 2009, 8, 1992–2003 Published on Web 02/02/2009

tears is challenging due to the small sample size as the amount of tears collected for a single sample is only about 5 µL,3 which is minute compared to other body fluids such as serum and saliva. Li et al.4 have attempted to characterize part of the tear proteome, using a combination of LC-ESI, LC-MALDI MS, and MS/MS to identify a total of 54 proteins present in tears. In that paper, they also reported several tear proteins which are glycosylated, such as lacritin and proline rich protein 1.4 In a more recent study, de Souza et al.5 have identified a total of 491 tear fluid proteins using LTQ-FT and LTQ-Orbitrap mass spectrometers. An et al. reported a glycomics study in human tears from ocular rosacea patients.6 Despite many studies on tear proteomics,4-8 there is no published study depicting the profile of N-linked glycoproteins in human tear fluid. Proteins are known to undergo post-translational modification (PTM), which may change protein structure and function. Commonly characterized PTMs are phosphorylation, ubiquitination, and glycosylation.9 Two of the most common protein glycosylation patterns are O-linked and N-linked glycosylation. N-glycosylated proteins in particular have been the focus of 10.1021/pr800962q CCC: $40.75

 2009 American Chemical Society

research articles

Quantification of N-Linked Glycoproteins in Tears

Figure 1. Strategy for quantitative analysis of N-glycoproteins in tear fluid using glycoprotein capture (hydrazide resin), iTRAQ labeling and nanoLC-MS/MS. Control group was labeled with iTRAQ reagent 116, while diseased group was labeled with iTRAQ reagent 117.

many proteomics studies.10-12 This is due in part to the fact that most N-linked glycoproteins are secreted.13 For example, glycoproteins can be found in body fluids such as plasma,14,15 urine,16 cerospinal fluid,17 and saliva.18 N-linked glycosylation occurs at the amide nitrogen of asparagine. The consensus motif for N-linked glycosylation is Asn-X-Thr/Ser (three-letter amino acid code), where X represents any amino acid except proline, though the less common motif Asn-X-Cys may also be found. These motifs allow the confirmation of the presence of an N-glycosylation site when analyzing peptide matches obtained from searches through protein databases. Therefore, the glycosylated proteins are of interest for their potential use as biomarkers and therapeutic targets. Some glycoprotein biomarkers currently in use include Prostate Specific Antigen for prostate cancer, CA125 for ovarian cancer, and c-erbB-2 for breast cancer.19 In fact, one of the targets in the immunotherapy of breast cancer is the Her2/neu glycoprotein receptor.20 It has also been established that there is a correlation between a difference in glycosylation patterns of proteins and the onset of cancer metastasis.21 As proteins can be involved in multiple biochemical pathways, it is believed that glycoproteins could play important roles in the pathology of other diseases as well. We have adopted the protocol first developed by Zhang et al.10 and subsequently used by other groups.14,17,18 A hydrazidefunctionalized resin was used to select for glycoproteins through the stable covalent hydrazone bonds formed between oxidized glycans and hydrazine functional groups. As nonglycoproteins exhibit only nonspecific binding to the resin, they can be easily washed away. Subsequently, N-linked glycoproteins were selectively cleaved from the resin using peptide-Nglycosidase F (PNGase F), an enzyme which specifically deglycosylates N-glycoproteins and not O-glycoproteins.22 Twodimensional nano-LC-nano-ESI-MS/MS was then used to analyze the peptides derived from formerly glycosylated Nlinked glycoproteins. Subsequent tagging methods, such as ICAT and iTRAQ reagents, can be used to determine the changes of occupancy of N-glycosylation at specific sites.10 We have applied this method in the analysis of tear proteins from patients suffering from a cornea disease known as climatic droplet keratopathy (CDK). This disease involves the spheroidal degeneration of the cornea, and yellowish deposits can be observed in the superficial corneal region.23 One recent study showed that these deposits contained the aggregation of

advanced glycation end products (AGEs), which are the final products of the reaction between sugars and proteins.24 AGEs take a long time to form, as enzymes are not involved in their aggregation to form the insoluble products. In another study, analysis of these droplets revealed that they are composed of at least 105 proteins, two of which were more highly expressed compared to the nondiseased condition.25 As CDK is a disease of the ocular surface, which is covered by tear fluid, the tear protein profile could potentially be informative for understanding the pathology of CDK. To quantify the relative difference in tear protein N-glycosylation levels between the diseased and nondiseased states, we have used iTRAQ amine-modifying isobaric tags to label the formerly glycosylated peptides and subject them to nanoLC-MS/MS analysis. Meanwhile, unfractionated tear samples from both control and CDK disease group were tryptic digested, labeled using iTRAQ reagents, and analyzed by nanoLC-MS/MS to quantitatively compare the overall levels of each identified tear protein. The overall workflow is shown in Figure 1.

Materials and Methods Chemicals and Enzymes. Acetonitrile (ACN), methanol (MeOH), and water (HPLC grade) were obtained from Fisher Scientific (Pittsburgh, PA). Ethanol (100%, HPLC grade), formic acid, hydrochloric acid, and sodium chloride were purchased from Merck (Darmstadt, Germany). Acetic acid, trifluoroacetic acid (TFA), ammonium acetate, ammonium bicarbonate, ethylenediaminetetraacetic acid (EDTA), formic acid, glycerol, sodium dodecyl sulfate (SDS), and urea were from Sigma (St. Louis, MO). 2-Amino-2-(hydroxymethyl)propane-1,3-diol (Tris), Affi-gel Oxidizer (sodium periodate), Affi-gel Hz Hydrazide Gel, and 10× Coupling Buffer were purchased from Bio-Rad Laboratories (Hercules, CA). Tris(2-carboxyethyl)phosphine (TCEP) was obtained from Pierce (Rockford, IL). Sequencing-grade trypsin was purchased from EMD Chemicals (Gibbstown, NJ, USA), while glycerol-free Peptide N-glycosidase F (PNGase F) was purchased from New England Biolabs (Ipswich, MA). The iTRAQ Reagents Multiplex Kit was obtained from Applied Biosystems (Foster City, CA). Patients and Tear Samples. Patients (n ) 13; average age, 72, all males) diagnosed as Climatic Droplet Keratopathy (CDK) and 11 control subjects (average age, 63; 9 males and 2 females) with clinically normal eyes and no history of eye disease or eye surgery were recruited for this study. The study was Journal of Proteome Research • Vol. 8, No. 4, 2009 1993

research articles

Zhou et al. a

Table 1. N-Linked Glycoproteins and Representative Formerly Glycosylated Peptides Identified Using MASCOT

no.

b

1

IPI no. IPI00004573

protein name Polymeric immunoglobulin receptor

2

IPI00012503

Isoform Sap-mu-0 of Proactivator polypeptide

3 4

IPI00020091 IPI00020487

5

IPI00021891

6 7

IPI00022429 IPI00022463

Alpha-1-acid glycoprotein 2 Extracellular glycoprotein lacritin Isoform Gamma-B of Fibrinogen gamma chain Alpha-1-acid glycoprotein 1 Serotransferrin

8

IPI00022488

Hemopexin

9 10

IPI00022974 IPI00023673

Prolactin-inducible protein Galectin-3-binding protein

11 12 13

IPI00026126 IPI00032328 IPI00166729

Mammaglobin-B Isoform HMW of KininogenAlpha-2-glycoprotein 1, zinc

14 15

IPI00178926 IPI00217778

Immunoglobulin J chain Isoform 2 of Phospholipid transfer protein

16

IPI00291262

Clusterin

17

IPI00298828

Beta-2-glycoprotein 1

18

IPI00298860

Lactoferrin

19

IPI00382606

20 21

IPI00383164 IPI00386524

22

IPI00418512

23

IPI00430842

Factor VII active site mutant immunoconjugate SNC66 protein CDNA FLJ25298 fis, clone STM07683, highly similar to Protein Tro alpha1 H, myeloma Isoform 4 of Deleted in malignant brain tumors 1 protein IGHA1 protein

24 25

IPI00550991 IPI00641737

Alpha-1-antichymotrypsin Haptoglobin

26

IPI00759659

27

IPI00784817

Isoform 2 of Golgi membrane protein 1 Anti-RhD monoclonal T125 gamma1 heavy chain

28

IPI00785067

29 30

IPI00788271 IPI00789477

31

IPI00790784

32

IPI00829767

Putative uncharacterized protein, IGH@ PROTEIN

Similar to lactotransferrin 73 kDa protein or Truncated lactoferrin

Isoform 2 of Alpha-1-antitrypsin Uncharacterized protein IGHG2 (Fragment)

error (Da)

Mascot score

formerly glycosylated peptide

Swiss-Prot N site identification, site no.c

m/z

z

842.72

3

-0.15

111

K.QIGLYPVLVIDSSGYVNPN#YTGR.I

Y, N186

699.85 937.06 801.71 928.41

2 3 3 3

-0.10 -0.16 -0.15 -0.14

70 94 117 59

K.VPGN#VTAVLGETLK.V R.ANLTNFPEN#GTFVVNIAQLSQDDSGR.Y R.LSLLEEPGN#GTFTVILNQLTSR.D K.DVVTAAGDMLKDN#ATEEEILVYLEK.T

Y, N469 Y, N90 Y, N421 Y, N80

525.25 966.15 699.83

3 3 2

-0.10 0.01 -0.00

32 74 76

R.TN#STFVQALVEHVK.E K.SVQEIQATFFYFTPN#KTEDTIFLR.E K.QFIEN#GSEFAQK.L

Y, N215 Y, N72 Y, N119

594.60

3

-0.09

65

K.DLQSLEDILHQVEN#K.T

Y, N78

966.15 739.33 839.32 868.87 703.29 707.32 674.84 835.41 452.75 699.36 1034.42 570.18 717.05 496.23

3 2 3 2 2 2 2 2 4 4 2 2 3 3

0.01 -0.09 -0.15 -0.13 -0.08 0.02 -0.01 -0.03 0.07 0.09 -0.10 -0.12 -0.04 -0.02

74 35 109 74 56 55 102 70 38 64 122 46 46 31

K.SVQEIQATFFYFTPN#KTEDTIFLR.E K.CGLVPVLAENYN#K.S R.QQQHLFGSN#VTDCSGNFCLFR.S K.ALPQPQN#VTSLLGCTH.R.SWPAVGN#CSSALR.W K.TFYWDFYTN#R.T R.ALGFEN#ATQALGR.A R.TVIRPFYLTN#SSGVD.K.FKQCFLN#QSHRTLK.N R.HGIQYFNN#NTQHSSLFMLNEVKR.A K.DIVEYYN#DSN#GSHVLQGR.F R.FGCEIENN#R.S R.IIVPLNNREN#ISDPTSPLR.T K.EGHFYYN#ISEVK.V

Y, N72 Y, N432 Y, N630 Y, N453 Y, N187 Y, N105 Y, N69 Y, N580 Y, N68 Y, N169 Y, N106,109 Y, N125 Y, N49 Y, N64

760.39 809.40 842.86 843.32 490.57 899.39 302.13 791.72 595.69

3 3 2 4 3 4 1 3 2

-0.11 0.03 -0.11 -0.17 -0.04 -0.04 -0.02960 -0.00190 -0.12

44 53 106 35 64 73 24 22 43

R.IYSN#HSALESLALIPLQAPLK.T K.MLN#TSSLLEQLNEQFNWVSR.L R.LAN#LTQGEDQYYLR.V R.DTAVFECLPQHAMFGN#DTITCTTHGN#WTK.L R.VYKPSAGN#NSLYR.D R.N#GSDCPDKFCLFQSETKNLLFNDNTECLAR.L PFLN#WTGPPEPIEAAVAR TAGWNIPMGLLFN#QTGSCK R.EEQYN#STYR.V

Y, N398 Y, N354 Y, N374 Y, N183, 193 Y, 162 N, N642 Y, 156 Y, 497 N, N529

741.90 741.90

4 4

0.00 0.00

87 87

R.LSLHRPALEDLLLGSEAN#LTCTLTGLR.D R.LSLHRPALEDLLLGSEAN#LTCTLTGLR.D

N, N288 N, N287

767.62

4

0.01

39

K.LEAHHN#CSFDYVEIFDGSLN#SSLLLGK.I

Y, N1818

741.90 881.13 601.03 674.83 599.28 644.32

4 3 4 4 3 2

0.00 0.09689 -0.07 -0.06 -0.17 -0.06

87 17 66 85 49 69

R.LSLHRPALEDLLLGSEAN#LTCTLTGLR.D LAGKPTHVN#VSVVMAEVDGTCY K.FN#LTETSEAEIHQSFQHLLR.T K.MVSHHN#LTTGATLINEQWLLTTAK.N K.VVLHPN#YSQVDIGLIK.L K.AVLVNN#ITTGER.L

Y, N144 Y, N340 Y, N106 Y, N184 Y, N241 Y, N09

558.26

3

-0.03

30

K.TKPREEQYN#STYR.V

N, N325

595.69 480.23

2 2

-0.12 -0.09

43 51

R.EEQYN#STYR.V K.TPLTAN#ITK.S

N, N325 N, N345

788.69 988.85 599.25 753.32

3 3 4 3

-0.05 -0.05 -0.08 -0.12

45 95 40 60

R.LAGKPTHVN#VSVVMAEVDGTCY.R.LSLHRPALEDLLLGSEAN#LTCTLTGLR.D K.FGRN#GSDCPDKFCLFQSETK.N R.FFSASCVPGADKGQFPN#LCR.L

N, N467 N, N271 N, N642 N, N187

713.57 678.59 699.34 1077.87 923.94

4 3 3 3 4

0.05 -0.11 0.00 -0.09 -0.06

20 66 87 76 71

R.LCAGTGEN#KCAFSSQEPYFSYSGAFK.C R.N#GSDCPDKFCLFQSETK.N R.TAGWNIPMGLLFN#QTGSCK.F R.TAGWNVPIGTLRPFLN#WTGPPEPIEAAVAR.F K.ADTHDEILEGLNFN#LTEIPEAQIHEGFQELLR.T

N, N199 N, N642 Y, N497 Y, N156 Y, N107

579.71

2

-0.09

43

R.EEQFN#STFR.V

Y, N176

a Number of proteins ) 32, number of peptides ) 54. N-glycosylation motifs are in bold and formerly glycosylated asparagine residues are shown as N#. b Numbers highlighted in bold are those N-glycoproteins that have not previously been reported in tear fluid. c Y, found in Swiss-Prot database; N, not found in Swiss-Prot database.

1994

Journal of Proteome Research • Vol. 8, No. 4, 2009

research articles

Quantification of N-Linked Glycoproteins in Tears

Table 2. Additional N-Linked Glycoproteins and Formerly Glycosylated Peptides Identified from iTRAQ Experimentsa

no.

1

b

accession

name

m/z

z

IPI00006114 Pigment 906.8152 3 epithelium-derived factor 2 IPI00022431 Alpha-2-HS-glycoprotein 837.4358 3 precursor 3 IPI00219825 PSAP 62 kDa protein, 804.6706 4 Prosaposin (Variant Gaucher disease and variant metachromatic leukodystrophy) 621.3466 3 4 IPI00400826 Clusterin isoform 1 or 914.9675 2 cDNA, FLJ94503, highly similar to Homo sapiens clusterin (complement lysis inhibitor, SP-40,40,sulfated glycoprotein 2, testosterone-repressed prostate message 2,apolipoprotein J) (CLU), mRNA 857.4370 3 5 IPI00472345 IGHG3 protein 651.8079 2 6 IPI00478003 Alpha-2-macroglobulin 770.1015 3 precursor 7 IPI00553177 Isoform 1 of 959.9837 4 Alpha-1-antitrypsin precursor 8 IPI00643623 LCN2 Lipocalin 2 or 665.8551 2 Uncharacterized protein LCN2 9 IPI00783987 Complement C3 806.0934 3 (Fragment) 10 IPI00784758 Gene_Symbol) 552.3154 2 LOC100126583 Putative uncharacterized protein DKFZp686M08189 11 IPI00877792 FGG 50 kDa protein, 690.7000 3 fibrinogen gamma chain

error (Da)

ProteinPilot score

formerly glycosylated peptide

Swiss-Prot N site identification, site no.c

0.03569

23

VTQN#LTLIEESLTSEFIHDIDR

Y, N285

0.05370

24

AALAAFNAQNN#GSNFQLEEISR

Y, N176

0.03940

23

DVVTAAGDMLKDN#ATEEEILVYLEK

Y, N80

-0.00098 0.00804

21 25

TN#STFVQALVEHVK LAN#LTQGEDQYYLR

Y, N215 Y, N374

0.02878 0.00015 0.02306

25 15 23

MLN#TSSLLEQLNEQFNWVSR EEQFN#STFR VSN#QTLSLFFTVLQDVPVR

Y, N354 N, N371 Y, N1424

0.00320

30

ADTHDEILEGLNFN#LTEIPEAQIHEGFQELLR

Y,N107

-0.00940

18

SYN#VTSVLFR

Y, N85

0.02445

24

TVLTPATNHMGN#VTFTIPANR

Y, N85

-0.01931

16

TPLTAN#ITK

N, N348

-0.01030

28

DLQSLEDILHQVEN#K(T)

Y, N78

a Number of proteins ) 11, number of peptides ) 13. N-glycosylation sites are in bold and asparagine residues which were formerly glycosylated are N#. b Numbers highlighted in bold are those N-glycoproteins that have not previously been reported in tear fluid. c Y, found in Swiss-Prot database; N, not found in Swiss-Prot database.

approved by the institutional review board of Catholic University of Cordoba, and the Interinstitutional Committee of Ethics in Health Research, Ministry of Health of the Province of Cordoba, Argentina. All patients received counseling, and the tear collection procedure was explained in the patient consent form. Tear collection was performed using 10-µL pipets with fire-polished tips. The tip of the capillary tube was laid into the space between the globe and the lid, and tears flowed by capillary action into the pipet. Tear samples were spun at 8000 rpm to remove cells and frozen at -80 °C until analysis. Enrichment of Tear Samples for N-Linked Glycoproteins Using Hydrazide-Functionalized Resin. The overall workflow for the prefractionation of tear samples for N-linked glycoproteins can be found in Figure 1. Protein quantitation for each sample was performed using the RC DC Protein Assay (BioRad Laboratories), according to the Microfuge Tube Assay Protocol. Protein standards used in obtaining a standard curve were prepared using Bovine Serum Albumin (Pierce). An aliquot from each diluted sample, equivalent to 500 µg of protein, was withdrawn into a centrifuge tube for N-glycoprotein enrichment. Prior to glycoprotein enrichment, ethanol precipitation was performed to desalt the tear samples. Four volumes of cold ethanol was added to each sample, mixing well prior to overnight incubation at -20 °C. The sample was then centrifuged at 13 000 rpm for 15 min at 4 °C and the supernatant was removed. The pellet was resuspended in 1× Coupling

Buffer (dilution of 10× Coupling Buffer, adjusted to pH 5.5 using 1.0 N HCl). Next, the glycoproteins were oxidized by adding sodium periodate solution up to a final concentration of 15 mM and incubating in the dark for 1 h at room temperature. Excess periodate was quenched by adding 200 mM glycerol to a final concentration of 20 mM and mixing for 15 min at room temperature. The mixture was then subjected to ethanol precipitation as described earlier, and then resuspended in 1× Coupling Buffer. Equilibration of the hydrazide gel resin, supplied in isopropanol, was performed by washing it four times in 2 vol of 1× Coupling Buffer (100 mM sodium acetate, 150 mM sodium chloride, pH 5.5, diluted from 10× Coupling Buffer). Oxidized glycoproteins were coupled to hydrazide-functionalized resins by incubating overnight at room temperature. Uncoupled nonglycoproteins in the supernatant were removed, which was stored at -20 °C for further analysis. The resin with bound glycoproteins was then washed six times with Urea Buffer A (8 M urea, 200 mM Tris, 0.05% SDS, and 5 mM EDTA, pH 8.3) to remove nonglycoproteins. Trypsin digestion was then carried out on the bound glycoproteins. The glycoproteins were reduced using 10 mM TCEP in Urea Buffer A, followed by blocking of the sulfhydryl groups on cysteinyl residues with iodoacetamide Blocking Agent (ProteoExtract All-In-One Trypsin Digestion Kit, CalbioJournal of Proteome Research • Vol. 8, No. 4, 2009 1995

research articles

Zhou et al.

Figure 2. MS/MS spectrum of N-linked glycopeptide fragment LSLHRPALEDLLLGSEAN#LTCTLTGLR (m/z ) 741.9, charge ) 4+) originated from SNC66 protein. (#) Represents the N-glycosylation site and the consensus motif of N-glycosylation is in bold. The mass difference of 115 Da between y9 and y10 ion confirms the deamidation of asparagine in the peptide.

Figure 3. MS/MS spectrum of N-linked glycopeptide fragment AVLVNN#ITTGER (m/z ) 644.31, charge ) 2+) originated from Golgi membrane protein 1. (#) Represents the N-glycosylation site and the consensus motif of N-glycosylation is in bold. The mass difference of 115 Da between y6 and y7 ion confirms the deamidation of asparagine in the peptide, while the mass difference of 114 Da between y7 and y8 ion indicates that the asparagine residue is not deamidated, and therefore, it is not a N-glycosylation site.

chem, EMD Biosciences). After six washes with Urea Buffer B (1 M urea, 25 mM Tris, pH 8.3), the resin was resuspended in Urea Buffer B. Trypsin digestion of glycoproteins was carried out overnight at 37 °C with mixing. To remove nontryptic peptides, the resin was then washed six times with 1.5 M NaCl, 1996

Journal of Proteome Research • Vol. 8, No. 4, 2009

three times each with 80% ACN/0.1% TFA in water, methanol, HPLC water, and six times with 100 mM NH4CO3. After washing, the resin was resuspended in 100 mM NH4CO3 to obtain an approximate 50% gel slurry. PNGase-F was added to the mixture to deglycosylate N-linked glycoproteins by incubating

research articles

Quantification of N-Linked Glycoproteins in Tears

Table 3. Increased and Decreased N-Glycosylation Levels of Five N-Glycoproteins in CDK-Disease Tear Samples Compared to Controla

no.

accession

1

IPI00431645.1

2

IPI00004573.2

protein name

N-sites

identified sequences

Up-Regulated in CDK Tears Haptoglobin N207, N211 NLFLN#HSEN#ATAK N241 VVLHPN#YSQVDIGLI Polymeric immunoglobulin N83, N90 AN#LTNFPEN#GTFVVNIAQLSQDDSGR receptor precursor N135 GLSFDVSLEVSQGPGLLN#DTK N186 QIGLYPVLVIDSSGYVNPN#YTGR N421 LSLLEEPGN#GTFTVILNQLTSR N469 VPGN#VTAVLGETLK Immunoglobulin J chain N49 EN#ISDPTSPLR Putative uncharacterized N470 LAGKPTHVN#VSVV protein DKFZp686M08189

3 4

IPI00178926.2 IPI00784758.1

5

IPI00020487.1 Extracellular glycoprotein lacritin precursor

N119

Down-Regulated in CDK Tears QFIEN#GSEFAQ

ratio (CDK/control) without N-glycopeptide capture, overall ratio (CDK/control) levels of with N-glycopeptide these 5 capture proteins

1.517 1.632 1.892

1.084 1.180

2.103 2.051 1.823 1.561 1.834 1.774

1.107 ND

0.701

0.959

a CDK samples were labeled with iTRAQ Reagent 117, while the control sample was labeled with iTRAQ Reagent 116. Overall levels of each identified tear protein between control and CDK tear samples were also quantitative assessed using unfractionated tear samples by iTRAQ. ND: no data.

overnight with mixing at 37 °C. Deglycosylated N-linked glycoproteins present in the supernatant were transferred into a new microfuge tube. The resin was washed with 80% ACN (in water) and the washes were pooled into their respective tubes. Solvent was removed using a SpeedVac. N-Linked Glycoprotein Identification Using 2D Nano-LCnano-ESI-MS/MS. The nano-LC system (DIONEX, LC Packings, Sunnyvale, CA) was coupled with a nano-ESI-MS/MS (Applied Biosystems, Q-Star XL, MDS Sciex, Concord, Ontario, Canada), which had TOF-MS and MS/MS capabilities. The PicoFrit microcapillary column with integrated spray tip was from New Objective (Woburn, MA), and was directly coupled with the Q-TOF mass spectrometer through a NanoSpray interface (Protana, Odense, Denmark). This column was self-packed to about 10 cm using Luna C18, 3 µm, 100 Å from Phenomenex (Torrance, CA). Formerly glycosylated N-glycopeptides were dissolved in the LC loading buffer (0.1% formic acid/2% ACN). A Famos autosampler (DIONEX, LC Packings) loaded the samples onto the first dimension, which was a strong cation exchange column (300 µm i.d. × 10 cm, porosity 10S SCX, DIONEX, LC Packings), followed by a trapping cartridge (C18, 300 µm i.d. × 5 mm, DIONEX, LC Packings) for 7 min at a flow rate of 30 µL/min and using 5% ACN/0.1% formic acid (in water) as loading solvent. Ten steps of 20 µL-injection salt plug elutions were used (10, 20, 30, 40, 50, 75, 100, 250, 500, and 1000 mM ammonium acetate). The system was then switched (Switchos, DIONEX, LC Packings) in-line to the C18 microcapillary column, which is the second dimension used in the 2DLC analysis. A linear gradient of 0.1% formic acid (in ACN) from 5% to 60% over 135 min at a flow rate of ∼300 nL/min was delivered (UltiMate solvent delivery system, DIONEX, LC Packings). For the nano-ESI-MS/MS system, the following parameters were used: ion spray voltage, 2200 V; curtain gas, 20; declustering potential (DP), 80 V; DP2, 15 V; focusing potential, 265 V; collision gas setting, 5 for nitrogen gas. The InformationDependent Acquisition (IDA) mode for the Analyst QS software (version 1.1, Applied Biosystems) was used to acquire the mass spectrometry data. The TOF-MS survey scan parameters used

were as follows: 1 s TOF-MS survey scan was performed in the mass range of 300-1200 Da, after which two product ion scans, each of 3 s, were carried out in the mass range of 100-1800 Da. The switching criteria were set at ions with m/z greater than 350 and smaller than 1200, a charge state of 2-4, and an abundance threshold of 8 counts/s, while former target ions were excluded for a total of 120 s. The Mascot database search engine (Matrix Science, London, U.K.) was used in protein identification. In all searches, the Human IPI database was chosen (version 3.39). The variable modifications used were carbamidomethylation of cysteine, oxidation of methionine, and deamidation of asparagine to aspartic acid. Two missed tryptic cleavage sites were allowed, and the mass tolerances for precursor ions and fragment ions were (0.30 and (0.15 Da, respectively. Top-matched spectra with scores greater than the threshold of 37 (5% threshold) were generally considered to be true and were manually examined. Spectra of peptide matches with scores ranging from 30 to 37 were also manually examined to ensure no significant matches were omitted. Confirmation of an N-glycosylation site was done using the following two criteria: (a) the presence of the N-glycosylation motif (Asn-X-Ser/Thr/Cys), and (b) the deamidation of asparagine to form aspartic acid. Peptide Quantitation Using iTRAQ. Tear samples were selectively enriched for N-glycoproteins according to the procedure described in section 2.3. Total tear protein (500 µg) pooled from the control group (11 samples) or CDK group (13 samples) was used for the iTRAQ experiments. Methane methylthiosulfonate (MMTS, from iTRAQ Reagents Multiplex Kit) was used as the blocking agent for cysteine residues. The schematic for peptide quantitation is as shown in Figure 1. Control group peptides were labeled with iTRAQ reagent 116, while those from the diseased group were labeled with iTRAQ reagent 117. Labeling was performed in accordance with the iTRAQ protocol. Briefly, the respective iTRAQ reagents were quantitatively transferred into the tubes containing the peptides, which were incubated for 1 h with mixing at room temperature. The contents of these two tubes were then combined before solvent removal using the SpeedVac system. Journal of Proteome Research • Vol. 8, No. 4, 2009 1997

research articles

Zhou et al.

Figure 4. (A) MS/MS spectrum of VVLHPN#YSQVDIGLIK (m/z ) 695.3889, charge ) 3+, originated from haptoglobin). (B) Relative quantification for N-glycosylated VVLHPN#YSQVDIGLIK at site 241 between CDK tear samples and control tear samples using iTRAQ. Control group peptides were labeled with iTRAQ reagent 116, while those from the diseased group were labeled with iTRAQ reagent 117.

Subsequently, the labeled peptides were dissolved in the LC loading buffer and subject to 2D-nanoLC-nanoESI-MS/MS analysis under the conditions described in section 2.4. 1998

Journal of Proteome Research • Vol. 8, No. 4, 2009

In a separate experiment, the overall levels of each identified tear protein were analyzed using unfractionated tear samples from both control and CDK disease group (50 µg of total protein

research articles

Quantification of N-Linked Glycoproteins in Tears from each group) by tryptic digestion, labeling with iTRAQ reagents using the same iTRAQ protocol, and 2D-nanoLCnanoESI-MS/MS. Analysis of iTRAQ Data. ProteinPilot software (version 2.01, Applied Biosystems, MDS Sciex) was used to analyze the MS/ MS data obtained from the iTRAQ experiments. The data was searched against the IPI Human database (version 3.39), with a PSPEP setting of “Reversed Protein Sequences”. Quantitation of results was chosen with emphasis on biological modifications, and the threshold used for detected proteins and matched peptides was set at 1.3 (95%). Isotope correction factors for the iTRAQ labels were those supplied with the reagent kit. N-glycosylation was similarly confirmed by the presence of both the N-glycosylation motif as well as a deamidated asparaginyl residue. Spectra were also examined at the lower m/z regions to confirm the presence of reporter ions for the peptides used in quantitation. Statistical calculations by the Paragon Algorithm used in the ProteinPilot software were used to determine whether the difference in protein expression between control and diseased samples was statistically significant.

Results Identification of N-Linked Glycoproteins Using Hydrazide Resin Capture and 2D-nano-LC-nano-ESI-MS/MS. Previous studies indicated that the hydrazide resin could be used to effectively select for N-linked glycoproteins. In our study, a total of 67 formerly N-glycosylated peptides from 43 unique N-linked glycoproteins were identified. The lists of proteins identified representing formerly N-glycosylated peptides are given in Table 1 (Mascot search results) and Table 2 (additional identification using ProteinPilot software). The overall confidence level used for positive peptide identification and siteassignment was 95%. The spectra of identified peptides also exhibited consecutive ions (at least 4) from either the b- or y-ion series. All identified N-linked glycopeptides have the consensus motif Asn-X-Thr/Ser. We found 12 N-glycosylated sites which had no annotation in Swiss-Prot database (Tables 1 and 2). Figure 2 showed the MS/MS spectrum of a peptide fragment LSLHRPALEDLLLGSEAN#LTCTLTGLR (m/z ) 741.9, charge ) 4+) originated from SNC66 protein. A mass difference of 115 Da between the y9 and y10 ion corresponds to the mass of aspartic acid and, therefore, confirms the deamidation of asparagine in the peptide. This can also be seen in the MS/MS spectrum of the peptide AVLVNN#ITTGER (m/z ) 644.31, charge ) 2+, from Isoform 2 of Golgi membrane protein 1), where the mass difference between the y6 and y7 ion was also 115 Da (Figure 3). In the same spectrum, the mass difference between the y7 and y8 ion was only 114 Da, indicating that the asparagine residue was not deamidated, and therefore could not have been formerly glycosylated. Relative Quantitation of Tear Protein N-Glycosylation Levels and Their Association with Climatic Droplet Keratopathy. In this study, we quantitatively compared the profiles of tear protein N-glycosylation levels from both control group (pooled from 11 samples) and CDK group (pooled from 13 samples). After hydrazide-resin capture and PNGase F deglycosylation, formerly glycosylated peptides were labeled using iTRAQ reagents to determine the relative tear protein N-glycosylation levels in control (labeled with iTRAQ reagent 116) and CDKdiseased samples (labeled with iTRAQ reagent 117). Unfractionated tear samples from both control and CDK disease group were also quantitatively evaluated using iTRAQ to determine

the overall levels of each identified tear protein. A given peptide labeled with different iTRAQ reagents would be detected at the same m/z in the MS spectrum. When the labeled peptide undergoes MS/MS fragmentation, the reporter groups of different masses would show up in the lower m/z region (reporter ions) of the MS/MS spectrum with intensities corresponding to concentration of the original labeled peptide. This allows us to quantify the difference in protein expression levels between two, or even up to eight samples if using the latest 8-plex reagents. A total of 5 N-linked glycoproteins were found to have significant changes in N-glycosylation levels between control and CDK samples, as shown in Table 3. However, the overall levels of these five proteins showed no appreciable changes between control and CDK samples (Table 3). Haptoglobin (at sites N207, N211, and N241), polymeric immunoglobulin receptor (at sites N83, N90, N135, N186, N421, and N469), immunoglobulin J chain (at site N49), and a putative uncharacterized protein DKFZp686M08189 (at site N470) were found to be significantly increased in their levels of N-glycosylation in CDK samples. In contrast, extracellular glycoprotein lacritin (at site N119) was found to be decreased in the level of N-glycosylation in CDK samples. Figure 4A shows a representative MS/MS spectrum of formerly N-glycosylated peptide (VVLHPN#YSQVDIGLIK, m/z ) 695.3889, charge ) 3+,) originated from haptoglobin. For haptoglobin, which is significantly up-regulated (Figure 4B), the peak area for the 117-iTRAQ label is larger than that of the control label (iTRAQ 116).

Discussion Tear proteins are complex and heterogeneous. A simple 2D shot-gun proteomics may miss the information of posttranslational modification such as glycosylation. 2D gel separation26,27 of tear proteins showed that as many as 500 spots could be observed. However, only less than 100 distinct proteins were identified, largely because of the presence of different isoforms and post-translational modifications with similar molecular weights, but different pI values (isoelectric points). In this study, prefractionation of tear sample was performed using a hydrazide-resin capture method and the glycan moieties were removed by enzyme PNGase F, which are highly selective for N-linked glycoproteins. Quantitative analysis of glycoproteins was carried out using iTRAQ tagging to determine the relative expression levels of N-glycosylated proteins in control and CDK-diseased samples. The iTRAQ labeling was done on the deglycosylated peptides (Figure 1). Such labeling methods can be used to determine individual N-glycosylation site occupancy.10,28 Each MS/MS glycopeptide spectrum was manually inspected to ensure the presence of the consensus motif Asn-X-Thr/Ser. Deglycosylation with PNGase F converts asparagine to aspartic acid with a mass increase of 1 Da (Figure 2). Caution must be taken in adapting the methods for tear fluid analysis as the scale is much smaller than analysis of other body fluids such as urine and serum. The biggest challenge for tear analysis is the extremely small, microliter, sample volume. Sometimes it is necessary to pool samples to obtain enough starting materials as the situation in this study. Individual tear analysis is also possible in some cases.29-31 Pooling of the samples facilitates the initial identification of any changes between control and disease group; however, the individual variation information is missing. Journal of Proteome Research • Vol. 8, No. 4, 2009 1999

research articles

Zhou et al.

Figure 5. (A) MS/MS spectrum of nonglycosylated form at the N-glycosylation site, N119 (QFIEN#GSEFAQK) from lacritin. The mass difference between y8 ion and y7 ion is 114 Da, whereas in (B), the mass difference between y8 ion and y7 ion is 115 Da in the glycosylated form. This shows the presence of both nonglycosylated and glycosylated forms (at the N-glycosylation site, N119) of lacritin in tears.

Typical tear proteins such as lactoferrin, lacritin, prolactininducible protein, zinc alpha-2-glycoprotein 1, Mammaglobin-B 2000

Journal of Proteome Research • Vol. 8, No. 4, 2009

precursor (Lipophilin-C), and lipocalin 2 (LCN2) are found to be N-glycosylated proteins (Tables 1 and 2). It is not surprising

research articles

Quantification of N-Linked Glycoproteins in Tears to see many immunoglobulins are N-glycosylated. There is relatively low abundance of the same type of N-linked glycoproteins in biological fluids, even though collectively they may constitute up to 50% of the secreted proteome. Using a selective capture strategy reduces sample complexity and minimizes masking of glycoproteins by high-abundance nonglycoproteins. In the case of human serum, albumin constitutes a major 50% of the total protein present,32 while lysozyme makes up 20-40% of the total tear fluid protein.33 This prefractionation method allowed identification of additional proteins from tear fluids. Compared with our previous study on the human tear proteome, we identified another 19 tear proteins which were not in the previous tear protein list.5,30 In this paper, we combined glycoprotein capture with iTRAQ labeling (Figure 1) to carry out relative quantitative analysis of N-linked glycoproteins in human tears. We have applied this strategy in the quantitative comparison of N-linked glycoproteins profiles in tears from CDK patients and normal controls. One finding is that the glycosylated form of some serum proteins such as haptoglobin was elevated in tears from CDK patients. However, our results also showed that there was no significant difference of the overall level of haptoglobin between control and CDK samples. It is known that glycosylation can modify protein function or protein folding structure.34 Haptoglobin is a hemoglobin-binding glycoprotein with high concentration in serum. Previous studies showed that the serum haptoglobin level is significantly elevated in inflammation and cancer. Changes in haptoglobin glycosylation patterns have been also found to be associated with disease development and progression.35-39 The origin and exact chemical composition of disease associated droplets remain unclear; however, proteins were found to be part of the constituents.40,24,25 Staining for lipids and calcium indicated their absence in CDK deposits.23,41,42 CDK deposits are usually located within Bowman’s membrane and the anterior stroma. A recent proteomic study of CDK droplets by Menegay et al. identified 105 proteins containing mainly secreted extracellular matrix (ECM) proteins and plasma proteins in CDK specimens. Previous cornea proteomics and bioinformatics have also suggested import of plasma proteins such as albumin, haptoglobin, hemopexin, transferrin, and immunoglobulin chains in the human cornea.43 It is likely ECM proteins and plasma proteins may contribute to the deposit formation. Epidemiology studies44-46 suggested that the risk factors for CDK are related to overexposure to ultraviolet irradiation and other climatic conditions (for example, aridity, constant winds). Inferior iris atrophy and depigmentation, frequently observed among patients with CDK,47 could also be related to this environmental condition. We also found that one lacrimal gland secreted protein, lacritin,48 whose N-glycosylation level was decreased in CDK tears. Lacritin is mitogenic for human corneal epithelial cells.49 Unglycosylated recombinant human lacritin also has mitogenic activity.48,49 Lacritin is a 12.3 kDa glycoprotein with 11 to 12 predicted O-glycosylation sites and one predicted N-linked glycoslyation site (at N119) toward the C-terminus.50 Both nonglycosylated and glycosylated forms at the N-glycosylation site, N119 in lacritin, were seen in tears. This was evident by the mass spectrum of peptide fragment (QFIEN#GSEFAQK) which originated from lacritin with the mass difference between y8 ion and y7 ion of 114 Da, whereas the mass difference between y8 ion and y7 ion was 115 Da in the glycosylated form (Figure 5). There was no obvious change of overall tear lacritin

levels between control group and CDK group (ratio of CDK/ control ) 0.959, Table 3). However, a decrease of N-glycosylation level was found when comparing its N-glycopeptide at site N119 in control and CDK tear samples (ratio of CDK/ control ) 0.701, Table 3). Tear lacritin levels have been found to be down-regulated in blepharitis patients.27 A recent study by Kaji et al.24 suggested that the formation of CDK deposits is related to the aggregation and accumulation of advanced glycation end products (AGEs) due to ultraviolet irradiation and/or the aging process. AGE formation is a very slow progress with no involvement of enzymes.24 They also suggested that the deposits in CDK would be aggregations of various AGE-modified proteins.24 Our results suggest that enzymatic glycosylation may also be involved in the CDK deposit formation because increased N-glycosylation levels of some serum-origin proteins were observed in CDK tears. It would be important to analyze the glycoprotein profile of the CDK deposit.

Conclusion We have successfully identified 43 N-linked glycoproteins in human tear fluid using a selective hydrazide-resin capture method followed by specific PNGase F deglycosylation of asparagine-linked proteins from their carbohydrate moieties, and finally two-dimensional nano-LC-nano-ESI-MS/MS analysis. Of these, 19 glycoproteins were not previously reported in the most comprehensive study of human tear fluid proteome thus far. Five N-linked glycoproteins were found to have significant changes in N-glycosylation levels when comparing tears from climatic droplet keratopathy patients and normal controls. Relative quantitation was accomplished by labeling the enriched fractions of formerly N-glycosylated proteins using iTRAQ isobaric labels and LC-MS/MS analysis. Increased N-glycosylation levels of four N-glycosylated proteins including haptoglobin (at sites N207, N211, and N241), polymeric immunoglobulin receptor (at sites N83, N90, N135, N186, N421, and N469), immunoglobulin J chain (at site N49), and an uncharacterized protein DKFZp686M08189 (at site N470), and decreased N-glycosylation levels of one down-regulated Nglycosylated protein, lacritin (at site N119), were observed in tears from CDK patients. However, the overall levels of these proteins in the tears did not change in CDK, rather the change was in the shift from the glycosylated to the nonglycosylated forms. Further analysis of the biochemical pathways of significant proteins, as well as using other biological assays, should improve our understanding of this disease and allow the development of prognostic or diagnostic biomarkers for CDK.

Acknowledgment. This work was supported by grants NMRC/0808/2003, NMRC/CPG001/2003, NMRC/0982/2005 and NMRC IBG from National Medical Research Council (NMRC), Singapore. References (1) Zhou, L.; Beuerman, R. W.; Foo, Y.; Liu, S.; Ang, L. P.; Tan, D. T. Characterisation of human tear proteins using high-resolution mass spectrometry. Ann. Acad. Med. Singapore 2006, 35 (6), 400– 407. (2) Grus, F. H.; Joachim, S. C.; Pfeiffer, N. Proteomics in ocular fluids. Proteomics Clin. Appl. 2007, 1, 876–888. (3) Zhou, L.; Huang, L. Q.; Beuerman, R. W.; Grigg, M. E.; Li, S. F.; Chew, F. T.; Ang, L.; Stern, M. E.; Tan, D. Proteomic analysis of human tears: defensin expression after ocular surface surgery. J. Proteome Res. 2004, 3 (3), 410–416.

Journal of Proteome Research • Vol. 8, No. 4, 2009 2001

research articles (4) Li, N.; Wang, N.; Zheng, J.; Liu, X. M.; Lever, O. W.; Erickson, P. M.; Li, L. A. Characterization of human tear proteome using multiple proteomic analysis techniques. J. Proteome Res. 2005, 4 (6), 2052– 2061. (5) de Souza, G. A.; Godoy, L. M. F.; Mann, M. Identification of 491 proteins in the tear fluid proteome reveals a large number of proteases and protease inhibitors. GenomeBiology 2006, 7 (8), 11. (6) An, H. J.; Ninonuevo, M.; Aguilan, J.; Liu, H.; Lebrilla, C. B.; Alvarenga, L. S.; Mannis, M. J. Glycomics analyses of tear fluid for the diagnostic detection of ocular rosacea. J. Proteome Res. 2005, 4 (6), 1981–1987. (7) Fung, K.; Morris, C.; Duncan, M. Mass spectrometric techniques applied to the analysis of human tears: a focus on the peptide and protein constituents. Adv. Exp. Med. Biol. 2002, 506 (Pt. A), 601–605. (8) Fung, K. Y.; Morris, C.; Sathe, S.; Sack, R.; Duncan, M. W. Characterization of the in vivo forms of lacrimal-specific prolinerich proteins in human tear fluid. Proteomics 2004, 4 (12), 3953– 3959. (9) Gevaert, K.; Van Damme, P.; Ghesquie`re, B.; Impens, F.; Martens, L.; Helsens, K.; Vandekerckhove, J. A la carte proteomics with an emphasis on gel-free techniques. Proteomics 2007, 7 (16), 2698– 2718. (10) Zhang, H.; Li, X.-j.; Martin, D. B.; Aebersold, R. Identification and quantification of N-linked glycoproteins using hydrazide chemistry, stable isotope labeling and mass spectrometry. Nat. Biotechnol. 2003, 21 (6), 660–666. (11) Zhang, H.; Yi, E. C.; Li, X.-j.; Mallick, P.; Kelly-Spratt, K. S.; Masselon, C. D.; Camp, I.; David, G.; Smith, R. D.; Kemp, C. J.; Aebersold, R. High throughput quantitative analysis of serum proteins using glycopeptide capture and liquid mass spectrometry. Mol. Cell. Proteomics 2004, 4, 144–155. (12) Lewandrowski, U.; Moebius, J.; Walter, U.; Sickmann, A. Elucidation of N-glycosylation sites on human platelet proteins: a glycoproteomic approach. Mol. Cell. Proteomics 2006, 5 (2), 226–233. (13) Taylor, M. E.; Drickamer, K. Introduction to Glycobiology; Oxford University Press, Inc.: New York, 2003. (14) Liu, T.; Qian, W. J.; Gritsenko, M. A.; CampIi, D. G.; Monroe, M. E.; Moore, R. J.; Smith, R. D. Human plasma N-glycoproteome analysis by immunoaffinity subtraction, hydrazide chemistry, and mass spectrometry. J. Proteome Res. 2005, 4 (6), 2070–2080. (15) Wang, Y.; Wu, S. L.; Hancock, W. S. Approaches to the study of N-linked glycoproteins in human plasma using lectin affinity chromatography and nano-HPLC coupled to electrospray linear ion trapsFourier transform mass spectrometry. Glycobiology 2006, 16 (6), 514–523. (16) Kreunin, P.; Zhao, J.; Rosser, C.; Urquidi, V.; Lubman, D. M.; Goodison, S. Bladder cancer associated glycoprotein signatures revealed by urinary proteomic profiling. J. Proteome Re. 2007, 6 (7), 2631–2639. (17) Pan, S.; Wang, Y.; Quinn, J. F.; Peskind, E. R.; Waichunas, D.; Wimberger, J. T.; Jin, J.; Li, J. G.; Zhu, D.; Pan, C.; Zhang, J. Identification of glycoproteins in human cerospinal fluid with a complementary proteomic approach. J. Proteome Res. 2006, 5 (10), 2769–2779. (18) Ramachandran, P.; Boontheung, P.; Xie, Y.; Sondej, M.; Wong, D. T.; Loo, J. A. Identification of N-linked glycoproteins in human saliva by glycoprotein capture and mass spectrometry. J. Proteome Res. 2006, 5 (6), 1493–1503. (19) Dube,D.H.;Bertozzi,C.R.Glycansincancerandinflammationspotential for therapeutics and diagnostics. Nat. Rev. Drug Discovery 2005, 4, 477–488. (20) Shepard, H. M.; et al. , Monoclonal antibody therapy of human cancer: taking the HER2 protooncogene to the clinic. J. Clin. Immunol. 1991, 11, 117–127. (21) Kobata, A.; Amano, J. Altered glycosylation of proteins produced by malignant cells, and application for the diagnosis and immunotherapy of tumors. Immunol. Cell Biol. 2005, 83 (4), 429–439. (22) Tarentino, A. L.; Gdmez, C. M.; Plummer, J.; Thomas, H. Deglycosylation of asparagine-linked glycans by peptide:N-glycosidase F. Biochemistry 1985, 24, 4665–4671. (23) Gray, R. H.; Johnson, G. J.; Freedman, A. Climatic droplet keratopathy. Surv. Ophthalmol. 1992, 36, 241–253. (24) Kaji, Y.; Nagai, R.; Amano, S.; Takazawa, Y.; Fukayama, M.; Oshika, T. Advanced glycation end product deposits in climatic droplet keratopathy. Br. J. Ophthalmol. 2007, 91, 85–88. (25) Menegay, M.; Lee, D.; Tabbara, K. F.; Cafaro, T. A.; Urrets-Zavalia, J. A.; Serra, H. M.; Bhattacharya, S. K. Proteomic analysis of climatic keratopathy droplets. Invest. Ophthalmol. Visual Sci. 2008, 49 (7), 2829–2837.

2002

Journal of Proteome Research • Vol. 8, No. 4, 2009

Zhou et al. (26) Green-Church, K. B.; Nichols, K. K.; Kleinholz, N. M.; Zhang, L.; Nichols, J. J. Investigation of the human tear film proteome using multiple proteomic approaches. Mol. Vision 2008, 14, 456–470. (27) Koo, B. S.; Lee, D. Y.; Ha, H. S.; Kim, J. C.; Kim, C. W. Comparative analysis of the tear protein expression in blepharitis patients using two-dimensional electrophoresis. J. Proteome Res. 2005, 4, 719– 724. (28) Ueda, K.; Katagiri, T.; Shimada, T.; Irie, S.; Sato, T. A.; Nakamura, Y.; Daigo, Y. Comparative profiling of serum glycoproteome by sequential purification of glycoproteins and 2-nitrobenzensulfenyl (NBS) stable isotope labeling: a new approach for the novel biomarker discovery for cancer. J. Proteome Res. 2007, 6 (9), 3475– 3483. (29) Ham, B. M.; Jacob, J. T.; Cole, R. B. Single eye analysis and contralateral eye comparison of tear proteins in normal and dry eye model rabbits by MALDI-ToF mass spectrometry using waxcoated target plates. Anal. Bioanal. Chem. 2007, 387 (3), 889–900. (30) Beuerman, R. W.; Zhou, L.; Prema, P.; Chan, C.; Ang, L. P. K.; Angayarkanni, N.; Foo, Y. H.; Liu,S. P.; Tan, D. T. H. Comprehensive characterization of human tear proteome using nano-liquid chromatography-QTOF tandem mass spectrometry and quantitative proteomics (iTRAQ). Association for Research in Vision and Ophthalmology, annual meeting, 2006. e-abstract: http://www.arvo. org/eweb/DynamicPage.aspx?site)arvo2&webcode)amabstractsearch. (31) Beuerman, R. W.; Zhou, L.; Cafaro, T.; Urrets-Zavalia, E.; UrretsZavalia, J.; Serra, H. Metabolomic and proteomic analysis of tears from climatic droplet keratopathy patients in Argentina. Association for Research in Vision and Ophthalmology, annual meeting, 2007. e-abstract: http://www.arvo.org/eweb/DynamicPage.aspx? site)arvo2&webcode)amabstractsearch. (32) Anderson, N. L.; Anderson, N. G. The human plasma proteome: history, character, and diagnostic prospects. Mol. Cell. Proteomics 2002, 1, 845–867. (33) Korb, D. R.; Craig, J.; Doughty, M.; Guillon, J. P.; Smith, G.; Tomlinson, A. The Tear Film: Structure, Function and Clinical Examination,.; Butterworth-Heinemann: Oxford, 2002, p 30. (34) Bosques, C. J.; Tschampel, S. M.; Woods, R. J.; Imperiali, B. Effects of glycosylation on peptide conformation: a synergistic experimental and computational study. J. Am. Chem. Soc. 2004, 126 (27), 8421–8425. (35) Turner, G. A. Haptoglobin. A potential reporter molecule for glycosylation changes in disease. Adv. Exp. Med. Biol. 1995, 376, 231–238. (36) Heo, S. H.; Lee, S. J.; Ryoo, H. M.; Park, J. Y.; Cho, J. Y. Identification of putative serum glycoprotein biomarkers for human lung adenocarcinoma by multilectin affinity chromatography and LC-MS/ MS. Proteomics 2007, 7 (23), 4292–4302. (37) Nakano, M.; Nakagawa, T.; Ito, T.; Kitada, T.; Hijioka, T.; Kasahara, A.; Tajiri, M.; Wada, Y.; Taniguchi, N.; Miyoshi, E. Site-specific analysis of N-glycans on haptoglobin in sera of patients with pancreatic cancer: a novel approach for the development of tumor markers. Int. J. Cancer 2008, 122 (10), 2301–2309. (38) Fujimura, T.; Shinohara, Y.; Tissot, B.; Pang, P. C.; Kurogochi, M.; Saito, S.; Arai, Y.; Sadilek, M.; Murayama, K.; Dell, A.; Nishimura, S.; Hakomori, S. I. Glycosylation status of haptoglobin in sera of patients with prostate cancer vs. benign prostate disease or normal subjects. Int. J. Cancer 2008, 122 (1), 39–49. (39) Smeets, M. B.; Sluijter, J. P. G.; Donners, M. M. P. C.; Velema, E.; Heeneman, S.; Pasterkamp, G.; de Kleijn, D. P. V. Increased arterial expression of a glycosylated haptoglobin isoform after balloon dilation. Cardiovasc. Res. 2003, 58, 689–695. (40) Duhaiman, A. S.; Aprahamian, S.; Gorban, A. M.; Shoukrey, N.; Tabbara, K. F. Biochemical analysis of climatic droplet keratopathy. Saudi Bull. Ophthalmol. 1988, 3, 147–149. (41) Garner, A.; Morgan, G.; Tripathi, R. C. Climate dropley keratopathy. II Pathologic findings. Arch. Ophthalmol. 1973, 89 (3), 198–204. (42) Johnson, G. J.; Overall, M. Histology of spheroidal degeneration of the cornea in Labrador. Br. J. Ophthalmol. 1978, 62 (1), 53–61. (43) Karring, H.; Thøgersen, I. B.; Klintworth, G. K.; Møller-Pedersen, T.; Enghild, J. J. The human cornea proteome: bioinformatic analyses indicate import of plasma proteins into the cornea. Mol. Vision 2006, 12, 451–460. (44) Rodger, F. C.; Cuthill, J. A.; Fydelor, P. J.; Lenham, A. P. Ultraviolet radiation as a possible cause of corneal degenerative changes under certain physiographic conditions. Acta Ophthalmol. 1974, 52, 777–785. (45) Freedman, A. Climatic droplet keratopathy. I. Clinical aspects. Arch. Ophthalmol. 1973, 89, 193–197.

research articles

Quantification of N-Linked Glycoproteins in Tears (46) Urrets-Zavalı´a, J. A.; Knoll, E. G.; Maccio, J. P.; Urrets-Zavalı´a, E. A.; Saad, J. A.; Serra, H. M. Climatic droplet keratopathy in the Argentine Patagonia. Am. J. Ophthalmol. 2006, 141 (4), 744–746. (47) Urrets-Zavalia, J. A.; Maccio, J. P.; Knoll, E. G.; Cafaro, T. A.; UrretsZavalia, E. A.; Serra, H. M. Surface alterations, corneal hypoesthesia and iris atrophy in patients with climatic droplet keratopathy. Cornea 2007, 26 (7), 800–804. (48) Sanghi, S.; Kumar, R.; Lumsden, A.; Dickinson, D.; Klepeis, V.; Trinkaus-Randall, V., Jr.; Laurie, G. W. cDNA and genomic cloning of lacritin, a novel secretion enhancing factor from the human lacrimal gland. J. Mol. Biol. 2001, 310, 127–139.

(49) Wang, J.; Wang, N.; Xie, J.; Walton, S. C.; McKown, R. L.; Raab, R. W.; Ma, P.; Beck, S. L.; Coffman, G. L.; Hussaini, I. M.; Laurie, G. W. Restricted epithelial proliferation by lacritin via PKCalphadependent NFAT and mTOR pathways. J. Cell Biol. 2006, 174, 689– 700. (50) McKown, R. L. ; Wang, N. ; Raab, R. W. ; Karnati, R. ; Zhang, Y. ; Williams, P. B. ; Laurie, G. W. Lacritin and other new proteins of the lacrimal functional unit. Exp. Eye Res. 2008, in press.

PR800962Q

Journal of Proteome Research • Vol. 8, No. 4, 2009 2003