The Proteome of the Wool Cuticle Henning Koehn,*,† Stefan Clerens,† Santanu Deb-Choudhury,† James D. Morton,‡ Jolon M. Dyer,† and Jeffrey E. Plowman† AgResearch, Growth and Development Section, Lincoln Research Centre, Christchurch, New Zealand, and Department of Wine, Food and Molecular Biosciences, Lincoln University, New Zealand Received December 2, 2009
The cuticle is responsible for important wool fiber characteristics such as handle and abrasion resistance, which impact on the fiber’s performance in both interior and apparel textiles. The cuticle proteome, however, is not well understood due to the difficulty in isolating pure wool cuticle and its significant resistance to protein extraction, which is attributed to the presence of extensive disulfide and isopeptide cross-linking. We investigated the proteome of highly pure Merino wool cuticle using a combined strategy of chemical and enzymatic digestion and identified 108 proteins, including proteins responsible for a variety of cellular processes. The majority of identified proteins belonged to keratin and nonkeratin protein families known to play an important role in molecular assembly and cellular structure. Keratinassociated, intermediate filament and cytoskeletal keratin proteins were identified as the most prominent keratinous cuticular constituents, while histones, tubulins, and desmosomes were the key nonkeratin structural proteins. We conclude that a variety of proteins contribute to cuticle structure and fiber characteristics, and that the keratinous protein families of IFPs and KAPs represent the most important cuticular constituents. Keywords: cuticle • wool • NTCB • cysteine • KAPs • keratin
Introduction All wool fibres have a common molecular structure, comprising two prominent layers: an inner cortex and an outer cuticle. The cortex is primarily responsible for the mechanical attributes of strength and rigidity,1 while the cuticle plays a critical role in the fiber’s durability, felting, and shrinking. Keratin intermediate filament proteins (IFPs) and keratinassociated proteins (KAPs) are the most prominent protein constituents, contributing approximately 90% of the fiber’s substance by weight.2 IFPs dominate the inner cortex, while the cuticle has been thought to be largely composed of KAPs. This class of proteins is also present in the matrix surrounding the intermediate filaments in the cortex, where they are believed to cross-link with the IFPs. The cuticle itself can be roughly divided into three regions: the endocuticle, exocuticle, and a-layer,3 which exhibit an increasing sulfur content moving from the innermost layer, the endocuticle, to the a-layer on the outside.3-5 These layers contain type I and type II IFPs, as well as a variety of KAPs. The type I IFPs are acidic proteins with molecular weights around 45-50 kDa, while type II IFPs are neutral-basic with molecular weights between 54 and 56 kDa.6 There are known to be 11 type I IFPs and six type IIs, though not all are expressed in hair and wool, with one type I being found only in vellus * To whom correspondence should be addressed. Henning Koehn, AgResearch, Private Bag 4749, Christchurch 8140, New Zealand. Tel.: +643-321-8800. Fax: +64 3 321 8811. E-mail:
[email protected]. † AgResearch. ‡ Lincoln University.
2920 Journal of Proteome Research 2010, 9, 2920–2928 Published on Web 04/27/2010
hair and one type II in cytokeratal extracts from the tongue.7 The proteins exhibit a high degree of homology, up to 92% in the case of some proteins in the type I IFP family, but between the two families, it is in the order of 30%. KAPs are divided into three groups depending on their sulfur content: high sulfur (HS), 30 mol % (KAP 4, 5, 9, and 17); and high glycinetyrosine (HGT) (KAP 6-8 and 18-22).4 Recently, additional KAP members, KAP 24.18 and KAP 26.1,9 have been described; these are stated to contain ∼9 mol % and ∼10 mol % cysteine, respectively, and are therefore included in the HS category. The majority of KAPs have a molecular weight of less than 20 kDa, although a few are significantly larger, such as KAP 10, which has an approximate weight of 46 kDa. Past research of the cuticle has had a strong focus on gene expression in the hair follicle of humans and sheep, which has highlighted the polymorphic nature of gene families associated with cuticle proteins, particularly those of the KAPs4,10-15 and IFPs.7,16 Information on the actual protein composition and location in the fiber is somewhat scarce by comparison.13,17,18 While data exist regarding the amino acid composition of particular parts of the wool fiber, including the cuticle, this is of little consequence when trying to identify and locate the actual proteins within the fiber, a necessary prerequisite before any conclusions can be drawn regarding how they interact to contribute to the fiber’s unique properties.19,20 In recent reports by Bringans21 and Koehn,22 a number of KAPs were identified in the a-layer and the endo- and exocuticle of Merino wool. However, identification of the major constituents of the cuticle 10.1021/pr901106m
2010 American Chemical Society
Proteome of the Wool Cuticle has been far from complete and the presence of additional cuticle proteins remains to be investigated. The lack of comprehensive information on the proteome of the wool cuticle is in part attributable to the difficulty in adequately separating the fiber cortex and cuticle for sound proteomic analysis. Another reason is the refractory character of parts of the cuticle to standard enzymatic digestion, believed to be caused by the extensive disulfide and ε-(γ-glutamyl)lysine cross-links.3,21,23 The high degree of sequence homology and identity between different KAP and IFP families further complicates correct and unequivocal identification of the proteins present in the wool cuticle.2,4,10-12 Nevertheless, some progress toward resolving ambiguity in the identification of cysteinerich proteins has been made recently with the development of a methodology which employs a combined approach of chemical and enzymatic digestion, showing clear benefits to the study of the wool cuticle.22 The purpose of the present study was to determine the identity of the major protein constituents of the cuticle proteome of Merino wool as these provide a basis for better understanding of the unique characteristics of the wool fiber and the cuticle in particular.
Materials and Methods Materials. Tris(2-carboxyethyl)phosphine hydrochloride (TCEP) was purchased from Fluka (Buchs, Switzerland). 2-Nitro-5-thiocyanobenzoic acid (NTCB), thiourea, and guanidinium HCl were from Sigma (St. Louis, MO). Glycine, urea, methanol, and chloroform were bought from Merck (Darmstadt, Germany). Ammonium bicarbonate, trifluoroacetic acid, acetic acid, and acetone were from BDH (Poole, England). Chloroform was supplied by Selby-Biolab (Clayton, Vic, Australia). Formic acid was from Ajax Finechem (Auckland, New Zealand). Acrylamide was from Bio-Rad (Hercules, CA). Ethanol was from SDS (Peypin, France). Water and acetonitrile were supplied by Mallinckrodt Baker (Phillipsburg, NJ). R-Cyano-4hydroxycinnamic acid (CHCA) was obtained from Bruker Daltonics (Bremen, Germany). All laboratory consumables were purchased from Eppendorf (Hamburg, Germany). Sequencing grade porcine trypsin was from Promega (Madison, WI). Sample Preparation. A midside sample of wool (∼20 µm diameter) was obtained from a 12 month old New Zealand Merino sheep, at which point the fibres are of commercial significance. Fibres from which ∼15 mm of material had been removed from the tip end were scoured according to previously published methods.21 Subsamples of 10 g were moistened with water and ground in a Retsch mill (Haan, Germany). Cuticular dust generated during the milling process was collected from the lid. Scanning Electron Microscopy. Samples were mounted onto aluminum stubs using conductive carbon adhesive tape and sputter coated from a gold/palladium leaf source to impart conductivity to the surface of the sample. The thickness of the coating was approximately 10 nm. Samples were studied using a Jeol JSM 7000F field emission gun scanning electron microscope (Tokyo, Japan). The microscope was operated at 5 kV and samples viewed at a working distance of 15 mm. Sample purity was assessed by an expert microscopist and was based on a cell count of the number of cuticle to cortical fragments per grid square over 10 grid squares, based on their different morphologies.24 Cuticle Fractionation. Soluble cuticular proteins were specifically extracted from purified cuticle using TCEP as reductant
research articles under similar conditions to those previously described for the digestion of whole wool.21 Cuticle samples of 10 mg were extracted in 200 µL of 7 M urea, 2 M thiourea, 50 mM Tris, and 50 mM TCEP, pH 4, at room temperature overnight. The soluble fraction was methanol/chloroform precipitated as previously described.25 The insoluble material was then either rinsed in three changes of 800 µL of charcoal-filtered water for NTCB treatment, or in 50 mM ammonium bicarbonate for tryptic digestion. Trypsin Digestion. The entire insoluble cuticle fraction remaining after the TCEP extraction and the entire precipitation pellet from the soluble fraction were used for trypsin digestion. Both insoluble and soluble fractions were separately reduced in 50 µL of 50 mM TCEP in 100 mM ammonium bicarbonate, pH 8, at 56 °C for 45 min, followed by alkylation in 50 µL of 360 mM acrylamide in 100 mM ammonium bicarbonate for 30 min. Trypsin (1 µg in 100 µL of 50 mM ammonium bicarbonate) was added and the extract digested at 37 °C for 18 h. NTCB and Trypsin Digestion. Both soluble and insoluble fractions generated from 10 mg of cuticle were digested in 200 µL of NTCB working solution (6 M guanidine HCl, 1 M glycine, 50 mM NTCB, pH 9) at 37 °C with light vortexing overnight. From each digest, 100 µL was taken for analysis (NTCB-only digest), while the remaining 100 µL was methanol/chloroform precipitated. Insoluble cuticular material remaining after NTCB digestion of the insoluble fraction was rinsed in three changes of 50 mM ammonium bicarbonate. Pellets from precipitation as well as the insoluble cuticular material were separately digested with 1 µg of trypsin in 100 µL of 50 mM ammonium bicarbonate at pH 8 overnight. LC-MALDI. LC-MALDI was performed on 600 µm AnchorChip plates using a Proxeon Easy-nLC, Proteineer fc fraction collection robot and an Ultraflex III MALDI-TOF/TOF mass spectrometer (Bruker, Bremen, Germany) in positive ion mode. Samples digested with trypsin only were diluted 60-fold, while samples treated with NTCB and trypsin were diluted 30-fold in 3% acetonitrile/0.1% trifluoroacetic acid in water, loaded on a trap (30 mm, 100 µm i.d.), and separated on an analytical (150 mm, 75 µm i.d.) column (both trap and column were inhouse packed with Microsorb C18 300-5 media (Varian, Palo Alto, CA), using a 3-55% B (acetonitrile/0.1% trifluoroacetic acid) gradient over 80 min with a flow rate of 300 nL/min). Saturated CHCA matrix solution in acetone was diluted 10fold in ethanol/acetone/ammonium phosphate (ratio 6:3:1) and mixed with the sample postcolumn at a flow rate of 700 nL/ min. A total of 380 fractions were collected from 8-84 min in 12 s time slices. Spectra were calibrated externally using a Peptide Calibration Standard (Bruker) containing Angiotensin I (m/z 1296.6848) and II (m/z 1046.5418), Substance P (m/z 1347.7354), Bombesin (m/z 1619.8223), ACTH-clip [1-17] (m/z 2093.0862), ACTH-clip [18-39] (m/z 2465.1983) and Somatostatin (m/z 3147.471) diluted 6-fold with matrix solution. LC-MS/MS. Samples were analyzed using an Ultimate nanoscale HPLC (LC Packings, Amsterdam, The Netherlands) with Famos autosampler and Switchos column switching module. The loading pump was an LC-10AT isocratic pump (Shimadzu) at a flow rate of 8 µL/min. Because additional peptides were liberated from the insoluble cuticle pellets during digestion, accurate protein estimation was not possible using conventional protein assays. Hence, optimal protein loads for analysis were determined empirically and applied as follows. Samples digested with NTCB and trypsin were diluted 30-fold, in 2% Journal of Proteome Research • Vol. 9, No. 6, 2010 2921
research articles
Koehn et al.
Figure 1. Experimental strategy for the identification of proteins from the wool cuticle. Samples analyzed directly after NTCB digestion were performed in duplicate (*), while all remaining sample analyses were performed as biological triplicates. In total, 38 analyses of 19 samples were performed. NTCB only digests (4 samples total) were performed in biological duplicate, while trypsin only digests (6 samples total) and NTCB and trypsin digests (9 samples total) on soluble and insoluble cuticle fractions were performed in biological triplicate, using LC-ESI and LC-MALDI (LCM).
acetonitrile/0.2% acetic acid in water, while samples digested with trypsin only were diluted 60-fold; 10 µL was used for analysis. Samples were loaded on the trap column (repacked 5 mm, 300 µm i.d. LC Packings cartridge) and separated on a 190 mm, 75 µm i.d. analytical column (both packed with Microsorb C18 300-5 media, Varian) coupled to a QSTAR Pulsar i mass spectrometer (Applied Biosystems, Foster City, CA) using a Proxeon stainless steel nanospray capillary at 2200 V. The gradient was 2-55% B (acetonitrile/0.2% formic acid) over 80 min with a flow rate of 150 nL/min. MS data was acquired from m/z 350-1200 and MS/MS from 40-1600 m/z accumulating three cycles over 1.3 s duration each. Data Analysis. For protein identification, data was searched against an NCBInr database (NCBInr_20091103) augmented with an in-house sheep sequence EST-database (AgResearch, Lincoln, New Zealand) using Mascot v2.2.03 (Matrix Science, London, U.K.), specifying Ovis aries as taxonomy. Propionamide and oxidation of methionine were specified as variable modifications for peptides generated using the standard protocol for trypsin digestion. Methionine oxidation, dehydroalanine, C-terminal amidation, and NTCB modification of N-terminal cysteine were used as variable modifications for peptides generated using NTCB only and NTCB in combination with trypsin. Enzyme specificity was set to semitrypsin and/or NTCB, where applicable. Error tolerance was set to 155 ppm for LC-MS, 100 ppm for LC-MALDI, and 0.5 Da for MS/MS. Data was compiled and analyzed using ProteinScape 2.1 (Bruker) with acceptance thresholds for peptide and protein scores set at 20 and 40, respectively. Protein and peptide lists were compiled using the Protein-Extractor functionality in ProteinScape including automatic assessment of true and false positive identification of proteins and peptides matches according to the protein and peptide settings detailed above. At least one peptide from each protein was required to exceed the identity threshold (p e 0.05), indicating correct protein identification. Only those results meeting or exceeding the threshold values and assessed as true matches were used for further analysis. 2922
Journal of Proteome Research • Vol. 9, No. 6, 2010
Results The purpose of this study was to investigate the protein constituents of the cuticle from Merino wool with a particular focus on its keratinous constituents. Cuticular flakes isolated from whole wool fibres and confirmed by SEM as highly pure (>95%) cuticle fractions were used for proteomic investigation as previously described.22,24 As a first step, proteins were extracted using a solution containing a powerful reducing agent (TCEP) to generate soluble and insoluble cuticle fractions. Then, a 3-fold approach was adopted whereby each fraction was individually digested with NTCB, trypsin, or both (NTCB and trypsin). In the latter case, the cuticle was first digested with NTCB, followed by digestion with trypsin. A total of 19 samples were analyzed using both LC-ESI and LC-MALDI (resulting in 38 analyses) as depicted in Figure 1. MS data acquired for each approach was searched against the NCBInr database augmented with in-house sheep sequences and search results were collated using ProteinScape. The analyses were performed in biological triplicates, with the exception of samples treated exclusively with NTCB, which were discontinued after analysis of biological duplicates, where search results were generally found to produce significantly fewer hits and did not contribute any additional proteins or unique peptides compared to samples analyzed after NTCB and trypsin digestion. In all, 108 cuticle proteins were positively identified, including proteins from various protein families responsible for a wide variety of cellular functions and processes (Table 1 and Figure 2). Among these, keratinous proteins such as KAPs and IFPs accounted for the majority of identified proteins and almost 70% of all matched peptides. Very high sequence coverage was obtained for type I and II IFPs (Figure 3) with nine of the 12 IFPs identified exhibiting sequence coverage of 80% or higher (Table 1). While the sequence coverage for KAPs was comparatively lower, a large number (29) of KAPs, including eight UHS, 17 HS, and four HGT were identified as constituents of the wool cuticle. Additional structural proteins were also identified, including six cytoskeletal keratins and 10 nonkeratinous proteins such as histones and tubulins. Several proteins known to
research articles
Proteome of the Wool Cuticle a
Table 1. Wool Cuticle Proteins accession
protein
gi|164652870
IFP Type II K85 (Partial K85, gi|246276, 1 AA difference) hair keratin type II intermediate filament IFP Type II K85 IFP Type I K33b IFP Type I K31 IFP Type II K86 IFP Type I K34 IFP Type II K81 IFP Type II K82 partial IFP Type II K83 IFP Type I K33a IFP Type I K 35 IFP Type I K32 Partial IFP Type I K39 Partial KAP 26.1 KAP 12.1 KAP 2.3 var3 KAP 19.3 KAP 11.1 KAP 6.1 KAP 10 like Partial KAP 2.3 KAP 2.3 var1 KAP 2.3 var2 KAP 4-like KAP 7 KAP4.25 KAP 4.12 KAP4.19 KAP 16.6 KAP 1.2 KAP4.24 KAP 1.4 KAP 13.1 KAP 1.4 KAP 24.1 Partial KAP 3.1 KAP 9.2 KAP 3.4 KAP 5.1 KAP 8.1 KAP 13.1 var1 KAP 15.1 Keratin 6A Type II Cytoskeletal Partial Keratin 5 Type II Cytoskeletal Keratin 18 Type I Cytoskeletal Keratin 7 Type II Cytoskeletal Keratin 10 Type I Cytoskeletal Keratin 14 Type I Cytoskeletal Junction plakoglobin Partial Desmoglein 4 Partial Desmoplakin isoform 1 partial Plakophilin 3 Partial Histone H4 Partial Histone H2A Histone H2B var1 H3 histone family 3A tubulin, alpha 1 tubulin, alpha 4 beta actin Tubulin beta Chain B, Tubulin-Colchicine-Phomopsin A: Stathmin-Like Domain Complex Histone 1 family member partial Heat Shock 70 kDa protein 5
gi|313239 gi|246276 gi|999000003 gi|125090 gi|999000010 gi|999000004 gi|999000007 gi|999000092 gi|125116 gi|999000002 gi|168693435 gi|999000056 gi|999000064 gi|999000054 gi|999000037 gi|999000042 gi|999000016 gi|999000018 gi|999000021 gi|999000052 gi|125644 gi|999000040 gi|999000041 gi|999000058 gi|224555964 gi|999000141 gi|999000050 gi|999000135 gi|999000034 gi|125624 gi|999000140 gi|259221166 gi|999000030 gi|259221163 gi|999000055 gi|229307 gi|999000072 gi|125662 gi|57546909 gi|999000043 gi|999000017 gi|999000032 gi|999000076 gi|999000083 gi|999000075 gi|999000079 gi|999000082 gi|999000077 gi|999000089 gi|999000067 gi|999000097 gi|999000068 gi|999000066 gi|999000070 gi|999000069 gi|165940904 gi|6755901 gi|6678467 gi|4501885 gi|999000044 gi|209870470 gi|999000104 gi|999000101
MW [kDa]
unique peptides
total peptides
SC (%)
pI
scores
groupb
37.9
1
69
97
4.8
5892
1
12.9 55.3 47.7 46.6 54.8 46.6 55.1 22 53.6 45.9 50.4 41.8 28.8 19.2 11.1 14.1 7.6 16.7 8.3 46.4 14.2 14 14 17.5 9.1 22.4 22.2 27.5 7.5 16 20.1 18.8 17.4 18.8 21 10.3 30.4 10.5 16.1 6.7 17.7 14.6 57.5 62.3 49 51.4 56.3 51.8 30.2 28.7 46.2 25.2 9.7 13.9 13.8 14.7 50.1 49.9 41.7 49.7 49.9
3 6 11 19 6 24 6 17 28 15 30 27 6 22 16 1 4 10 10 48 1 1 2 1 5 1 2 1 1 5 2 5 11 1 8 3 7 2 2 4 1 1 6 6 7 6 6 4 5 3 6 4 17 13 17 10 1 1 15 2 1
21 90 77 81 103 83 97 27 93 66 54 40 6 22 16 16 7 10 10 48 10 15 12 11 5 14 17 19 4 12 10 13 12 9 8 3 7 2 2 4 2 1 19 19 10 13 8 6 5 3 6 4 17 13 17 10 20 18 15 15 14
94 94 90 89 88 87 87 85 84 82 74 69 22 85 75 75 74 72 66 65 61 58 58 57 57 49 48 47 44 44 42 40 39 37 33 24 23 21 17 16 8 7 23 21 17 17 14 13 21 15 14 13 73 62 59 57 43 41 37 32 29
10.8 6.1 4.6 4.6 5.7 4.8 6 4.5 5.4 4.7 4.8 4.7 5.3 9.5 11.4 12.5 9.9 10.5 9.5 9.7 12.5 12.6 12.6 12.7 9.6 12.4 12.8 12.7 10.5 7.2 12.6 7.1 10.1 7.1 9.4 7.9 11.8 8 12.6 9.3 10.7 9.9 7.6 8.6 5.2 5.8 4.9 4.9 7.9 4.1 9 10.6 11.1 11.4 10.8 11.8 4.8 4.8 5.2 4.6 4.6
1058 7103 6294 6486 7897 5905 7909 2134 6741 5144 4417 2683 280 1260 815 710 489 625 680 2611 463 594 520 410 223 558 752 835 193 658 363 1026 475 577 546 176 280 201 214 139 75 43 691 762 404 691 261 233 311 126 170 148 983 693 1061 473 1039 973 682 994 985
1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 5 5 5 5 5 5 5 5 5
15.9 72.4
5 10
5 10
23 19
11.2 4.90
281 382
5 6
KAP info
HS HS HS HGT HS HGT HS HS HS HS UHS HGT UHS UHS UHS UHS HS UHS HS HS HS HS HS UHS HS UHS HGT HS HS
Journal of Proteome Research • Vol. 9, No. 6, 2010 2923
research articles
Koehn et al.
Table 1. Continued accession
protein
gi|60592792 gi|78558752 gi|4503545 gi|165940892 gi|999000091 gi|78557679 gi|999000102 gi|78558068 gi|999000099 gi|224797873 gi|57163961 gi|157780891 gi|112696
heat shock 90kD protein 1, alpha ribosomal protein S15a ribosomal protein S11 ribosomal protein L6 Elongation factor 1 alpha eukaryotic translation initiation factor 4A Eukaryotic translation elongation factor 2 eukaryotic translation initiation factor 5A Poly(A) binding protein, cytoplasmic 1 hairy/enhancer-of-split related with YRPW motif 2 stratifin casein kinase I epsilon 14-3-3 protein zeta/delta (Protein kinase C inhibitor protein 1) (KCIP-1) Leucine aminopeptidase 3 Partial G protein-coupled receptor, family C, group 5, member D pituitary adenylate cyclase-activating polypeptide type 1 receptor Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) glycogen phosphorylase kappa-casein Leucine Rich Repeat Containing 15 Partial Cathepsin D precursor Interleukin 1 family, member 10 Similar to V-set and Ig domain containing 8 partial fibulin 1-D MHC class II MHC class I antigen toll-like receptor 7 laminin receptor precursor integrin, beta 2 PREDICTED: similar to lectin, galactose binding, soluble 7 Galectin 3 GDP dissociation inhibitor 2 Retinoic acid receptor alpha partial ATPase, Cu++ transporting, beta polypeptide PREDICTED: NudC domain containing 1 ATP-binding cassette subfamily B member 1 RAS (RAD and GEM)-like GTP-binding 1 prepro serum albumin peptidylarginine deiminase ubiquitin and ribosomal protein L40 precursor Selenium binding protein 1 Partial S100 calcium binding protein A3 Cyclophilin A (tentative) stem cell factor
gi|999000059 gi|999000062 gi|18086551 gi|6166169 gi|57163939 gi|42495564 gi|999000051 gi|18203300 gi|999000060 gi|999000111 gi|78499391 gi|1094015 gi|108792435 gi|168831416 gi|146674809 gi|57164157 gi|999000095 gi|999000093 gi|240849297 gi|999000103 gi|57619187 gi|999000096 gi|57526446 gi|240849273 gi|57164373 gi|57526451 gi|4507761 gi|999000047 gi|999000063 gi|999000038 gi|1881771
MW [kDa]
unique peptides
total peptides
SC (%)
84.7 9.2 10.3 20.4 49.7 13.6 95.3 16.8 70.6 26.4 27.8 46.5 27.8
5 2 2 2 15 4 12 2 8 3 13 13 5
5 2 2 2 15 4 12 2 8 3 17 13 9
54.9 36.1
7 5
43.4
groupb
pI
scores
7 27 18 9 36 27 16 14 13 11 61 32 31
4.8 11.2 10.8 11 9.7 5.8 6.4 4.9 10 10.6 4.5 9.9 4.6
195 79 80 110 682 149 494 80 265 74 1283 524 391
6 7 7 7 8 8 8 8 8 8 9 9 9
7 5
14 11
6.1 5.3
243 224
9 9
2
2
5
4.9
53
9
34.7 97.2 12.6 60.3 39.8 16.7 15.9 19.6 24.7 41.5 120.1 25.5 84.3 15.4
7 5 1 33 6 4 6 3 3 4 2 3 3 11
7 5 1 33 6 4 6 3 3 4 2 3 3 11
26 4 17 49 17 33 17 17 11 9 2 16 3 86
8.8 6.7 5.7 8 6.6 4.5 9.3 4.4 5 6.4 6.8 5.2 6.1 6.1
271 126 69 2111 237 181 320 67 84 98 63 143 64 488
10 10 11 12 12 13 13 13 13 13 13 14 14 15
29.2 50.4 48.2 160.9 17.8 141.9 32.9 69.1 74.7 14.7 48 11.2 16.8 29.2
7 7 3 12 4 9 2 2 14 3 32 7 9 1
7 7 3 12 4 9 2 2 14 3 32 7 9 1
27 14 8 8 8 6 6 4 27 17 71 52 38 4
9.7 5.9 9.2 6.1 4.5 9.7 10.3 5.8 5.50 10.7 6.3 4.9 9.7 5.1
346 164 88 306 145 253 70 73 512 104 2212 463 339 44
15 15 15 15 15 15 15 15 16 16 17 17 17 17
KAP info
a All identified proteins are listed and grouped according to the family of proteins that best fits them. Information on the number of unique peptides and total peptides identified in each protein, overall sequence coverage, pI, and molecular weight are shown. Proteins identified from the in-house sheep database are indicated by accession numbers starting with “999”. In the case of IFP type II K85, both the C-terminal fragment, which differs by one amino acid, as well as the full sequence were identified. While these database matches were treated as a single hit for K85 in determining which of the different type II IFPs were identified, they were treated as independently identified IFPs for all subsequent data analysis. While hair keratin type II intermediate filament (gi|313239) shares homology with IFP type II K83, it exhibits a significantly different N-terminal protein sequence. However, hair keratin type II intermediate filament (gi|313239) is currently not assigned according to the Knn nomenclature. As a result, it was not included as a positive IFP type II identification. It was, however, treated as a general positive identification of IFP for all other data analysis. For statistical presentation (Figures 2 and 3), related protein families were combined as indicated (#, §). b Legend (Group, Protein Family):1, IFPs; 2, KAPs; 3, Cytoskeletal Keratins; 4, Desmosomal Proteins#; 5, Structural Proteins; 6, Protein Stabilisation§; 7, Protein Synthesis§; 8, Translation & Transcription Regulation; 9, Signalling; 10, Glycolysis§; 11, Metabolism§; 12, Proteases§; 13, Immune Response; 14, Cell Adhesion#; 15, Binding & Transport Proteins; 16, Protein Metabolism & Modification§; 17, Unknown Function.
be involved in protein modification and metabolism such as protein binding and transporter proteins, and proteases and enzymes involved in glycolysis and cell metabolism, protein synthesis, and protein stabilization, were also identified, with 2924
Journal of Proteome Research • Vol. 9, No. 6, 2010
sequence coverage generally below 30%. Proteins associated with immune response and DNA translation were also identified. In addition, a number of signaling proteins and transcription factors with good sequence coverage (>30%) were detected.
Proteome of the Wool Cuticle
research articles
Figure 2. Distribution of unique peptides, all peptides, and identified protein families in the wool cuticle. Protein families involved in protein modification and metabolism (*) include those playing a role in synthesis, metabolism, stabilization, modification, and glycolysis as well as proteases. Detailed information on the individual proteins from each group is summarized in Table 1.
A significant number of cuticle proteins known to be responsible for cell adhesion processes, such as laminin and integrin, as well as plakoglobin, desmoglein, desmoplakin and plakophilin, were also identified. Prominent proteins in hair and wool of unknown function such as selenium binding protein 1 and S100 calcium binding protein A3 were both found to be present in the wool cuticle, exhibiting sequence coverage of 71% and 52%, respectively. As a next step, the data was manually validated to ensure each identified protein contained at least one unique peptide, not present in any other identified protein (Supplementary Table 1). IFP and KAP family members are known to exhibit extensive sequence homology which was reflected in the relatively low percentage of unique peptides identified within the IFP (22%), KAP (55%), and cytoskeletal keratin families (47%); other protein families almost exclusively exhibited unique peptides, with the exception of structural and adhesion proteins (Figure 3). However, all IFPs featured several unique peptides with the exception of keratin 85 (gi|164652870, only one unique peptide, Table 1), allowing unambiguous identifi-
cation. A total of 10 identified KAPs, including KAPs 1.4, 2.3, 2.3 var 1 and var 3, 4-like, 4.19, 4.25, 13.1 var 1, 15.1 and 16.6, were found to exhibit only one unique peptide. Intriguingly, with the exception of KAP 4.25, all of these carried the N-terminal cysteine modification associated with NTCB-mediated cleavage of peptides. Although a significant number of peptides were found to match the sequence of KAP 4.2 (19 peptides), KAP 4.8 (18 peptides), and R tubulin (21 peptides), none of these were unique (Supplementary Table 1); as a result, these proteins were excluded from the list of positively identified proteins (Table 1). Five nonkeratin proteins also had only a single unique peptide: stem cell factor, kappa casein, tubulincolchicine-phomopsin A, R tubulin 1 and 4. All other keratin and nonkeratin proteins exhibited a minimum of two unique peptides (Supplementary Table 1). Under the chosen conditions, 23% of all unique peptides were identified in both trypsin, and NTCB and trypsin digests, while 22% and 55% were specific to either trypsin, and NTCB and trypsin, respectively (Supplementary Figure 1). Journal of Proteome Research • Vol. 9, No. 6, 2010 2925
research articles
Koehn et al.
Figure 3. Distribution of peptides in identified protein families of the wool cuticle. The percentage of unique peptides for each protein family compared to all 825 unique peptides identified is shown in green (front panel). The average sequence coverage is shown in light blue (middle panel). The percentage of unique peptides compared to the total number of matching peptides within each protein family is shown in red (back panel).
Discussion The cuticle of Merino wool exhibits proteins associated with a variety of cellular functions, including several proteins which are known to be involved in keratin organization and linking. Four desmosomal proteins (plakoglobin, desmoglein, desmoplakin, and plakophilin) were identified, which are associated with linking the intermediate filament networks of neighboring cells.26 Other identified nonkeratin proteins known to play an important role in cellular adhesion and intermolecularlinking were laminin, integrins, actins, and tubulins. However, keratins were by far the most abundant structural proteins, most notably KAPs and IFPs. Overall, 12 out of a possible 17 IFPs were identified, including seven type I and five type II IFPs. Hence, with the exception of K87, all possible IFP type II proteins were identified. While IFPs accounted for over half of all identified peptides, they contributed fewer than one-fourth of all unique peptides, which is attributable to the extensive sequence homology between IFPs (Figure 2). IFP-derived peptides dominated MS-spectra and >80% sequence coverage was obtained for the vast majority of identified IFPs. Hence, it appears that type I and II IFPs are important constituents of the wool cuticle. It is difficult to assess the true abundance of IFPs as the acquired MS-data does not allow for quantitative analysis. However, it is possible that IFPs may not be as abundant as this data suggests. Compared to KAPs, for example, IFPs are relatively large proteins with fewer cysteines and an advantageous distribution of tryptic cleavage sites that facilitate the detection of IFPs by MS. Although the sequence coverage and number of unique peptides for KAPs was comparatively lower, a large number (29) of KAPs were positively identified (Figure 2), covering the UHS (8), HS (17), and HGT (4) categories (Table 1). 2926
Journal of Proteome Research • Vol. 9, No. 6, 2010
The isolation of pure wool cuticle represents a significant challenge, as, unlike the human hair cuticle, which forms several full cuticle layers, the wool cuticle exists exclusively as a single layer surrounding the cortex of the wool fiber. While this study used highly pure cuticle, as determined by SEM analysis, the possibility of a low degree of cortical contamination cannot be completely discounted, especially if, during their dislodgement from the fiber, cuticle flakes retained miniscule cortical appendages that may have escaped detection. Because IFPs are abundant in the cortex, they may thus have contributed to the observation of type I and II IFPs in the cuticle. Nevertheless, the finding that type I and II IFPs represented important constituents of the wool cuticle was indeed congruent with similar observations in human hair.18 These results are consistent with the hypothesis that IFPs are not only crucial for cortical structure, but play an important role in the molecular assembly and rendering of cuticle properties/ structure as well. Fibres from human hair and wool share extensive similarities; hence, it is not surprising to see that many of the proteins identified in this project are also reported to be present in human hair shafts.18 Even so, there are some intriguing differences between some of the proteins identified in our study, and those identified by Lee et al.,18 which are probably attributable to the unique protein composition of the cuticle and the use of NTCB for protein digestion, rather than inherent differences between hair and wool proteins. By comparison with the study on the human hair shaft proteome, the present study succeeded in identifying additional KAPs belonging to KAP families 5, 6, 7, 8, 15, 16, 24, and 26. Interestingly, the authors investigating the hair shaft proteome noted that the apparent difficulty in successfully identifying KAPs was based
research articles
Proteome of the Wool Cuticle on their relatively small size and paucity of tryptic cleavage sites. Despite additional endeavors using both enzymatic and chemical digestion, including the use of CNBr, they were not successful in identifying additional KAPs in the hair shaft according to their report. The strategy used in the present study, specifically targeting cysteine for chemical cleavage with NTCB, clearly succeeded in generating unique peptides, which not only proved crucial for the positive identification of a number of KAPs in the Merino cuticle (Supplementary Table 1), but also generated additional information for the identification of other proteins, including IFPs, by specific cysteine modification and cleavage as depicted in Figure 4. This approach has proven particularly useful in generating additional peptide fragments in protein regions exhibiting significant cysteine content. The identification of cysteine-rich proteins such as UHS and HS KAPs is a classic example of the benefit of chemically induced N-terminal cysteine cleavage compared to tryptic digestion, as all of the unique peptides critical for the identification of KAPs 1.4, 2.3, 2.3 var 1 and var 3, 4-like, 4.19, 13.1 var 1, 15.1, and 16.6 were generated through NTCB-mediated protein cleavage. Because of its comparatively small size, NTCB was able to penetrate the cuticular complexes more easily than large enzyme molecules, which generated access to cysteine-rich regions and facilitated subsequent enzyme treatment with trypsin. Hence, compared to classic tryptic digestion, a combined approach of chemical and enzymatic digestion exhibited synergistic effects, producing additional peptide fragments which ultimately resulted in increased sequence coverage and a higher number of identified proteins. It appears likely that the adoption of this approach would also improve the chances of identifying additional KAPs and other cysteine-proteins if applied to the study of human hair. It has been speculated that the poor extractability of the cuticle of human hair and wool is attributable to the high degree of disulfide and isopeptide cross-linking.10,21,23 The large number of UHS and HS KAPs as well as other cysteine proteins identified in this study affirms the probability that disulfide bonds play a significant role, but it was not possible to directly investigate the presence of isopeptide cross-links in the cuticle. It is interesting to note, however, that known substrates for isopeptide crosslinking in wool and hair such as trichohyalin,27,28 involucrin,29 and loricrin30 and cross-linking enzymes such as transglutaminases31-33 were not detectable in this study under the chosen conditions. These findings differ from those of the human hair shafts where, despite the absence of loricrin and involucrin, trychohyalin was found sporadically, while transglutaminases were found at significant levels. It is possible, however, that these proteins were primarily derived from the medulla, which exclusively exists in the inner cortex and is known to contain both trichohyalin and transglutaminases. Hence, despite existing reports suggesting the presence of considerable amounts of ε-(γ-glutamyl)lysine cross-links in the wool cuticle based on its amino acid composition,19,20 concrete proteomic evidence in support of this hypothesis remains elusive. Further investigation is clearly necessary, but in the absence of commercially available isopeptidases, which can cleave ε-(γ-glutamyl)lysine cross-links34 and thus may render the cuticle more amenable to further digestion, this will likely continue to represent a challenge.
Figure 4. Comparison of characteristic MS/MS spectra for peptides generated using trypsin only and NTCB and trypsin digestion. MALDI (A and B) and ESI MS/MS spectra (C) of unique peptides matched to the sequence E.CVEADSGR.L (A) and R.EAECVEADSGR.L (B and C), respectively, of IFP Type II K83. Spectrum A shows the y and b ion series fragmentation pattern generated by NTCB-mediated N-terminal cysteine cleavage, resulting in an N-terminal NTCB-modified cysteine residue (C*1). In the absence of cysteine cleavage, a tryptic peptide is observed as displayed in spectra B and C which exhibit the fragmentation pattern associated with propionamide (C*2) and dehydroalanine (C*3) modification of cysteine, which are typical of acrylamide alkylation and NTCB-mediated β-elimination, respectively.
Conclusion The results of this study show that the wool cuticle consists of a diverse range of proteins. A majority of these were keratin and nonkeratin proteins which are known to play an important role in cellular structure, with KAPs exhibiting the highest number of identified proteins of any protein category. Until now, the paucity of tryptic cleavage sites combined with the comparatively small size of KAPs and the extensive sequence homology within as well as between KAP families have, in many Journal of Proteome Research • Vol. 9, No. 6, 2010 2927
research articles cases, presented a challenge for the complete characterization of KAPs in both wool and hair. The employment of combined enzymatic and chemical cleavage in this study was particularly successful in the identification of cysteine-peptides, which proved crucial for unambiguous identification of a number of KAPs, but also contributed significantly in the identification of other sulfur/cysteine-rich proteins. In this study, the core protein constituents from the cuticle of Merino wool were successfully identified and thus represent a valuable basis for future studies on how these contribute to fiber and textile quality. Clearly, much remains to be learned about the tissue/sublayer specific expression of keratinous proteins and their interaction with each other as well as nonkeratin proteins in our endeavor to better understand the mechanisms of molecular assembly of wool and hair proteins.
Acknowledgment. Dr. Henning Koehn was supported by a Wool Research Organisation of New Zealand. Inc. and the New Zealand Wool Industry Charitable Fellowship. We also thank Gail Krsinic for her help in cuticle isolation and SEM analysis, Ancy Thomas for preparing samples for tryptic digestion, Dr. Zhidong Yu for providing cDNA sheep sequences, Dr. Duane Harland for the image of the cuticle layers and Anita Grosvenor for critical editing of the manuscript. Supporting Information Available: Supplementary Table 1: list of all identified proteins and corresponding peptide sequences from digests of the wool cuticleThis material is available free of charge via the Internet at http://pubs.acs.org. References (1) Parry, D. A. D. Steinert, P. M. , Intermediate Filament Structure; Springer-Verlag: Heidelberg, Germany, 1995; pp 1-183. (2) McLaren, R. J.; Rogers, G. R.; Davies, K. P.; Maddox, J. F.; Montgomery, G. W. Linkage mapping of wool keratin and keratinassociated protein genes in sheep. Mamm. Geneome 1997, 8 (12), 938–40. (3) Zahn, H.; Wortmann, F.-J.; Hoecker, H. Considerations on the occurrence of loricrin and involucrin in the cell envelope of wool cuticle cells. Int. J. Sheep Wool Sci 2005, 53 (1), 24–36. (4) Rogers, M. A.; Schweizer, J. Human KAP genes, only the half of it? Extensive size polymorphisms in hair keratin-associated protein genes. J. Invest. Dermatol. 2005, 124 (6), vii–ix. (5) Roper, K.; Fohles, J.; Klostermeyer, H. Complete enzymatic hydrolysis of wool and its morphological components. Methods Enzymol. 1984, 106, p. 58–69. (6) Powell, B. C. Rogers, G. E. Differentiation in hard keratin tissues: hair and related structures. In The Keratinocyte Handbook.; Cambridge University Press: Cambridge, 1994; pp 401-436. (7) Langbein, L.; Rogers, M. A.; Winter, H.; Praetzel, S.; Schweizer, J. The catalog of human hair keratins. II. Expression of the six type II members in the hair follicle and the combined catalog of human type I and II keratins. J. Biol. Chem. 2001, 276 (37), 35123–32. (8) Rogers, M. A.; Winter, H.; Langbein, L.; Wollschlager, A.; PraetzelWunder, S.; Jave-Suarez, L. F.; Schweizer, J. Characterization of human KAP24.1, a cuticular hair keratin-associated protein with unusual amino-acid composition and repeat structure. J. Invest. Dermatol. 2007, 127 (5), 1197–204. (9) Rogers, M. A.; Langbein, L.; Praetzel-Wunder, S.; Giehl, K. Characterization and expression analysis of the hair keratin associated protein KAP26.1. Br. J. Dermatol. 2008, 159 (3), 725–9. (10) Rogers, M. A.; Langbein, L.; Winter, H.; Beckmann, I.; Praetzel, S.; Schweizer, J. Hair keratin associated proteins: characterization of a second high sulfur KAP gene domain on human chromosome 21. J. Invest. Dermatol. 2004, 122 (1), 147–58. (11) Rogers, M. A.; Langbein, L.; Winter, H.; Ehmann, C.; Praetzel, S.; Korn, B.; Schweizer, J. Characterization of a cluster of human high/ ultrahigh sulfur keratin-associated protein genes embedded in the type I keratin gene domain on chromosome 17q12-21. J. Biol. Chem. 2001, 276 (22), 19440–51.
2928
Journal of Proteome Research • Vol. 9, No. 6, 2010
Koehn et al. (12) Rogers, M. A.; Langbein, L.; Winter, H.; Ehmann, C.; Praetzel, S.; Schweizer, J. Characterization of a first domain of human high glycine-tyrosine and high sulfur keratin-associated protein (KAP) genes on chromosome 21q22.1. J. Biol. Chem. 2002, 277 (50), 48993–9002. (13) Shimomura, Y.; Aoki, N.; Schweizer, J.; Langbein, L.; Rogers, M. A.; Winter, H.; Ito, M. Polymorphisms in the human high sulfur hair keratin-associated protein 1, KAP1, gene family. J. Biol. Chem. 2002, 277 (47), 45493–501. (14) Rogers, G.; Koike, K. Laser capture microscopy in a study of expression of structural proteins in the cuticle cells of human hair. Exp. Dermatol. 2009, 18 (6), 541–7. (15) Rogers, M. A.; Langbein, L.; Praetzel-Wunder, S.; Winter, H.; Schweizer, J. Human hair keratin-associated proteins (KAPs). Int. Rev. Cytol. 2006, 251, 209–63. (16) Langbein, L.; Rogers, M. A.; Winter, H.; Praetzel, S.; Beckhaus, U.; Rackwitz, H. R.; Schweizer, J. The catalog of human hair keratins. I. Expression of the nine type I members in the hair follicle. J. Biol. Chem. 1999, 274 (28), 19874–84. (17) Powell, B. C. The keratin proteins and genes of wool and hair. Wool Technol. Sheep Breed. 1996, 44 (2), 100–118. (18) Lee, Y. J.; Rice, R. H.; Lee, Y. M. Proteome analysis of human hair shaft: from protein identification to posttranslational modification. Mol. Cell. Proteomics 2006, 5 (5), 789–800. (19) Bradbury, J. H. The amino acid analysis of cuticle-rich material from wool. Text. Res. J. 1960, 30, 128–135. (20) Bradbury, J. H.; Ley, K. F. The chemical composition of wool. XI. Separation and analysis of exocuticle and endocuticle. Aust. J. Biol. Sci. 1972, 25 (6), 1235–47. (21) Bringans, S. D.; Plowman, J. E.; Dyer, J. M.; Clerens, S.; Vernon, J. A.; Bryson, W. G. Characterization of the exocuticle a-layer proteins of wool. Exp. Dermatol. 2007, 16 (11), 951–60. (22) Koehn, H.; Clerens, S.; Deb-Choudhury, S.; Morton, J. D.; Dyer, J. M.; Plowman, J. E. Higher sequence coverage and improved confidence in the identification of cysteine-rich proteins from the wool cuticle using combined chemical and enzymatic digestion. J. Proteomics 2009, 73 (2), 323–30. (23) Zahn, H.; Fo¨hles, J.; Nlenhaus, M.; Schwan, A.; Spel, M. Wool as a biological composite structure. Ind. Eng. Chem. Prod. Res. Dev. 1980, 19 (4), 496–501. (24) Dyer, J. M.; Plowman, J.; Krsinic, G.; Deb-Choudhury, S.; Koehn, H.; Millington, K.; Clerens, S. Proteomic evaluation and location of UVB-induced photomodification in wool. Photochem. Photobiol., B 2010, 98, 118–127. (25) Wessel, D.; Flugge, U. I. A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Anal. Biochem. 1984, 138 (1), p. 141–3. (26) Yin, T.; Green, K. J. Regulation of desmosome assembly and adhesion. Semin. Cell Dev. Biol. 2004, 15 (6), 665–77. (27) Alibardi, L. Fine structure and immunocytochemistry of monotreme hairs, with emphasis on the inner root sheath and trichohyalinbased cornification during hair evolution. J. Morphol. 2004, 261 (3), 345–63. (28) Tarcsa, E.; Marekov, L. N.; Andreoli, J.; Idler, W. W.; Candi, E.; Chung, S. I.; Steinert, P. M. The fate of trichohyalin. Sequential post-translational modifications by peptidyl-arginine deiminase and transglutaminases. J. Biol. Chem. 1997, 272 (44), 27893–901. (29) Candi, E.; Oddi, S.; Terrinoni, A.; Paradisi, A.; Ranalli, M.; FinazziAgro, A.; Melino, G. Transglutaminase 5 cross-links loricrin, involucrin, and small proline-rich proteins in vitro. J. Biol. Chem. 2001, 276 (37), 35014–23. (30) Candi, E.; Melino, G.; Mei, G.; Tarcsa, E.; Chung, S. I.; Marekov, L. N.; Steinert, P. M. Biochemical, structural, and transglutaminase substrate properties of human loricrin, the major epidermal cornified cell envelope protein. J. Biol. Chem. 1995, 270 (44), 26382–90. (31) Thibaut, S.; Candi, E.; Pietroni, V.; Melino, G.; Schmidt, R.; Bernard, B. A. Transglutaminase 5 expression in human hair follicle. J. Invest. Dermatol. 2005, 125 (3), 581–5. (32) Thibaut, S.; Cavusoglu, N.; de Becker, E.; Zerbib, F.; Bednarczyk, A.; Schaeffer, C.; van Dorsselaer, A.; Bernard, B. A. Transglutaminase-3 enzyme: a putative actor in human hair shaft scaffolding. J. Invest. Dermatol. 2009, 129 (2), 449–59. (33) Yoneda, K.; Akiyama, M.; Shimizu, H.; Imamura, S. Transglutaminase 1 in human hair follicle. Exp. Dermatol. 1999, 8 (4), 368–9. (34) Loewy, A. G.; Blodgett, J. K.; Blase, F. R.; May, M. H. Synthesis and use of a substrate for the detection of isopeptidase activity. Anal. Biochem. 1997, 246 (1), 111–7.
PR901106M
