Investigation of the Putative Biological Substrates for Human Kallikrein-Related Peptidase 7

INVESTIGATION OF THE PUTATIVE BIOLOGICAL SUBSTRATES FOR HUMAN KALLIKREIN-RELATED PEPTIDASE 7

Yijing Yu

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Graduate Department of Laboratory Medicine and Pathobiology

University of Toronto

Investigation of the putative biological substrates for human kallikrein-related peptidase 7

Yijing Yu

Doctor of Philosophy

Department of Laboratory Medicine and Pathobiology

University of Toronto

2017 Abstract

Kallikrein-related peptidases (KLKs) are a group of serine proteases widely expressed in various tissues and involved in a wide range of physiological and pathophysiological processes. KLK7, known as stratum corneum chymotryptic enzyme, has been associated to several skin

(patho)physiological conditions, with little existing knowledge regarding its substrates and potential functional roles. Therefore, a degradomics approach was employed to systemically investigate the endogenous substrates of KLK7, in an effort to understand the molecular pathways underlying its action in human skin. We identified several previously known as well as novel KLK7 substrates. Our most promising candidates were validated using targeted quantitative proteomics (selected reaction monitoring assays, known as SRM assays) and in vitro recombinant protein digestion assays. Our study revealed midkine, CYR61 and tenascin-C as novel endogenous substrates for KLK7. Interestingly, some of these substrates (e.g. midkine) were prone to proteolytic cleavage only by KLK7 (and not by other skin-associated KLKs), whereas others (e.g. CYR61 and tenascin-C) could be digested by several KLKs. Furthermore, using melanoma cell lines, we showed that KLK7-mediated cleavage of midkine results in an overall reduction in the pro-proliferative and pro-migratory effect of this growth factor. Next, in order to further validate those putative KLK7 substrates identified by degradomics, we

ii performed sweat proteomics and peptidomics to explore the skin endogenous proteins and peptides. We identified 1,928 unique sweat proteins (861 from the proteomic study and 1,651 from the peptidomic study) and 32,818 endogenous sweat peptides. Several skin proteases and endogenous protease inhibitors were identified in human sweat reflecting the intense proteolytic activity of human skin. The presence of several antimicrobial peptides supports the functional roles of sweat in host defense and innate immunity. More importantly, putative KLK7 substrate peptides (i.e CTGF and Annexin-2) were identified in sweat peptidomics, which further corroborated our degradomics data. In summary, our degradomics approach revealed many novel endogenous substrates for KLK7, which may shed more light on the pathobiological roles of

KLK7 in human skin. Similar systematic substrate screening approaches could be applied to discover the biological substrates of other human proteases.

iii

Acknowledgments

I would like to express my special gratitude to many people for their support of my graduate studies in the Department of Laboratory Medicine and Pathobiology (LMP), University of Toronto.

First, I would like to thank my supervisor Dr. Eleftherios P. Diamandis, our LMP graduate department coordinator, Dr. Harry Elsholtz and the admission committee of LMP for recruiting me to study in this great program. I had a great experience in our program.

I want to thank Dr. Diamandis for his supervision of my Ph.D research project at Mount Sinai Hospital. Thank you for providing me wonderful research learning and working opportunities. Without your support, my work could not have been done successfully. Thank you for your challenging questions and inspiring thoughts during the past meetings. The scientific thinking I learned from you will continue to benefit me throughout my career. I deeply appreciate your great support and help to my graduate life.

I would like to thank my committee members, Dr. David Irwin and Dr. George Yousef. Thank you for your great feedback and advice. Without your support, it would be more difficult for me to complete my Ph.D study. I want to thank my internal defense members, Dr. Vinod Chandran and Dr. George Charames for their generous help in reviewing my Ph.D. thesis and for sitting on my Ph.D. Oral Examination committee. I would like to extend my sincerest thanks to my external examination committee member Dr. Maria Brattsand, for her thoughtful and careful review on my thesis.

I am thankful to all the members of ACDC lab. Thank you for your presence and support in the past years. Specially thanks to: Ioannis Prassas, Carla Muytjens, Antoninus Soosaipillai, Ihor Batruch, Apostolos Dimitromanolakis, Davor Brinc, Natasha Musrap, Azza Eissa, Ana Konvalinka, Hari Kosanam, and Julie Van. I not only have learned useful technical skills, but also gain critical ideas and great help with paper wiring and editing from all of you.

Special thanks to Ferzeen Sammy, Rama Ponda, Conchita Ferrao, Amanda Goddard, Susan Chou and Cindy Todoroff for their great help during my Ph.D. study.

Lastly but most importantly, I am deeply grateful for the support from my family. Especially, I would like to thank my husband, who has accompanied me living out of my home country in the past several years and stayed in Toronto for his PDF training. I want to thank my parents for their great support to my life and education experience. And my son, thank you for pushing me to be a better mom and a better person.

Table of Contents

Acknowledgments...... iv

Table of Contents ...... vi

List of Abbreviations ...... xi

List of Tables ...... xvi

List of Figures ...... xvii

List of Appendices ...... xx

Chapter 1 ...... 1

1.1 Serine Protease ...... 2

1.1.1 General properties ...... 2

1.1.2 Catalytic mechanism, specificity and regulation of serine protease ...... 3

1.1.3 Kallikrein-related peptidases ...... 6

1.2 Kallikrein-related peptidase 7 ...... 9

1.2.1 Discovery of the KLK7 ...... 9

1.2.2 Organization of KLK7 gene and protein sequence ...... 9

1.2.3 Tissue expression and cellular localization ...... 10

1.2.4 Regulation of KLK7 expression and activity ...... 13

1.2.5 KLK7 inhibitors ...... 13

1.3 Skin kallikrein-related peptidases ...... 15

1.3.1 Normal skin structure and function ...... 15

1.3.2 Skin kallikrein-related peptidases and their (patho)physiology ...... 16

1.3.3 Skin phenotype in a KLK transgenic mice model ...... 19

1.3.3.1 Association of KLK7 with skin dysfunction ...... 20

1.4 Putative substrates of kallikrein-related peptidases ...... 21

1.4.1 Regulation of kinin signaling pathway by KLKs...... 22

1.4.2 Semen function ...... 22

1.4.3 Skin-related KLK substrates ...... 23

1.4.4 KLK Substrates involved in innate immunity ...... 24

1.4.5 KLK substrates in the central nervous system ...... 24

1.4.6 Extracellular matrix-related KLK substrates ...... 25

1.4.7 Secreted substrates of KLKs ...... 26

1.4.8 KLK-regulated surface receptors ...... 26

1.4.9 Other KLK substrates ...... 27

1.4.10 KLKs in human biofluids...... 31

1.5 Strategies for substrates finding ...... 31

1.5.1 Proteomic identification of protease cleavage sites (PICS) ...... 32

1.5.2 Protein topography and migration analysis platform (PROTOMAP) ...... 33

1.5.3 Cell surface protease degradomics ...... 33

1.5.4 Terminal amino isotopic labelling of substrates (TAILS) ...... 34

1.6 Rationale, hypothesis and objectives ...... 37

1.6.1 Rationale ...... 37

1.6.2 Hypothesis ...... 38

1.6.3 Objectives ...... 38

Chapter 2 ...... 39

2 Expression and Characterization of kallikrein-related peptidase-7 ...... 40

2.1 Introduction ...... 40

2.2 Materials and Methods ...... 41

2.2.1 Cloning, production and purification of recombinant KLK7 ...... 41

2.2.2 Characterization of recombinant KLK7 ...... 42 vii

2.2.2.1 SDS-PAGE and Western blotting ...... 42

2.2.2.2 Gelatin zymography ...... 42

2.2.2.3 Fluorogenic AMC substrate profiling ...... 43

2.2.2.4 pH profiling of KLK7 ...... 43

2.2.2.5 Effect of cations on KLK7 activity ...... 43

2.2.3 Tissue expression patterns of KLK7 in humans ...... 44

2.3 Results ...... 45

2.3.1 Yeast expression of pPIC9-KLK7 ...... 45

2.3.2 Production of recombinant mature form of KLK7 ...... 48

2.3.3 Characterization of recombinant mature form of KLK7 ...... 50

2.3.4 Analysis of KLK7 expression pattern in human tissues ...... 54

2.4 Discussion ...... 56

2.5 Author Contributions ...... 57

Chapter 3 ...... 58

3 Novel Biological Substrates of Kallikrein-related peptidase 7 Identified Through Degradomics ...... 59

3.1 Introduction ...... 59

3.2 Materials and Methods ...... 61

3.2.1 Reagents, cells and antibodies ...... 61

3.2.2 Preparation of cell culture media for proteomic analysis ...... 61

3.2.3 Selected reaction monitoring assay ...... 63

3.2.4 In vitro recombinant protein digestion assay ...... 64

3.2.5 Cell proliferation assay...... 64

3.2.6 Migration and invasion assay ...... 64

3.2.7 Microarray profiles ...... 65

3.2.8 Data analysis ...... 65

viii

3.3 Results ...... 66

3.3.1 Preparation of cell culture media for proteomic analysis ...... 66

3.3.2 Validation of candidates with an SRM assay in WM9 cultures ...... 69

3.3.3 Validation of midkine, tenascin-C and CYR61 as novel substrates for KLK7 by in vitro recombinant protein digestion assay...... 73

3.3.4 In vitro digestion of midkine, CYR61 and tenascin-C by other skin KLKs ...... 73

3.3.5 In vitro monitoring of proteolytic processing of substrates by KLK7 ...... 76

3.3.6 Biological effects of KLK-mediated cleavage of new substrates ...... 76

3.4 Discussion ...... 81

3.5 Author Contributions ...... 84

Chapter 4 ...... 85

4 Proteomic and peptidomic analysis of human sweat with emphasis on proteolysis ...... 86

4.1 Introduction ...... 86

4.2 Materials and Methods ...... 88

4.2.1 Sample collection and preparation ...... 88

4.2.2 Strong cation exchange chromatography ...... 91

4.2.3 Mass spectrometry and data analysis ...... 91

4.2.4 Bioinformatic analysis ...... 92

4.2.5 SDS-PAGE and Western blotting-SDS-PAGE ...... 92

4.2.6 Immunodetection of KLKs in sweat (ELISA) ...... 92

4.2.7 In vitro recombinant protein digestion assay ...... 93

4.3 Results ...... 94

4.3.1 Characterization of sweat proteome ...... 94

4.3.2 Characterization of human sweat peptidome ...... 101

4.3.3 Sweat antimicrobial peptides ...... 105

4.3.4 Detection of KLKs in human sweat ...... 108

4.3.5 Detection of putative KLK7 substrate peptides ...... 110

4.4 Discussion ...... 115

4.5 Author Contributions ...... 120

Chapter 5 ...... 121

5 Summary and Future Directions ...... 122

5.1 Summary ...... 122

5.2 Future direction ...... 124

References: ...... 128

Appendices ...... 154

A.1 Tables and Figures ...... 154

A.2 Ongoing Studies ...... 169

A.2.1 Characterization of mature form of KLK13 ...... 169

A.2.2 Identification of putative substrates of KLK9 ...... 170

List of Abbreviations

FA formic acid

2D-DIGE two-dimensional difference gel electrophoresis

AD atopic dermatitis

AMPs anti-microbial peptides

AMC 7-amino-4-methylcoumarin

Asp aspartic acid

BK Bradykinin

CAMP collection of anti-microbial peptides

CAN Acetonitrile

CCN cyr61, ctgf, nov

CDSN Corneodesmosin

CE corneocyte envelope

COFRADIC combined fractional diagonal chromatography

CSF cerebrospinal fluid

CRISPR clustered regularly interspaced short palindromic repeats

CTGF connective tissue growth factor

CVF cervical vaginal fluid

CYR61 cysteine rich angiogenic inducer 61

DCD Dermecidin

DFS diseases-free survival

DSC1 desmocollin 1

DSCs desmocollins

DSG1 desmosoal protein desmoglein 1

DSGs desmogleins

DTT Dithiothreitol

ECM extracellular matrix

ELISA enzyme-linked immunosorbent assay

EphB2 ephrin type-B receptor 2

FBS fetal bovine serum

FPLC fast protein liquid chromatography

FPR false positive rate

GDF 15 growth differentiation factor 15 hCAP18 human cationic antimicrobial protein 18

His histidine

HPGs hyperbranched polyglycelrols

HPLC high performance liquid chromatography

HSE human skin equivalent hTGF beta-1 human transforming growth factor-beta 1

ICAT isotope-Coded Affinity Tags

xii

ICE IL-1 beta-converting enzyme

IGFBPs insulin-like growth factors binding proteins

IL-1beta interleukin 1beta iTRAQ isobaric Tags for Relative and Absolute Quantification

KLKB1 plasma kallikrein

KLKs kallikrein-related peptidases

LC liquid chromatography

LC-MS/MS liquid chromatography-tandem mass spectrometry

LEKTI lmphoepithelial Kazal-type-related inhibitor

LK low molecular weight kininogen

MBP myelin basic protein

MDK midkine

Met methionine

MIF macrophage migration inhibitory factor

MMPs matrix metalloproteinases

NHEKs normal human epidermal keratinocytes

NOV nephroblastoma overexpressed

NS netherton syndrome

ORF open reading frame

OS overall survival

xiii

PARs proteinase-activated receptors

Phe phenylalanine

PICS proteomic identification of protease cleavage sites

PROTOMAP protein topography and migration analysis platform

PSA prostate-specific antigen

PSL positional scanning substrate libraries

RCL reactive-center loop

SB stratum basale

SC stratum corneum

SCCE stratum corneum chymotryptic enzyme

SDS-PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis

Ser serine

SERPINs serine protease inhibitors

SFTI-1 sunflower trypsin inhibitor

SG stratum granulosum

SILAC stable isotope labeling by amino acids

SKALP skin-derived antileukoproteinase

SLPI secretory leukocyte proteases inhibitor spink serine protease inhibitors of kazal-type

SRM selected reaction monitoring

xiv

SS stratum spinosum

STS steroid sulfatase

SVM support vector machine

TAILS terminal amino isotope labelling of substrates

TFA trifluoroacetic acid

TGF beta-1 transforming growth factor-beta 1

TGs transglutaminases

TNC tenascin-C

TPA 12-O-tetradecanoylphorbol-13-acetate

Tyr tyrosine. uPAR urokinase-type plasmainogen activator receptor

VEGF vascular endothelial growth factor

XLI X-linked ichthyosis

List of Tables

Table Title Page

1.1 Summary of the putative physiological substrates of each kallikrein 28 (insufficient data exist for KLKs 9, 10 and 11)

1.2 Overview of the techniques commonly used for protease substrate 35 identification

2.1 Kinetic parameters for the hydrolysis of synthetic AMC substrates by KLK7 51

3.1 Candidate substrates of KLK7 based on the screening results from melanoma 70 and HaCaT cell lines

3.2 Selected reaction monitoring assay validation of specific peptides in cell 72 culture

4.1 Selected list of proteases and protease inhibitors identified in sweat proteome 99

4.2 KLKs identified in this proteomics study 111

4.3 Previous known putative KLKs substrates were identified in human sweat 112

4.4 KLK7 substrates peptides found in substrate screening experiment (Chapter 3) 113 were detected

xvi

List of Figures

Figure Title Page

1.1 Catalytic mechanism of serine proteases 5

1.2 Genetic and proteomic characterization of kallikrein-related 8 peptidases

1.3 Six transcripts of the KLK7 gene 12

1.4 Physiological KLKs substrates in various processes 30

1.5 Summary of current MS-based methods towards identification 36 of the substrate repertoire of a protease

2.1 Analysis of pPIC9-KLK7 46

2.2 Gene sequencing result of pPIC9-KLK7 47

2.3 Characterizations of purified recombinant active of protein 49 KLK7.

2.4 Regulation of the activity of KLK7 by pH 52

2.5 Regulation of the activity of KLK7 by different cations 53

2.6 Expression pattern of KLK7 in 8 types of human normal and 55 cancer tissues

3.1 Workflow of our degradomics approach and criteria for 67 candidate selection

3.2 The number of proteins identified after applying the selection 68 criteria in three cell lines samples

3.3 In vitro proteolyic processing of substrate candidates by KLK7 74

xvii

3.4 processing of In vitro proteolyic substrate candidates by other 75 KLKs. KLK5, 7, 8, 13 or KLK14 was tested

3.5 SRM-based time course analysis of KLK7-mediated protein 78 cleavage

3.6 Functional studies of MDK and KLK7-treated MDK (MDKT) 79 in the WM9 cell line

3.7 A, expression pattern of MDK in 8 types of human normal and 80 cancer tissues were analyzed from publicly available microarray database. B, expression pattern of MDK and KLK7 in normal skin and melanoma

4.1 Workflow for sweat proteomics and peptidomics sample 90 preparation

4.2 (A)Venn diagrams of total proteins identified from female and 96 male sweat pools. (B-F) Overlap of our sweat proteome with previously published sweat proteomic datasets

4.3 Gene ontology classification of proteins identified by sweat 97 proteomics

4.4 Tissue specificity of proteins identified in sweat proteomics 98 study

4.5 Venn diagram of peptides (A) and proteins (B) identified from 102 female and male sweat pools. (C) Venn diagram of proteins found in the sweat proteome and peptidome

4.6 Gene ontology analysis for proteins identified in sweat 103 peptidomics samples

4.7 Tissue specification of proteins identified in sweat peptidomics 104

xviii

study

4.8 Protein sequence coverage map and endogenous proteolytic 107 map of dermcidin (DCD)

4.9 Western blotting analysis of KLKs in human sweat 109

4.10 In vitro proteolyic processing of CTGF by KLK7 114

xix

List of Appendices

Content Title Page

Table A1 Sweat proteases identified in this study 154

Table A2 Sweat protease inhibitors identified in this study 157

Table A3 Selected Sweat identified dermcidin peptides 159

Table A4 Corneodesmosomes proteins identified in this study 162

Table A5 Cystatin proteins identified in this peptdiomics study 164

Table A6 Kinetic parameters for the hydrolysis of synthetic AMC substrates by 171 KLK13

Figure A1 The number of proteins identified after applying the selection criteria 165 in human normal keratinocyte cell (HaCaT) secretome as the procedure described in Fig 3.1

Figure A2 Expression pattern of substrate proteins in 8 types of cancer and 166 normal tissues

Figure A3 Characterization of purified recombinant active of protein KLK13. 172

Figure A4 Catalytically active KLK13 was detected in sweat with immune- 173 capture activity assay

Chapter 1

Literature Review

Some sections of this chapter were published in Biological Chemistry.

Yijing Yu, Ioannis Prassas, Eleftherios P. Diamandis.

“Putative kallikrein substrates and their (patho)biological functions.”

Biol Chem. 2014 Sep; 395(9): 931-43.

1.1 Serine Protease

1.1.1 General properties

Proteases and protease inhibitors account for over 2% of the number of human gene (1). All proteases are classified based on their sequence and structure similarities, and a great summary of the relationships among all human proteases can be found in the MEROPS database (2). In this classification system, protease clans are defined based on their catalytic mechanism and protease families are grouped based on their most common ancestry. As of now (Sept 2016, version 10.0), 1,208 known and putative human peptidases and an additional 241 non-peptidase homologues are listed in MEROPS. Based on their catalytic mechanisms, these proteases can be grouped into five classes:

1) aspartic acid proteases (N=322),

2) metalloproteinases (N=291),

3) cysteine proteases (N=322),

4) serine proteases (N=393), and

5) threonine proteases (N=50).

As the names indicate, the key catalytic residue for cysteine proteases, serine proteases and threonine proteases, are cysteine, serine, and threonine, respectively. Among all the classes, serine proteases represent the largest proteolytic class, which contains 13 clans and 40 families. Collectively, they account for more than one-third of the known proteases. In all serine proteases, the nucleophilic serine is located in the active site of enzyme, and it attacks the substrate carbonyl carbon atom to form a tetrahedral intermediate (3). The nucleophilicity of serine is dependent on the catalytic triad, which in serine proteases is typically composed of Aspartic acid (Asp or D), Histidine (His or H) and Serine (Ser or S) (4). Among all serine proteases families, the S1 family of the PA clan superfamily contains the largest number serine proteases. A majority of them exhibit trypsin-like activity with preferred cleavage site after arginine (Arg or R) or lysine (Lys or K). Only a few of them have chymotrypsin-like and elastase-like specificity. Chymotryptic-proteases can cleave substrate at the carboxyl side of amino acids such as

2 phenylalanine (Phe or F), tyrosine (Tyr or Y) and tryptophan (Trp or W), while elastase-like proteases proteolysis cleavage site after alanine (Ala), glycine (Gly) and valine (Val) (5,6).

1.1.2 Catalytic mechanism, specificity and regulation of serine protease

The majority of (chymo)trypsin-like serine proteases are produced as a typically pre-pro-active- form peptide. The pre-peptide is signaling peptide which allows the pro-active-form peptide secreted out of cells. The pro-peptides usually consist of 5-7 amino acids (excepting a 37 aa pro- peptide for KLK5), and cleavage of these peptides occurs after a specific residue (i.e. Arg or Lys). The release of pro-peptides causes a conformational change within the protease, which facilitates the enzyme-substrate binding. For most of the S1 family clan PA serine proteases, three amino acids (serine (i.e. S195), histidine (i.e. H57) and aspartic acid (i.e. D102)) form a classical catalytic triad with aspartic acid in the active pocket. As shown in Figure. 1.1, in serine proteases, two tetrahedral intermediates occur during the proteolytic reaction. Generally, the hydroxyl oxygen atom of serine from the proteases attacks the carbonyl of the substrates peptide bond, by utilizing histidine (eg. H57) as a general base. A positively charged pocket known as the oxyanion hole exists in the active site, which is contributed by the backbone nitrogen atoms of amino acids (i.e. Gly193 and Ser195). A new acyl-enzyme intermediate and H57 help to stabilize the new N-terminus. The free polypeptide fragment is replaced and attacked by a water molecule. The oxyanion hole of the second intermediate collapses and generates a new C- terminus in the substrate (4,7).

The substrate specificities of serine proteases can be determined by the interaction between the enzyme‟s residue (named as “S”) with the substrate‟s amino acid (named as “P”). According to Schechter and Berger nomenclature, the enzyme‟s amino acids located in the active pocket are designated as Sn-S1, S1‟-Sn‟, while corresponding substrate residues are designated as Pn-P1, P1‟-Pn‟ (8). Generally, based on the S1 binding site, S1 family PA proteases can exhibit trypsin, chymotrypsin or elastase-like activity, respectively. For trypsin-like proteases, they have a negatively charged S1 substrate-binding pocket, which facilitates the interaction with positively charged amino acid Arg and Lys on substrates. For chymotrypsin-like enzymes, they contain a hydrophobic S1 site, and thus an aromatic or bulky non-polar P1 residues (i.e. Tyr, Phe or Trp)

3 on substrates are preferred. As compared to chymotrypsin-like enzymes, elastase-like proteases have an opposite substrate specificity, as bulky amino acids at S1 on protease interact with P1 hydrophobic amino acids on substrates (5,6).

The activities of serine proteases are regulated at various levels, such as transcriptional, post- transcriptional and post-translational levels (9). As mentioned above, all serine proteases are synthesized as inactive precursors (as zymogens). The activation of zymogen occurs when the proteases reach their desired sites of proteolytic action. In addition, the activities of these serine proteases are also controlled by endogenous protease inhibitors (10). Based on the mechanisms of action, the inhibitors are divided into three groups: canonical inhibitors, non-canonical inhibitors and serine proteases inhibitors (i.e. serpin) (11). For canonical inhibitors, an exposed convex loop from the inhibitors tightly binds to the active site of the proteases and then gets cleaved by the proteases. For non-canonical inhibitors, they not only interact with the active site of proteases via their N-terminus, but also have a secondary inhibitory site that is located outside of the active site of proteases. Serpins also bind to the proteases; however, the conformation of these inhibitors gets changed once their peptides are cleaved by the proteases (11,12).

Figure 1.1 Catalytic mechanisms of serine proteases. First, once the polypeptides substrate binds to the surface of the serine proteases, the scissile bond is inserted into the active site of the enzyme, which results in the generation of a tetrahedral intermediate. Second, the scissile bond gets broken, leading to the formation of an acyl-enzyme intermediate. Third, water comes in to the reaction, which replaces the N-terminus of the cleaved peptide, and attacks the carbonyl carbon. Fourth, the bond between the oxygen of the water and the carbon of serine is formed. Overall, this generates another tetrahedral intermediate. In the final reaction, the bond between the serine and the carbonyl carbon gets broken, and a new bond forms when oxygen of serine attacks the hydrogen of histidine. As a result, the C-terminus of the polypeptide is now ejected. Figure modified from https://en.wikipedia.org/wiki/Serine_protease

1.1.3 Kallikrein-related peptidases

Since the first report by Eugen Werle in 1934 about an “unknown substance present in large amounts in the pancreas of humans” (he named it kallikrein by derivation from the Greek word for pancreas), research on the kallikrein has grown exponentially. At present, the PubMed query “Kallikreins” retrieves more than 30,000 research papers and 3,000 reviews related to these enzymes. The human kallikrein-related peptidases (KLKs) belong to the serine protease family S1 within the PA clan. The 15 KLK genes are located on the human chromosome 19q13.3–q13.4, which is distant from the plasma kallikrein (KLK1B) that is located in chromosome 4q35 (9). All KLK genes have five coding exons of similar size, and a conserved intron phase pattern (9). These enzymes are initially produced as inactive pre-pro-enzymes, which subsequently get activated through proteolytic removal of their N-terminal pre-pro peptide sequences. Upon activation, the activity of mature KLKs is tightly regulated by various endogenous inhibitors, such as metal ions, lymphoepithelial kazal-type inhibitor, serpins and macroglobulins (10).

Sequence comparison of the 15 KLKs reveals about 45-65% similarity among the classic KLKs (KLKs 1-3) and around 30-40% identity among the remaining members of the family (KLK4- 15). The most conserved regions of sequences are those adjacent to the catalytic domains of KLKs, which are mainly formed by three amino acids: histidine (His), aspartic acid (Asp) and serine (Ser) (13). The crystal structures of several human KLKs, such as KLKs 1, 3, 4, 5, 6 and 7, were recently resolved, which provide the inspiring information regarding the distinct regulatory, activation and inhibitory mechanisms of each KLK (14-18).

Within the KLK family, KLK1 was initially known for its hypertensive role(19). Similar to plasma kallikrein (KLKB1), KLK1 is an essential component of the kallikrein-kinin system, which could regulate blood pressure (13). While from a clinical perspective, prostate specific antigen (PSA), which is KLK3, is undoubtedly the most studied KLK (an established marker for the prostate cancer). Other than PSA, it has been demonstrated that many KLKs are also aberrantly expressed in a plethora of cancer types (e.g. ovarian, breast, colon, lung), and thus it is suggested that these KLKs can be potentially utilized as single or combined cancer markers (20,21). Furthermore, it is now established that KLKs participate in a wide range of physiological and pathophysiological processes, such as skin desquamation, neuron degeneration,

6 tissue remodeling and wound healing (21-23); these findings were largely based on the results of recently developed KLK transgenic and knockout models (24-26).

Figure 1.2 Genetic and proteomic characterization of human kallikrein-related peptidases. A): Localization of KLKs in the human chromosome. The KLK locus is located on chromosome 19q 13.4. Each arrow represents for a certain KLK gene. B): Using the longest transcript (protein coding) of KLK7 as an example, 5 coding exons (yellow boxes) and 4 intervening introns with a conserved intron phase pattern (I, II, I, 0). Most of the KLKs have one or two non-coding exons. C): KLKs are synthesized into conserved precursors which contain a pre-peptide (secreted signal sequence), a pro-peptide and serine proteases domains (the active form of serine proteases, containing catalytic triad). D): Tertiary and primary structure of KLK7. Left, the overall structure of KLK7 with the AAF-CMF inhibitor (18). Catalytic triad (H70, D112 and S205) is depicted in the S1 pocket. Right, Sequence alignment of KLK4, KLK5, KLK6 and KLK7 (27). Conserved residues are shown in red boxes, while catalytic triad is indicated in green box with an asterisk.

1.2 Kallikrein-related peptidase 7

1.2.1 Discovery of the KLK7

Kallikrein-related peptidase 7 is coded by gene KLK7, also known as SCCE (skin stratum corneum chymotryptic enzyme) or PRSS6, was first reported in 1991 by Egelrud group (28). Egelrud found a chymotrypsin-like protease extracted from skin stratum corneum which was associated with degradation of the desmosomal protein desmoglein 1(DSG1), and the protease was further studied and named as “stratum corneum chymotryptic enzyme” (28,29). The first cDNA study of this gene was extracted and purified from plantar corneocytes, while murine KLK7 cDNA was obtained from the mouse tail by the same group (30). The genomic organization of KLK7 was reported by our group, through mapping with chromosomal localization and sequence homology analysis (31).

1.2.2 Organization of KLK7 gene and protein sequence

KLK7 is located between KLK6 and KLK8 on chromosome 19q13.3-13.4 and is transcribed from telomere to centromere (Fig. 1.2). Six transcripts of KLK7 have been reported, which are named as KLK7-(001-005) and KLK7-201 in the Ensembl database (http://useast.ensembl.org/index.html). Four of the transcripts (i.e. KLK7-001, -003, -004 and - 201) contain open reading frame (ORF) which can lead to protein translation, while the other two mRNAs (i.e. KLK7-002 and -005) go through nonsense-mediated decay and no protein production (Fig 1.3). Although KLK7-001 and KLK7-201 transcripts have the different 5‟ UTR (untranslated region), they encode the same isoform of KLK7 protein, which is also the longest form of KLK7 (253aa). KLK-003 has a different 5‟ UTR, generating a shorter isoform (181aa) while KLK-004 not only has a different 5‟ UTR, but also lacks a coding exon, resulting in the shortest isoform (69aa). Here KLK7-001 is used as an example to illustrate the gene structure of KLK7: it contains 6 exons (5 coding exons + 1 non-coding exon), which is then translated into a classical KLK pre-pro-serine proteases domain protein. The pre-peptide (the first 1-22aa) is a secreted signaling peptide, which can lead the transportation of the KLK precursor from cytoplasm to extracellular region of the cells. The pro-peptide (23-30aa) maintains the latency of KLK precursor, and is cleaved to generate the active-form KLK7 (30). The serine protease

domain (31-224aa) of KLK7 contains the characteristic catalytic triad (H70, D112 and S205), while a predicted N-glycosylation site is located at position 246 of KLK7 (30). Through sequence alignment, KLK7 shows a 40-60% similarity with other KLKs (31). As compared to other KLKs (i.e. KLK4, 5, 6), crystal structure studies revealed that the most distinguishing features of KLK7 are the short 70-80 loop and the unique S1 pocket. Asn189 replaces Asp189 in the S1 pocket of KLK7, and Ala190 (unlike Ser190 in KLK4, KLK5 and KLK6) exists in KLK7, which make the S1 pocket of KLK7 hydrophobic. Thus, the S1 pocket of KLK7 prefers P1 aromatic acid tyrosine (32).

1.2.3 Tissue expression and cellular localization

Earlier studies of KLK7 only focused on the skin system (28). Later studies suggested that KLK7 was also expressed in salivary gland, uterus, thymus, brain, thyroid, trachea, placenta, testis, kidney and cerebellum using RT-PCR assays (31). According to human protein atlas, KLK7 protein was found to be expressed in the bone marrow, lymph node, tonsil, spleen, skeletal muscle, kidney and skin (http://www.proteinatlas.org/; version 15). Our group had shown that KLK7 was detected in biofluids such as breast milk, cervical vaginal fluid (CVF), cerebrospinal fluid (CSF), ovarian ascites, breast cyst fluid, saliva and amniotic fluid (33). Moreover, KLK7 was identified in the stratum granulosum of skin by immunohistochemical staining (34,35).

In addition KLK7 is expressed in several types of cancer tissues, including cervical, head and neck cancer, ovarian, pancreatic and skin cancer (13). For instance, KLK7 was reported to be significantly elevated in ovarian tumor (36-39), and served as an unfavorable prognostic marker for overall survival (OS) and disease-free survival (DFS) of ovarian cancer patients (40). High expression of KLK7 was identified in squamous cervical cancers by several groups (41,42), and high serum KLK7 level correlated with poor prognosis of cervical cancer patients (43). Aberrant expression of KLK7 was also reported in breast cancer samples, although it is still controversial regarding KLK7 as an unfavorable prognostic biomarker in breast cancer (44-49). Moreover, KLK7 was found with increased expression in colon cancers and its expression was significantly associated with shorter OS and DFS (50-54). Over-expression of KLK7 was observed in pancreatic adenocarcinomas and associated with excessive degradation of E-cadherin and

10 desmoglein 2 (55,56), and the increased shedding of urokinase-type plasminogen activator receptor was associated with elevated KLK7 (57). KLK5, 7, 8 and 10 were reported with elevated expression level in oral squamous cell carcinoma(58). Interestingly, decreased KLK7 expression was reported in lung cancer and prostate cancer samples (59-61), and loss of KLK5 and KLK7 in prostate cancer tissues were correlated with worse disease-specific and OS (62). Thus, the role of KLK7 in the pathogenesis of certain cancers is complex and needs to be further investigated.

Figure 1.3 Six transcripts of the KLK7 gene. The start site is labeled with arrow and the stop codon with a stop sign . Grey blocks represent coding exons, while white blocks are non- coding exons. Transcript KLK7-005 doesn‟t contain coding exon, which results in no protein being. In transcripts KLK7-201 and KLK7-001, coding exons 2, 3 and 5 are important for proteolytic activity as they code the amino acids creating the catalytic triad (H70, D112 and S205). The figure is modified from the Ensembl database.

1.2.4 Regulation of KLK7 expression and activity

Earlier studies suggested that several KLKs (i.e. KLK1-3) are expressed in certain tissues and regulated by hormones (63). However, little is known about transcriptional regulation of skin KLKs, including KLK7. Morizane et al reported that IL-4 and IL-13 induced KLK7 mRNA expression in normal human epidermal keratinocytes (NHEKs) (64). Calcium, vitamin D derivatives and retinoic acids also could induce expression of KLK5 and KLK7 in keratinocytes (65).

Activation or degradation of KLKs can control their activity. The zymogen-form of KLKs can be activated through proteolytic cleavage by several other proteases, including metallopeptidases and meprin and members of the KLK family (66,67). For instance, meprins have been reported to proteolyze the pro-KLK8 (67). In skin, a KLK activation cascade has been hypothesized: KLK5 initializes the cascade by auto-activation and then activates the downstream pro-KLKs, such as pro-KLK7, pro-KLK8 and pro-KLK14; activated KLK14 can further activate pro-KLK5 which results in a positive feedback (68). In addition to auto-activation, several KLKs, such as KLK4, 6 and 12, can undergo autolysis that result in loss of their own activities.

The optimal activities of KLKs are dependent on pH (e.g. KLK7 exhibits the highest activity at basic pH) (16). The physiological pH of the skin surface is around 5.5, which causes around 9- fold decrease in the activities of KLKs compared the optimized in vitro conditions. Elevated pH on skin surface was reported in atopic dermatitis patients, which may contribute to the aberrant activities of some serine proteases (e.g. KLK7 and KLK11) in these patients (69,70). Moreover, the activity of KLKs has been reported to be regulated by metal ions. It has been demonstrated that Cu2+ and Zn2+ inhibit KLK7‟s activity with fluorogenic AMC (7-amino-4-methylcoumarin) substrate (71).

1.2.5 KLK7 inhibitors

The studies on Netherton Syndrome, KLKs and serine protease inhibitors of kazal-type (spink) have dramatically expanded our knowledge on skin biology and pathobiology. Netherton syndrome is rare genetic skin disease, caused by SPINK5 mutation, which resulting in loss of the

13 function of the serine protease inhibitor Lymphoepithelial Kazal-type-related inhibitor (LEKTI). LEKTI is expressed in the spinous and granular layer of the epidermis (72). Full-length LEKTI inhibits the proteolytic activity of trypsin, plasmin, subtilisin A, cathepsin G, and human neutrophil elastase (10,73). Certain recombinant LEKTI fragments inhibit the activities of several KLKs including KLK7, while LEKT2 and SPINK6 selectively inhibit tryptic proteases (74).

SERPINs (i.e. serine protease inhibitors or classified inhibitor family I4) are the largest and most broadly distributed superfamily of protease inhibitors. They irreversibly inhibit their specifically targeted proteases via a conformational change, which disrupts the active site of proteases. In humans, 36 serpins have been identified (75), and 5 of them inhibit the activities of certain KLKs including KLK7. All 5 of these SERPINs are involved in skin physiology and pathobiology (75), including SERPINA1, SERPINA3, SERPINA, SERPINA5, SERPINE1 and SERPINF2. For example, SERPINA1 was reported as a potential cancer biomarker of cutaneous squamous cell cancer (76), while SERPINA3 contributes to skin wound repair and its degradation by neutrophil elastase was associated with non-healing of human wound (77). Moreover, Jendrny et al synthesized a library of inhibitory compounds by fusing reactive-center loop (RCL) of serpins to sunflower trypsin inhibitor (SFTI-1), and found that three of these compounds effectively inhibited the activities of KLK7 at the nanomolar level (78). Recently, de Veer et al designed several KLK7 inhibitors by manipulating the sequence of SFTI, and the best synthesized inhibitor displayed very high potent (to picomole) against KLK7. Moreover, the in vitro study with skin tissue model indicated that this KLK7 inhibitor alone could sufficiently block the corneocyte shedding(79).

In addition to serpins, two members of the trapping family: secretory leukocyte proteases inhibitor (SLPI; alternative name: antileukoproteinase) and elafin (alternative name: trappin-2), have been reported to inhibit the activity of KLK7 (80).

Strategies such as high throughput screening, in silico screening, de novo design followed by experimental screening have been used to identify novel proteases inhibitors. Experimental compounds screening was carried on KLK5 with the use of reported public chemical databases, and was able to identify compounds which inhibit the activities of KLK5, KLK7 and KLK14

(81). A later study by the same research group revealed several new small-molecule inhibitors for KLK5, KLK7 and KLK14, which were designed through the extensive structure and ligand- based mining of the ChemBridge Library (82). In addition, through the tailored de novo chemical synthesis, KLK7 inhibitors were identified, such as isomannide derivatives and nitrogen- containing heterocyclic compounds. IC50 of these two compounds showed that these inhibitors act at the nanomolar level (83).

1.3 Skin kallikrein-related peptidases

1.3.1 Normal skin structure and function

The skin is the largest and most visible organ of the human body. It accounts for almost of 15% of the body weight and serves as the primary defense against a challenging environment, chemical and physical damages, UV-radiation and pathogens (84). Three major layers, the epidermis, dermis and subcutaneous tissues, compose the skin (85,86). Mature epidermis is composed of four layers (from outer to inner side): stratum corneum (SC), stratum granulosum (SG), stratum spinosum (SS) and stratum basale (SB) (86). Keratinocytes are the major cell type in the epidermis, and additional cell types such as melanocytes, Langerhans cells, Merkel cells and adjacent epithelial cells are also found in the epidermis layer. Furthermore, three appendages (i.e. sweat glands, the pilosebaceous follicles and the nails) are present in epidermis. The dermis is located between subcutaneous and dermis-epidermal junction, and provides nutrients and circulatory support, and the majority of cells in this layer are fibroblasts, while subcutaneous tissue mainly contains fat cells (adipocytes).

The uppermost epidermal layer is the stratum corneum and forms the first line of skin barrier, which is renewed every 2-4 weeks. This renewing process is controlled by an ingenious differentiation program of keratinocytes (87,88). Keratinocytes originating from the stratum basal layer proliferate upwards into stratum spinous and granular layers, which results in an upward migration of the formed cells. During the shift of keratinocytes to corneocytes (in the stratum corneum), the plasma membranes become embedded with the cross-linked corneocyte envelope (CE), and the cellular keratin intermediate filaments is aggregated, leading to the flattened squamous cell cytoskeleton. Keratinocytes eventually undergo the terminal

15 differentiation process into the non-viable cells known as corneocytes (87). Corneocytes are a special type of “dead” cells that don‟t contain nuclei, sub-cellular organelles or protein synthesis(89). Shedding off the coenocytes is the final step of stratum corneum renewal process, which is also known as “desquamation”.

Constant assembly and disassembly of desmosomes is essential for desquamation (90). Desmosomes are an adhesive intercellular junction that maintains cell-cell contact and adhesion. Under electron microscopy, desmosomes are electron-dense discs with a diameter of 0.2-1µm (91). In the stratum corneum, the modified desmosomes, known as corneodesmosomes, are generated during terminal keratinocyte differentiation, which occurs mainly through crosslinking specific proteins (i.e. desmogleins) by transglutaminases (TGs) (87). The (corneo)desmosomes mainly consist of three family proteins: 1) the cadherins (desmogleins (DSGs) and desmocollins (DSCs)), 2) the armadillo proteins (plakoglobin, plakophilin), and 3) the plakins (desmoplakin, envoplakin, periplakin and plectin). In addition, a unique adhesion protein, called as corneodesmosin (CDSN), is also expressed in corneodesmosome (92). In human skin, 4 desmogleins (DSG1-4) and 3 desmocollins (DSC1-3) are identified (92). The extracellular domains of the DSGs and DSCs mediate the cell-cell adhesion, while the cytoplasmic tails of those cadherins bind to other desmosome proteins (i.e. armadillo, plakin family proteins and keratin) to maintain the cellular cytoskeleton (92).

Proteases and other factors (i.e. proteases inhibitors, Ca2+, lipid) are required for at least of 3 processes of skin differentiation: cornified-envelope formation (activation of TGs), degradation of cellular components (i.e. loss of nuclei and mitochondria by lysosomal proteases, calpains and caspases) and desquamation (i.e. proteolysis of corneodesmosome proteins by stratum corneum proteases) (92-95) The enzymes that are responsible for desquamation process are not yet well known, although several possible candidates (e.g. KLK5 and KLK7) have been indicated (87).

1.3.2 Skin kallikrein-related peptidases and their (patho)physiology

Numerous evidence has suggested that KLKs play important roles in many physiological processes such as semen liquefaction and skin desquamation. Moreover, they are also involved

16 in various pathophysiological processes, such as cancer and inflammatory diseases (96). In the skin, the mRNA levels of 9 KLKs were detected by RT-PCR, including KLK1, 4-8, 11 and 13- 14 (97). At the protein level, 8 of the 15 KLKs have been found in human epidermis by ELISA, including KLK1, 5, 6, 8 and 11-14 (33). Another study from our lab was able to detect KLK5-8, KLK10-11, and KLK13-14 in human SC and sweat (98). KLK5 and KLK7 were firstly extracted in the 1990s, named as stratum corneum tryptic/chymotryptic enzyme, with are predominantly involved in desquamation. Among the activities of KLKs detected in the skin SC, KLK7 represents all chymotryptic-like activity while KLK5 ascribes around 50% of tryptic activity. Moreover, in situ hybridization experiments indicated the presence of KLK1, 4, 6, 9, 10, 11, 13 and 14 in the sweat gland (97).

The first report of KLKs involved in skin function was published in the 1990s and showed that KLK5 and KLK7 were associated with the desquamation (28,29,99). Since then, a large number of studies have been reported and supported the notion that KLKs serve as desquamation enzymes. Through in vitro assays, KLK5 was found to be able to cleave CDSN, DSG1 and DSC1, while KLK7 can proteolyze CDSN and DSC1 (100). Additional studies from our lab suggested that KLK6, KLK13 and KLK14 also were able to cleave the recombinant DSG1 protein (101).

Antimicrobial peptides (AMPs) are part of the host defense response, and help fight against pathogenic infections. The activities of AMPs have been reported to be regulated by KLKs. In skin, β-defensins and cathelicidins are the two major AMPs families. KLK5 and KLK7 regulate the activity of cathelicidin by processing the precursors into active forms or proteolyze the active form into non-active small fragments (102) And KLK5 can cleave β-defensin KLK5 in vitro (103). Interestingly, KLKs inhibitors, such as LEKTI, potentially regulate AMPs processing through inhibiting the activity of KLKs, as increased AMPs activity was found in mice lacking the expression of KLKs inhibitors (102). In rosacea, a common skin condition with red or blush face, the increased activity of cathelicidin is related to the elevated level of KLK5 (104). In addition, other studies demonstrated that KLKs also contribute to the normal skin barrier function. For instance, KLKs (i.e KLK5 and KLK7) may be involved in the maintenance of skin permeability via the degradation of the lipid processing enzymes (105,106).

Besides the physiological roles, KLKs seems to contribute to the pathogenesis of multiple skin diseases as well. For instance, atopic dermatitis (AD) is a common skin inflammatory disease, with symptoms like dry and scaly skin. The skin expression of several KLKs, such as KLK5, KLK6, KLK7, KLK8, KLK10, KLK11, KLK13 and KLK14, were significantly increased in AD patients, compared to healthy controls (70,107). Among these elevated KLKs, the increased expression of KLK7 in the SC of AD patients was found to be predominant as compared to - KLKs with tryptic-like activity (107). Moreover, increased expression levels of KLK7, KLK11 and plasmin, correlated with the elevated proteolytic activity in the lesional skin of AD patients compared to the non-lesional skin or healthy skin (70). Furthermore, enhanced KLK7 in the AD lesion was found by immunohistochemistry staining, and was correlated with the IL-4 serum level in AD patients (64). Recent studies in a murine AD model suggested that elevated pH resulted in enhanced expression of KLK5, and further led to the skin barrier dysfunction via activation of the protease-activated receptor 2-thymic stromal lymphopoietin pathway (108).

Psoriasis is an autoimmune disease and patients with psoriasis exhibit symptoms including a red, itchy and scaly skin. The expression of several KLKs is significantly up-regulated in the lesional SC compared to the non-lesion SC of psoriasis patients, including KLK 5-8, KLK10-11 and KLK13-14 (109). Moreover, serum levels of KLK 5 and KLK11 in psoriasis patients were found to decrease significantly after therapeutic treatment (i.e. etretinate).

Another skin disease in which KLKs are involved is Netherton syndrome, which is a rare skin disease with the incidence of around 1/200,000 births (http://www.orpha.net/consor/cgi- bin/OC_Exp.php?Expert=634). Netherton syndrome patients typically present with mild to high skin barrier dysfunction, severe desquamation and atopic-like inflammatory symptoms (110). Netherton syndrome is caused by a genetic mutation of the SPINK5 gene, and this gene encodes the serine protease inhibitor LEKTI, which is a 120-kDa protein that contains 15 domains in humans. Both full-length and fragmental domains of LEKTI can inhibit a list of serine proteases, including plasmin, subtilisin A, cathepsin G, trypsin and human neutrophil elastase (111). Moreover, it has been demonstrated that KLK5 and KLK7 activity are elevated in both Netherton Syndrome patients and spink5 knockout mice, and excessive degradation of DSG1 in spink5 knockout mice was associated with elevated activity of KLK5 and KLK7 (112,113). Knockdown of KLK5 or KLK7 can partially rescue the impaired epidermal architecture and it

18 can increase the expression of several cornedesomesome proteins in spink5-/- mice (112,114,115).

1.3.3 Skin phenotype in a KLK transgenic mice model

Genetically modified mice have been generated as great models to study the (patho)physiological functions of KLKs. For instance, a tissue-specific KLK5 transgenic mice model (Tg-KLK5) was reported recently, and in this model, overexpression of human KLK5 was induced under the human involucrin promoter in the granular layer of the murine epidermis (116). As expected, elevated expression and activity of KLK5 were detected in Tg-KLK5 mice, with an increased activity level of KLK5 downstream targets such as KLK7, KLK14 and elastase-2. Disrupted skin barrier function and inflammatory skin phenotypes, such as exfoliative erythroderma with scaling, growth delay and hair abnormalities, were also observed in these Tg-KLK5 mice. Moreover, the SC was detached via desmosomal cleavage, and expression of inflammatory cytokines and chemokines, and the infiltration of immune cells in skin were enhanced (116). Thus, KLK5 was suggested to be involved in maintaining the skin barrier function and contribute to the process of skin pathology such as Netherton Syndrome and atopic dermatitis. Recent study indicated that neither ablation of KLK5 or KLK7, but deletion of both KLKs could rescue the lethal phenotype in LEKTI-deficient mice. And the triple deficient mice exhibited completely rescued epidermal barrier, comparing to double deficient mice (spink5A135X/A135X /klk5-/-; spink5A135X/A135X /klk7-/-) (117).

Transgenic mice expressing human KLK7 were also generated (24). Over-expression of KLK7 protein and elevated proteolytic activity of KLK7 were observed in these transgenic mice. Pathologic skin phenotype was found to be similar to chronic lesions of atopic dermatitis patients, including increased epidermal thickness, hyperkeratosis, dermal inflammation and severe pruritus (24). This study suggested that KLK7 may play an essential role in the pathogenesis of inflammatory skin diseases.

KLK8 knockout mice were generated and studied by Kirihara et al (118). Under normal conditions, there was no notable histological difference in the epidermis between wild-type and

KLK8-/-mice. However, under ultraviolet light irradiation, KLK8-/- mice exhibited skin inflammation and the delayed skin recovery. 12-O-tetradecanoylphorbol-13-acetate (TPA) can induce KLK8 mRNA level which cause epidermal proliferation and hyperkeratosis in wild-type mice (119); however, TPA induction resulted in less proliferation in the stratum corneum of KLK8-/- mice, confirming that KLK8 was essential for TPA-induced epidermal proliferation and hyperkeratosis (120). Moreover, less cleavage of adhesion proteins desmoglein 1 (DSG1) and corneodesmosin (CDSN) were associated a delay in corneocyte shedding in KLK8-/- skin, resulting in the hyperkeratosis phenotype. All these studies indicated that KLK8 may be important to maintain skin normal function.

1.3.3.1 Association of KLK7 with skin dysfunction

Since KLK7 was first discovered as a stratum corneum chymotrypsin-like protease, its function in the skin has been well studied in desquamation (100,121). Additional studies indicated that through interaction with various substrates, KLK7 helps maintain skin barrier function and contributes to skin innate immunity (96,121). As mentioned in Chapter 1.2.3, aberrant expression or activity of KLK7 was associated with several skin diseases, such as psoriasis, Netherton Syndrome, and atopic dermatitis (70,107,109,112,113). It has been reported that KLK7 is also involved in the processes of other skin dysfunctions that are listed as follows:

(1) Skin cancers: KLK7 expression was found in around 80% of primary squamous cervical tumors and 20% of primary adenocarcinomas, although KLK7 was undetectable in normal cervical keratinocytes (41). Melanoma is one of the most metastatic skin cancers (122). Previous studies using gene expression profiling with human primary cutaneous melanoma have shown that KLK4, KLK7 and KLK11 are associated with metastatic dissemination and overall survival of patients with primary melanoma (123). Moreover, protein expressions of KLK6 and KLK7 were significantly higher in melanoma tissues than in common nevi (124). Another high-throughput microarray study of melanoma tissues indicated that expression of KLKs is associated with melanoma progression before the establishment of distant metastasis (125).

(2) X-linked ichthyosis (XLI): XLI is a recessively inherited ichthyosis, and its typical syndromes are dry, polygonal and thick dark scale on the skin. XLI is associated with gene deletion of steroid sulfatase (STS); however, recent studies indicated that half of the patients also harbored KLK7 mutation (126).

(3) Impaired epidermal barrier function of infants: it is known that the neonatal stratum corneum (SC) rapidly hydrates and the skin surface acidifies to the adult level by ~4 weeks of age. Compared to adult SC, the hyper-activity of KLK7 and reduced level of filaggrin- derived natural moisturizing factors were found in the neonatal SC (127).

(4) Human skin equivalent (HSE) models: research on these models suggested that reduced activity of KLK5 and KLK7 might lead to the excessive thickening of the stratum corneum (128). Although the protein level of KLK7 was found to be elevated in HSE, the activity of KLK7 was low, which could be due to the abundant and diffuse epidermal protease inhibitors, such as lymphoepithelial Kazal-type-related inhibitor (LEKTI) and skin-derived antileukoproteinase (SKALP), or due to reduced activity and subsequent activation of KLK7.

1.4 Putative substrates of kallikrein-related peptidases

Despite the great advances in our understanding of the KLK (patho)biology, the identification of high-throughput proteomic techniques, together with the development of KLK-specific animal models, have recently offered ample new insights regarding the putative KLK substrates and their implications in health and disease. Undoubtedly, elucidation of the exact KLK substrate repertoires will benefit greatly from the ongoing development of KLK-specific activity-based probes and highly selective KLK inhibitors. The physiological substrates of these enzymes remain largely an open question. The advent of an updated overview of all physiological KLK substrates is presented in Table 1.2.

1.4.1 Regulation of kinin signaling pathway by KLKs

Since its initial discovery, the functional roles of KLK1 have been thoroughly studied (129). Similar to plasma kallikrein (KLKB1), KLK1 targets the kinin-mediated signaling pathway and is an important regulator of blood pressure, cell proliferation, neovascularization promotion and platelet aggregation through cleavage of the low molecular weight kininogen (LK) into bradykinin (13). The exact cleavage sites of LK by KLK1 have been identified as GFSPFR↓SSRIG and MISLM↓KRPPG (130). Using kinin-releasing assays, it has been shown that KLK2 and KLK13 may also display some activity against LK, however at significantly lower levels compared to KLK1 (131,132). Whether other KLKs (other than KLK1) can release vasoactive kinin from kininogens is a subject of an active debate in the literature (133).

1.4.2 Semen function

Among 11 KLKs detected in seminal plasma, KLK3 (also known as prostate specific antigen, PSA) is the major component of seminal fluid, with a concentration of 0.5-3 mg/mL (134). The majority of KLK3 is found in seminal fluid in its active state, free from any inhibitors (134). In the 1980s, the first identified natural substrates for seminal KLK3 were semenogelin I, II and fibronectin (135,136). Since then, fibronectin has been proposed as a putative substrate for more KLKs, including KLK2, 4, 5, 6, 7, 8 and 14 (21,101,137-141). The spermatozoa mixture of semenogelins and fibronectin creates a gel like coagulum, which prevents sperm from acidic pH, but limits the mobility of sperm. KLK3-mediated degradation of semenogelins and fibronectin results in semen liquefaction, which allows sperm to be mobile. Both semenogelins I and II can be digested by KLK3 into multiple peptides in vitro, at KLK3-preferred cleavage sites (P1: leucine or tyrosine)(142,143). KLKs 2, 5 and 14 have also been shown to be able to hydrolyze semenogelins in vitro, but whether these enzymes possess a functional in vivo role in semen remains questionable (144). For instance, the activity of KLK2 in seminal fluid is drastically compromised by immediate binding to endogenous inhibitors (e.g. inhibition of KLK2 from protein C inhibitor) (137). Interestingly, KLK14 was found to be significantly lower in asthenospermic cases and in patients with delayed liquefaction. It is now established that KLK14

22 can activate pro-KLK3, which suggests a possible role for KLK14 as a main regulator of semen‟s coagulation and liquefaction cascade (145).

1.4.3 Skin-related KLK substrates

In human epidermis, 9 KLKs (KLK1, 5-8, 10-11, 13-14) have been found in stratum corneum (SC), stratum granulosum (SG) and epidermis appendages (such as eccrine sweat glands, hair follicles and nerves). Active forms of KLK5, KLK7 and KLK14 have been detected in stratum corneum extracts, while active KLK8 has been recently demonstrated in human sweat (67,68). Desquamation is a physiological biweekly process, in which terminally differentiated keratinocytes (known as corneocytes) are shed off from skin surface in a continuous renewal process. Regulated degradation of corneodesmosomes is crucial for balanced barrier breakdown and maintenance of skin barrier integrity. Corneodesomosomes are junctional proteins, consisting of corneodesmosin (CDSN), desmogleins (DSG 1, 4) and desmocollins (DSC 1), which collectively mediate keratinocytes adhesion (100). KLK 5 and 7 were initially detected as the major desquamatory enzymes, since they are highly expressed in granular keratinocytes and intercellular spaces of SC. Using in vitro digestion systems, both KLK5 and KLK7 could potently cleave CDSN, DSC1, while DSG1 was mainly degraded by KLK5 (100). Interestingly, DSG1 has also been shown to be cleaved by other skin KLKs, such as KLK6, 13 and 14 (101). The KLK-mediated degradation of DSG1 can be inhibited by epidermal KLK inhibitor lymphoepithelial-kazal-type 5 inhibitor (LEKTI) (146). Loss of function of LEKTI can cause Netherton syndrome (NS), a severe disease with aberrant skin barrier and deregulated proteolysis of SC. Excessive degradation of DSG1 and DSC1, has been observed in NS and Spink 5-/- mice (Spink: the gene encoding serine protease inhibitor Kazal-type 5), mainly due to hyperactivity of KLK5 and KLK7 (113,147). Finally, KLK5 has also been recently shown to proteolytically activate pro-filaggrin to natural moisturizing factor highlighting a critical role for KLK5 in the maintenance of skin barrier function (148).

1.4.4 KLK Substrates involved in innate immunity

Recent research has suggested that KLKs contribute to human innate immunity through activation or degradation of antimicrobial peptides. Cathelicidin-related antimicrobial peptides are a family of polypeptides found in macrophages and leukocytes that play critical roles in microbial defense. Cathelicidins vary in size from 12-80 amino acid residues and their activity is controlled by proteolytic processing of the pro-form (hCAP18 in humans) to a mature peptide (LL-37). In skin, activation of human cathelicidin peptides has been attributed to KLK5 and KLK7. According to Yamasaki et al, KLK5 and KLK7 can activate precursor hCAP18 into mature form LL-37, and further generate higher activity antimicrobial peptides (middle size), while excessive digestion of cathelicidin by the two enzymes results in significant loss of antimicrobial activities (102). More recently, our group has shown that KLK8 and KLK14 can also cleave synthetic LL-37 into shorter active antimicrobial peptides, suggesting additional antimicrobial roles for these enzymes in the skin (67). Similar regulation of microbial defense by KLKs has been also demonstrated in studies with cervico-vaginal fluids (CVF). Defensing-1α, which is abundantly expressed in CVF have been proposed as putative substrates for KLK5. Interestingly, similar fragmentation pattern was found in ex vivo cleavage of defensing-1α mix with CVF, however the biological relevance of KLKs in this process is under investigation (63).

1.4.5 KLK substrates in the central nervous system

KLK6 is one of the most abundant serine proteases in human brain the dysregulation of which has been associated with many neurological diseases, including Parkinsons disease, Alzheimer‟s disease, multiple sclerosis and spinal cord injury (149-152). Most of the insights into the functional roles of KLKs in the neural system come from studies in relevant animal models. For instance, in 2002 Bernett et al reported that KLK6 was able to degrade human myelin basic protein (MBP), a finding that implicates KLK6 in the pathogenesis of several demyelinating diseases (e.g. multiple sclerosis) (14). Interestingly, the use of KLK6-neutralizing antibodies significantly relieved the pathological symptoms of in a rodent model of multiple sclerosis (153). Moreover, α-synuclein, which is involved in the aggregation of Lewy bodies in Parkinsons disease, can also be cleaved by KLK6. In fact, using a α-synuclein transgenic mouse model,

Spencer B et al showed that lentiviral delivery of KLK6 resulted in clearance of α-synuclein (154). Furthermore, proteolysis of amyloid precursor protein by KLK6 has also been demonstrated, indicating a potential contribution of KLK6 in the development of Alzheimer disease (155). Lastly, in spinal cord injury studies, it has been shown that KLK6 can decrease neurite outgrowth through proteolytic digestion of certain laminins (151). Other than KLK6, KLK8 is also abundant in brain tissues. KLK8 has been shown to also mediate cleavage of MBP (156). More recently, KLK8 has been involved in memory and anxiety regulation through cleavage of ephrin type-B receptor 2 (EphB2) in the amygdala (157).

1.4.6 Extracellular matrix-related KLK substrates

The extracellular matrix (ECM) is the non-cellular constituent found in all tissues that provides vital physical support for the cells and regulates central biochemical functions, such as cell differentiation, cell-cell interaction and tissue homeostasis. KLKs have long been known to interact with many proteins of the ECM. For example, KLK 3, 5, 6, 13 and 14 are all able to cleave recombinant laminins in vitro (13,136,138,158,159). Similarly, collagens can also be cleaved by KLK2 (160), KLK5 (138), KLK6 (21,155) and KLK13 (159). As mentioned above, fibronectin (another major ECM glycoprotein) is also a common substrate for several KLKs (21,136,138-140). Recent studies with KLK12 have revealed that this enzyme mediates signaling properties through cleavage of the mitogen CTGF, Cyr61 and NOV proteins (CCN family), which all belong to ECM-associated signaling protein family (161). Additionally, through in vitro degradomics, a list of extracellular matrix proteins was found to be cleaved by KLK7, including Cyr61, CTGF and tenascin (162). KLKs can also regulate urokinase-type plasminogen activator receptor (uPAR) and its ligand serine protease urokinase-type plasminogen activator (uPA), which is essential for cell signaling, cell-ECM interaction and ECM proteolysis (163). It has been reported that both KLK2 and KLK4 can convert zymogen pro-uPA into active uPA, which may contribute to increased invasion potential of the prostate (164,165) and ovarian cancers (166). Lastly, KLKs can also interact with cell adhesion ECM proteins, by transiently expressing human KLK6 in the HEK293 cells, Klucky et al observed a significantly enhanced cleavage of E-cadherin (167). Along with this line, KLK6 induced keratinocyte proliferation and migration in KLK6 transgenic mice, was also associated with decreased level of E-cadherin in

25 epidermal keratinocytes. Similarly, KLK7 was able to generate soluble E-cadherin fragments via in vitro proteolysis (55).

1.4.7 Secreted substrates of KLKs

KLKs have been implicated in the regulation of several hormones, growth factors and cytokines. For instance, KLK1 has been reported to activate pro-insulin into mature form insulin (168). Moreover, there are several studies suggesting that KLKs regulate several growth factors. KLKs 2, 3, 4, 5 and 14 have all been shown to cleave insulin-like growth factors binding proteins (IGFBPs) (169-172). Cleavage of IGFBPs by KLKs and concomitant loss of IGF-binding capacity has been associated with increases in IGF bioavailability, which may partially account for the involvement of KLKs in tumor growth and metastasis (171). Furthermore, KLKs have been implicated in the pathogenesis of many inflammatory diseases via regulation of cytokines, such as IL-1β. IL-1α belongs to interleukin 1 family, which have an essential role in inflammation, tumorigenesis and autoimmune diseases (173). IL-1β is synthesized as precursor proteins, which is activated by caspase-1 (also known as IL-1β-converting enzyme (ICE)). In skin, keratinocytes express IL-1β but not ICE (174). On the contrary, KLK7 rather than ICE converts pro-IL-1β into mature form IL-1β (175). Finally, for more than 20 years now it is known that KLK3 can cleave latent hTGF-α into mature form in the conditioned medium of prostate cancer cells (176). More recently it was shown that more KLKs (4, 5, 6 and 7) can cleave TGFβ-1, indicating KLKs as important mediators of tumor progression through regulation of TGFβ-1 signaling (177).

1.4.8 KLK-regulated surface receptors

Proteinase-activated receptors (PARs) constitute a 4-member family of G protein-coupled receptors with ubiquitous expression and crucial signaling roles in the pathobiology of several pathologies, including cancer, central nervous pathologies, gastrointestinal diseases and cardiovascular abnormalities (178,179). These seven-transmembrane receptors are known targets of extracellular proteases, which activate PARs through proteolytic cleavage and unveiling of a

26 tethered ligand that stimulates their activity. Emerging literature has established a tight connection between PARs and KLK activity (180). We have recently shown that KLKs 1, 2, 4, 5, 6 and 14 can cleave and regulate PAR activity (181). Along with this line, KLKs 5, 6 and 14 (and not KLK7 and 8) were shown to activate PAR2 (182), while KLK14 was also shown to be able to inactivate PAR1(22,180). It is becoming evident that the interaction of KLKs with PARs is tissue and disease-dependent. For instance, in the setting of prostate cancer, KLK4 has been shown to co-localize with PAR2 and to trigger calcium signaling through PAR1 and PAR2 activation (183). In melanoma, KLK6 activation has been linked to increased metastasis and proliferation through enhancing of PAR2 activation (184). Similarly, a critical association between PARs and several KLKs has been suggested in many types of neurodegenerative diseases (e.g. multiple sclerosis) (185). Lastly, in skin pathologies, hyperactivity of KLK5 (a known state in Netherton Syndrome) can promote the expression of proinflammatory cytokines and chemokine via activation of PAR2 (186). The interplay between KLKs and PARs in the skin is definitely an exciting subject of ongoing research.

1.4.9 Other KLK substrates

A unique role for KLK4 has been established in the tooth, during enamel maturation (25). Elegant work in the field has revealed that ameloblast-produced MMP20 activate KLK4, which, in turn, cleaves amelogenin, enamelin and ameloblastin leading to enamel maturation (187). Recently, in vitro degradation assays have suggested that KLK14 can cleave complement C3 to C3a fragments, implicating a potential role for KLK14 in inflammatory response regulation (188). Lastly, KLK7 has been shown to cleave procaspase-14 into intermediate forms during terminal differentiation (189).

Table 1.1 Summary of the putative physiological substrates of each KLK (insufficient data exist for KLKs 9, KLK10 and KLK11). KLK Putative Natural substrates Reference

KLK1 Kininogen, Proinsulin (168,190)

KLK2 Kininogen, Semenogelins, Fibronectin, (132,137,165,171)

pro-KLK3, Collagen, IGFBP 2-5, uPA,

PAR2

KLK3 IGFBP3, Semenogelins, Fibronectin, (136,137,171,176)

Laminin, hTGF- beta

KLK4 Enamelin, amelogenin, ameloblastin, (139,164,172,183,191)

pro-KLK3, PAR1, PAR2,

Fibronectin,collagen, uPA, IGFBPs

KLK5 DSG1, DSC1, CDSN, ProKLK7, (68,100,138,182,191)

PAR2, Cathelicidin, Defensing-1 α,

Fibronectin, Laminin, Collagen,

semenogelins, pro-fillagrin, IGFBPs,

Mucin 4, Mucin 5B

KLK6 PAR1, PAR2, Amyloid beta protein, (14,22,154,155)

MBP, Fibronectin, Collagen type 1, IV,

α-synuclein, laminin, DSC1, E-cadherin

KLK7 DSC1, CDSN, Pro-MMP9, (100,140,175,189,192)

Cathelicidin, Fibronecin, Pro-IL-1beta,

E-cadherin,

Procaspase-14

KLK8 MBP, EphB2, Fibronectin, (67,141,156)

LL-37(synthetic peptides)

KLK12 CCN1-6 (161)

KLK13 Fibronectin, Collogen type I, type IV, (131,159)

Laminin, DSG1, LK

KLK14 Kininogen, KLK5, Pro-KLK3, (68,145,169,182,188,193)

Desmoglein-1, Vitronectin precursor,

Fibronectin, Collogen type I, type IV,

Laminin, IGFBP2-3, PAR1, PAR2,

Semenogelin, LL-37(synthetic

peptides), Complement C3

KLK15 Pro-KLK3 (194)

Figure 1.4 Physiological KLKs substrates in various processes. Centre, The pro-form of KLKs is proteolyzed into active form KLKs. The active KLKs are involved in different biological processes including cell adhesion, desquamation, semen liquefaction, neuron cells and kinin system. 1) Substrate proteins involved in cell adhesion include cadherin, extracellular- matrix proteins, cytokines, cytokines, growth factors, hormones and signaling receptors; 2) Corneodesmosomes proteins (such as desmoglein 1, desmocollin1 and corneodesmosin) are reported to be cleaved by KLKs.; 3) KLKs can cleave semenogelin and fibronectin which results in semen liquefaction; 4) KLKs are reported to proteolysis MBP, α-synuclein and amyloid β protein in neuronal synapses.

1.4.10 KLKs in human biofluids

KLKs are widely expressed in various tissues of the human body. Co-expression of KLKs were reported in several biological fluids, including sweat, seminal plasma, breast milk, follicular fluid, breast cancer cytosol, amniotic fluid, breast cyst fluid, ovarian ascites, saliva, urine, CVF and CSF (33,98). Interestingly, different isoforms of KLKs were reported in various biological fluids. For instance, differential N-glycosylation of KLK6 was detected between ovarian cancer ascites fluid and CVF (195). Although active forms of KLKs were found in several biofluids (i.e. seminal plasma, sweat), which highlights their potential proteolytic function roles in responding tissues (67), KLKs complex (KLKs binding to inhibitors) were commonly detected in biofluids (i.e seminal plasma, ovarian ascites) (196). To date, seminal plasma is one of the biofluids with characterized KLKs function. In seminal plasma, KLKs coexist with their putative substrates (i.e. semenogelins and fibronectin). Proteolysis of these substrate proteins result in semen liquefaction and enhanced sperm motility, while proteolytic reaction is tightly controlled by multiple environmental factors (i.e protein C inhibitors, Zn2+ concentration).

1.5 Strategies for substrates finding

Traditionally, the classic method of determining whether a protein is a substrate for a given protease, is the direct incubation of the candidate substrate with the protease of interest followed by subsequent examination for candidate proteolytic products (197). This hypothesis-driven approach has been widely used for the identification of putative KLKs substrates in the past. For instance, in the 1990s it was observed that the stratum corneum tryptic enzyme (KLK5) and chymotryptic enzyme (KLK7) were abundantly expressed in granular keratinocytes and stratum corneum of the skin (30), however their exact physiological substrates were only revealed in 2004, when Caubet et al demonstrated that co-incubation of KLK5 and KLK7 with corneodesmosin (CDSN), desmoglein 1 (DSG1) and desmocollin 1 (DSC1) led to degradation of all three proteins (100). Since then, the same approach has been used for the identification of a several putative KLK substrates, including extracellular matrix collagens by KLK6 (21), fibronectin by KLK7 (140) and laminin by KLK5 (138). The obvious disadvantage of this method is the lack of robustness and the bias that emanates from the arbitrary pre-selection of the

31 candidate substrates. Furthermore, the fact that a protease can cleave a protein in vitro does not necessarily mean that this cleavage is also happening in vivo, where proteases and endogenous substrates are under completely different micro-environmental dynamics.

Combinatorial scanning of peptide libraries, phage display and matrix substrate library screening are additional methods that have been used for the profiling of KLK specificities (172,198,199). With the exception of KLKs 9 and 15, these methods have been widely used for the characterization of the active-site profiling of the prime and non-prime preferences of most KLKs (160,172,199-204). Despite some minor discrepancies, all three methods clearly distinguish KLKs in two groups based on their P1-preference: a) the tryptic-like KLKs (4, 5, 6, 11 and 14), which display a strong preference for basic P1 residues (arginine/lysine) and b) the chymotryptic-like KLKs (3, 7 and 9), which prefer aromatic or more bulky P1 amino acid residues (phenylalanine/tryptophan). A common limitation of the above mentioned methods is their inability to provide direct insights into how KLKs interact with their substrates in an ex vivo or in vivo setting (Table 1.2 summarizes the limitations of each of these methods). This problem has been recently tackled by the emergence of powerful mass spectrometry-based technologies, as discussed in detail below.

All the following techniques have been used for the identification of KLKs physiological substrates (as discussed in the previous Chapter). Of note, some MS-based methods (e.g. CONFRADIC method) that have been successfully used for the characterization of other enzymatic families, has not been utilized in the KLK scene yet (205,206). A summary of the MS-based approaches that has been used for the characterization of KLK substrates is depicted in Fig. 1.5.

1.5.1 Proteomic identification of protease cleavage sites (PICS)

Proteomic identification of protease cleavage sites (PICS) is a powerful new method, which allows the determination of both prime and non-prime site preference PICS of an enzyme after incubation with a proteome-derived peptide library (197). The ingenuity of the method lies in the chemical protection of the primary amines of all peptides, which allows enrichment of the neo-

N-termini of the prime-side protease products after incubation with the protease of interest. The newly generated termini are biotinylated and detected with liquid-chromatography coupled to mass spectrometry (LC-MS), while the non-prime sites are predicted by blasting the prime sequence against the human proteome database. This technique has been successfully used to delineate the physiological substrates of many proteases, including thrombin, matrix metalloprotease 2, cathepsin G and caspase-3 (207) and KLK2 (208).

1.5.2 Protein topography and migration analysis platform (PROTOMAP)

Protein topography and migration analysis platform (PROTOMAP) is another method which was recently invented by Ben Cravatt et al at the Scripps Research Institute (209). This method allows characterization of global proteolytic events in biological systems, by combining 1D SDS-PAGE with liquid chromatography (LC) - tandem mass spectrometry (LC-MS/MS). In a classical PROTOMAP analysis, complex biological samples are first separated on a 1D gel, based on their size (MW). Gel bands are then excised, trypsin-digested and analyzed by LC- MS/MS. Data from each gel band is incorporated into multiple “photographs”, which facilitate identification of modifications in migration and topography between control (low or no enzyme activity samples) and experimental (protease treated) samples.

Based on this data, candidate substrate proteins are identified and a quantitative assessment of the efficiency of cleavage by the enzyme is provided. This approach has been recently applied to analyze several proteolytic networks in humans (209-211). Among others, PROTOMAP has also been used to study KLK4 substrates in a prostate cancer cell model, which identified KLK4 as a major regulator of several of TGFβ1-related proteins (212).

1.5.3 Cell surface protease degradomics

Secreted proteases are in a constant interplay with extracellular molecules (cytokines, chemokines, and growth factors), adhesion proteins and cell surface receptors. The objective of degradomics studies is the characterization of all proteolytic fragments that follow the treatment of cells (or tissues) with an active enzyme. In cell culture models, cleaved peptides can be

33 detected in the culture supernatant, with the use of tandem mass spectrometry and database mining in a standard fashion. Quantification of these methods can be achieved with differential tagging of proteins prior to enzymatic treatment, such as in the case of Isobaric Tags for Relative and Absolute Quantification (iTRAQ) and Isotope-Coded Affinity Tags (ICAT), or with metabolic labeling of cells such as in the case of Stable isotope labeling by amino acids (in cell culture models)(197,213). For instance, the degradome of KLK12 has been recently investigated using a similar approach. By comparing the spectral count of peptides among different biological samples (enzyme treated vs non-treated controls), a novel extracellular matrix protein, cysteine rich angiogenic inducer 61 (CYR61), has been identified as a putative KLK12 substrate in breast cancer cells (161).

1.5.4 Terminal amino isotopic labelling of substrates (TAILS)

Terminal amino isotopic labelling of substrates (TAILS) is another quantitative MS-based method that has been recently implemented towards the identification of the substrate repertoire of selected KLKs. This method is based on quantitation of N-terminal fragments of each protein (N-terminal peptides) through identification of alterations in relative protein abundance among treated and non-treated samples (214). Enrichment occurs through modification of N-termini and lysine residues before trypsin digestion. Blocked peptides cannot bind to hyperbranched polyglycerols (HPGs) columns, which is specifically used in TAILS. The trypsin digested non- N-terminal peptides binding to the column are removed from the samples (214). Most importantly, the amines blocked by reductive dimethylation can use both heavy (i.e. d(2)C13- formaldehyde) or light (i.e. d(0)C12-formaldehyde) labeling for different biological samples, which result in quantitative comparison of multiple samples (214). This method has been recently applied in the KLKs field. The degradomics profiling of KLKs 4, 5, 6 and 7 have been investigated in an ovarian cancer cells model (177). Among the lists of putative KLKs substrates the most significant findings included the growth differentiation factor 15 (GDF 15) and the macrophage migration inhibitory factor (MIF) (177).

Table 1.2 Overview of the techniques commonly used for protease substrate identification.

Primary Natural Limitations/Disadvantages Specificity Substrate Methods Information Information

In vitro Yes Yes Pre-selection of putative substrates allows Fluorescent Assay biases. Lack of throughput.

Phage display Yes No Possible bias via under-representation of particular sequences; false-positive interaction of peptide sequence with the solid support, system is lacking protein- protein, or protein-cell interactions common in physiological conditions.

PSL Yes No Induced-fit neighboring bias (binding of an amino acid influences can depend on the type of amino-acids that is beside it), lack of information about nonprime sites; the system is lacking protein-protein, or protein-cell interactions common in physiological conditions.

PICS Yes No Overlapping cleavage sites between the enzymes used to create the library and the enzyme of interest, inefficient re- constructed nonprime sites.

2D-DIGE No Yes Low throughput, time-consuming (2-3 days per run), not suitable for membrane bound or small (<10KDa) proteins.

PROTOMAP No Yes Non-optimal sensitivity, non-complete coverage of explicit proteolysis sites.

Cell

Surface Possible Yes Absolute need for quantitation (e.g. Degradomics ITRAQ, ICAT, SILAC) to reduce false positives.

COFRADIC Yes Yes Time and labor intensive.

Figure 1.5 Summary of current MS-based methods towards the identification of the substrate repertoire of a protease.

1.6 Rationale, hypothesis and objectives

1.6.1 Rationale

Kallikrein-related peptidases (KLKs) are a group of 15 homologous serine proteases, which are widely distributed in various tissues and body fluids (9). The majority of KLKs display trypsin- like activity, while a few of them have chymotrypsin-like activities. KLKs are known to participate in a well-characterized proteolytic cascade that regulates desquamation, barrier function and innate immunity of the skin (96). KLK7, known as stratum corneum chymotryptic enzyme, is abundantly expressed in the skin and is also the sole chymotryptic serine proteases in skin (121). Aberrant regulation of skin KLKs (e.g. KLK7) has been linked to severe skin pathologies (i.e. psoriasis, atopic dermatitis, Netherton Syndrome and melanoma) (121,125). Previous work has suggested that KLKs may be involved in multiple physiological and pathophysiological processes through activating or inactivating targeted substrates in various tissues (215). Identification of the direct endogenous substrates of kallikrein-related peptidases remains an unmet objective (215). Traditional strategies (i.e. combinatorial peptide libraries scanning) have been used in KLK substrate identification with several known limitations, such as lacking of protein-protein, or protein-cell interactions common in physiological condition (215). Thus, a new strategy is needed to systemically investigate the physiological substrates of KLK7.

The advanced MS-based approach offered new insights regarding the putative substrates of many proteases and their implications in health and disease. However, most of these substrates identification strategies are mainly performed on the in vitro cell culture system. Thus, investigation of human biofluids, such as sweat, will potentially reveal the endogenous proteins/peptides substrates of KLKs, which helps depict how KLKs act as part of an interconnected web of proteases. Previous studies indicated that 8 of 15 KLKs were detected in skin and sweat, respectively (33,98). Several skin KLKs were found in their active forms, which highlight their potential roles involved in skin function. A plethora of proteomics and peptidomics studies of several biofluids, such as urea, plasma, urine, cerebrospinal fluid and saliva, were reported in the last decade, but human sweat has not been well characterized (216- 218).

1.6.2 Hypothesis

KLK7 is involved in the skin physiology and pathobiology through regulating the activities of enzyme-specific substrates, and targeting specific KLK7 substrates can be developed as novel therapies for various skin diseases

1.6.3 Objectives

Objective I. Characterization of purified active-form KLK7 recombinant protein

1. To generate pPIC9-KLK7 yeast-expression construct 2. To generate a stable KLK7 yeast expression system 3. To produce and purify active KLK7 recombinant protein 4. To characterize the purified KLK7 with SDS-PAGE, western blot and mass spectrometry 5. To measure the activity of KLK7 with zymogen gel and fluorogenic substrate assay

Objective II. Identification of putative substrates for KLK7 with degradomics combined with substrate specificity analysis 1. To perform proteomic analysis of cell lines overlaying media at different experimental conditions (active KLK7-treated/vehicle control) 2. To validate the specificity of cleaved peptides in different biological replicate samples under different experiment conditions 3. To validate potential candidates by using in vitro recombinant protein digestion assay 4. To establish selective reaction monitoring assay (SRM) assay targeting specific substrate peptides generated by KLK7 cleavage 5. To analyze the kinetic of enzymatic digestion of candidate substrates by KLK7 using SRM assay 6. To perform biological function assay with KLK7 substrate candidate

Objective III. Putative KLK7 substrate peptides identified in human sweat

1. To perform proteomic and peptidomics analysis of human sweat samples 2. To detect the KLKs in sweat with ELISA and western blot 3. To validate putative KLK7 substrate identified in sweat

Chapter 2

Expression and characterization of recombinant kallikrein-related peptidase-7 protein

Sections of this chapter were published in Biological Chemistry.

Yu Y, Prassas I, Dimitromanolakis A, Diamandis EP.

“Novel biological substrates of human kallikrein 7 identified through degradomics.”

J Biol Chem. 2015 Jul 17; 290(29): 17762-75.

2 Expression and Characterization of kallikrein-related peptidase-7 2.1 Introduction

Kallikrein-related peptidase 7 was first discovered by Egelrud et al as a novel stratum corneum chymotryptic enzyme in the 1990s (29). As expected, KLK7 was abundantly expressed in skin with essential functions involved in skin (patho-) physiology, although KLK7 was also detected in many other tissues (33). In skin, KLK7 was indicated to interact (i.e. activate or degrade) a list of substrate proteins that are essential to maintain normal skin functions (215). The regulation of KLKs activity is tightly controlled by many microenvironment factors, such as pH, ion concentration and more importantly, proteases inhibitors (10). Dysregulated expression or activity of KLK5 and 7 was associated with many skin pathophysiologies (e.g. psoriasis, atopic dermatitis, rosacea, Netherton Syndrome and melanoma) (96,112,121,125,181). Among the 15 KLKs, KLK7 has attracted the most attention for the development of KLK-based therapeutics (10). Currently, a KLK7-targeting depsipeptide is undergoing clinical trials as a novel therapy for skin barrier disruption (219) .

Similar to other KLKs, KLK7 is originally translated into a pre-pro-serine protease (inactive form) peptide. The pre-peptide (1-22aa) is the secretion signaling peptide, which can lead the KLK7 precursor peptides to be secreted into extracellular space (9). Pro-peptide (23-30aa), which functions to maintain the latency of KLK7, will be further cleaved to generate the active form (serine proteases domain: 224-250aa) of KLK7 (30). The serine proteases domain contains a characteristic catalytic triad (His70, Asp112 and Sir205) (16), and a predicted N-glycosylation site is located at position 246 (31). Since KLK7 is known as a chymotryptic-protease, it has P1 specificity against Phe or Try (31).

To investigate the function of KLK7 in skin, I here expressed and purified the active-form KLK7 using P. pastoris yeast expression system. The recombinant KLK7 was subsequently analyzed via various biochemical assays. In addition, I analyzed the tissue expression patterns of KLK7 in humans.

2.2 Materials and Methods

2.2.1 Cloning, production and purification of recombinant KLK7

Recombinant KLK7 was produced in the P. pastoris yeast expression system (Invitrogen). Briefly, a DNA fragment encoding the mature form of KLK7 (amino acids 30-253; NP_005037) was amplified from a klk7 cDNA clone (named PCMV-XL5-KLK7, Origene). The primers used were KLK7XhoI-forward “TGCTCGAGAAAAGAATTATTGATGGCGCC” and KLK7EcoRI-reverse “CCGAATTCTTAGCGATGCTTTTTCATGGT” (The bold font indicates KLK7 sequence). The PCR reaction (per 20µL) was performed as following: 0.4µL DNA (vector template, ~ 50ng/µL), 4µL of 5x Phusion buffer (ThermoFisher Scientific), 0.4µL of 10mM dNTPs (Invitrogen), 1µL of 10µM forward primer, 1µL of 10µM reverse primer, 0.2µL of Phusion High-Fidelity DNA polymerase (10U/µL, ThermoFisher Scientific) and 13µL water. The PCR cycling steps were performed as follows: 30s of denaturation at 98°C, 30s of annealing at 56°C plus 30s of extension at 72°C with 30 PCR cycles. The PCR product was digested and further ligated into a pPIC-9 vector containing a α-secretion signal peptide sequence in front of multiple cloning sites. The digestion reaction (total volume: 20µL) was followed as 1µL DNA (~1µg/µL of vector or PCR product), 1µL XhoI (10U/µl, Fermentas), 1µL EcoRI (10U/µl, Fermentas), 4µL of 10x Tango buffer (Fermentas) and 13µL of water. The ligation reaction was performed with the use of the Rapid DNA Ligation Kit (Thermo Scientific): 4µL of 5x ligation buffer, digested PCR product (at 3:1 molar excess over vector), 10-100ng of digested vector, 1µL T4 DNA ligase (5U/µL) and water up to 20µL. The ligation product was transformed into E. coli TOP10 chemical component cells. The transformed cells were further selected by Ampicillin. Sequencing-confirmed pPIC-9-klk7 construct was linearized by SacI enzyme digestion and electroporated into P. pastoris KM71 cells.

A stable KM71 transformant was cultured in BMGY media for 2 days. The yeast cells were pelleted and grown in BMMY medium. Methanol was added to the culture to induce protein expression. At day 6 of the culture, the supernatant was collected and concentrated 10x using a positive pressure ultrafiltration system (Millipore Corp., Bedford, MA) with a 5-kDa cutoff regenerated cellulose membrane (Millipore). Purification was performed with an automated AKTA FPLC system on a pre-equilibrated 5-mL cation-exchange HiTrap high performance Sepharose HP-SP column (GE Healthcare) (buffer A: 0.01M acetic acid, 50mM NaCl, pH 6.0;

Buffer B: 0.01M acetic acid, 1M NaCl, pH 6.0; flow rate at 1 ml/min, gradient elution: 5 min buffer A, followed by linear gradient of 5% buffer B in 25 min, 10% buffer B for 25 min, 15% buffer B for 15min, then 25-100% buffer B in a 30 min gradient). The collected fractions were further concentrated (10x) using 3-kDa Amicon® Ultra centrifugal filters (Millipore), and stored at -80°C for further use. The fractions were analyzed by SDS-PAGE and Western blotting. The concentration of purified protein was measured with coomassie protein assay (Thermo scientific), as well as a KLK7-specific ELISA (33). The N-termini of separated bands on SDS-PAGE gels were analyzed by N-terminal Edman sequencing.

2.2.2 Characterization of recombinant KLK7

2.2.2.1 SDS-PAGE and Western blotting

The Mini-PROTEAN® Tetra cell system and 4-12% gradient polyacrylamide gels at 200V for 45 min (BioRad) were used in this study. Gels were strained with Bio-Safe Coomassie Stain (Invitrogen) or by silver staining (PlusOne Silver Staining kit, protein, GE Healthcare). For Western Blotting, the proteins on the gel were transferred to PVDF membrane (Trans-Blot® TurboTM Mini PVDF Transfer Packs from Bio-Rad) with Trans-Blot® TurboTM Transfer Starter System. The membrane was blocked with 5% milk, and further incubated with primary antibody (1:1000 diluted in 1% milk; in-house made anti-KLK7 polyclonal antibodies (33)) overnight at 4°C. Next, membranes were washed 3 times with PBS and further incubated with horseradish peroxidase-labeled secondary antibodies (1:20000) for 1 hour at room temperature. After 3 times washing with PBS, membranes were incubated with ECL western bolting detection reagent and further exposed to X-ray film (GE Healthcare).

2.2.2.2 Gelatin zymography

The activity of purified KLK7 was monitored by gelatin zymography (Novex 10% Zymogram, Invitrogen). Briefly, KLK7 was separated on the gelatin gel, which was further incubated with renaturing buffer (Invitrogen) for 1 hour and then with developing buffer (Invitrogen) for 4 hours. Finally, the gel was stained with Bio-Safe Coomassie Stain (Invitrogen) and destained until the

42 white bands, which corresponded to the areas of protease activity, were visible against a dark blue background.

2.2.2.3 Fluorogenic AMC substrate profiling

Briefly, a panel of fluorogenic 7-amino-4-methylcoumarin (AMC) peptides (final concentration: 0.25mM) were mixed with the purified KLK7 (final concentration: 6nM) in buffer (0.1M phosphate buffer, 0.01% Tween, pH 8.5). The panel of AMC fluorogenic substrates included two chymotrypsin-like enzyme substrates (LLVY-AMC and AAPF-AMC), five trypsin-like substrates (VPR-AMC, GGR-AMC, EKK-AMC, GPK-AMC, QGR-AMC) and a negative control: AAPV-AMC. Freely released AMC fluorescence was measured with a fluorometer (PerkinElmer Life Sciences) at 380nm excitation, 480nm emission, with 30 second intervals for 10 min at 37°C. To calculate the kinetic constants (Kcat/Km) of KLK7, the recombinant KLK7 was incubated with increasing concentration of LLVY-AMC and AAPF-AMC substrates (0.015, 0.03, 0.06, 0.12, 0.25, 0.50 and 1mM), respectively. Kinetic results were analyzed with the Enzyme Kinetics Module (Sigma Plot, SSPS, Chicago).

2.2.2.4 pH profiling of KLK7

The effect of pH on the activity of KLK7 was performed with different pH buffers, including 0.1M MES (pH 6), 0.1M phosphate buffer, 0.01% Tween (pH 7.0, 7.5, 8.0 and 9.0) and 0.1 M sodium hydroxide/sodium bicarbonate (pH 10-11). Enzyme (6nM) and substrate (LLVY, 0.25mM) were incubated in the corresponding reaction buffer, and the fluorescent activity was monitored as described above. The highest activity of KLK7 in optimal pH buffer was considered as 100%.

2.2.2.5 Effect of cations on KLK7 activity

Briefly, KLK7 (6 nM) was incubated without (control, 0.1M phosphate buffer, 0.01% Tween, pH 8.5) or with different cation salts (i.e. NaCl, MgCl2, CaCl2 and ZnCl2 ) at the concentration

43 range of 0.1-100mM at 37°C for 15 min. After incubation, the substrate (LLVY, 0.25mM final concentration) was added and the reaction was measured as described above. The activity of KLK7 plus substrate alone was considered as 100%.

2.2.3 Tissue expression patterns of KLK7 in humans

Publicly available microarray data profiles from cancer and normal human tissue across 8 cancer types were downloaded from the NCBI GEO repository (220) of microarray experiments. All data were selected to be in a common array platform: Affymetrix HG-U133 Plus2. Suitable experiments were located and data were downloaded as raw CEL files for further processing. The CEL files were processed and normalized using R 2.15.2 and the Bioconductor platform v2.8 (221). Quality control metrics were evaluated for each array (average background, RNA degradation, scale factors, percent present) by Simpleaffy. As per common Affymetrix quality control procedures, samples that showed high 3‟ to 5‟ ratio for control genes (>2.5 for ACTB, >2 for GAPDH) or flagged as outliers in other metrics were excluded. After quality control, expression data were normalized using the gcRMA array normalization (222).

2.3 Results

2.3.1 Yeast expression of pPIC9-KLK7

Active-form KLK7 sequence was amplified from KLK7 cDNA, and the PCR product was visualized by loading on the 1.5% agarose gel. As predicted, a 691 bp length band was observed on the gel (Fig. 2.1B). They were further ligated to the eukaryotic pPIC9 vector, followed by transformation into TOP10 competent E. coli bacteria. The bacteria containing pPIC9-KLK7 were selected and followed by plasmid purification. The purified plasmids were subjected to double-enzyme digestion, and confirmed that KLK7 gene was successfully inserted into pPIC9 vector (Fig. 2.1 C). A 8 Kb vector band and an insert band (~700 bp) were observed on the gel. Sequencing-validated pPIC9-KLK7 plasmid was linearized by SacI enzyme digestion, and the linearized vectors were then electroporated into P. pastoris KM71 cells (a eukaryotic expression system) (Fig. 2.1 D and Fig. 2.2). After the colonies grew up, PCR was performed to confirm that KLK7-expressing vectors were successfully transfected into the selected yeast colonies, before the KLK7 expression was induced by methanol. Twelve PCR-validated yeast colonies were randomly selected and cultured, and one colony displayed the highest expression level of KLK7 on day 6 of methanol induction.

Figure 2.1 Analysis of pPIC9-KLK7. A, Sequence map of pPIC9. The red arrows indicate the insert region of KLK7. B, PCR product of KLK7 on agarose gel. The PCR product (733bp) amplified from pCMV-KLK7 were obtained with the specific primers and then subjected to agarose gel electrophoresis. C, Digestion products of pPIC9-KLK7 on agarose gel. Plasmid extracted from TOP10/ pPIC9-KLK7 colony was double digested by using XhoI and EcoRI at 37°C. D, Single digestion product on agarose gel. pPIC9-KLK7 was digested with SalI.

Figure 2.2 Gene sequencing result of pPIC9-KLK7. Sample= sequence from plasmid extracted from the colony of TOP10/pPIC-KLK7. Sequence alignment was preformed through ClustalW2 program. The arrows indicate the start site and stop site of KLK7 sequence in the construct, respectively.

2.3.2 Production of recombinant mature form of KLK7

The mature form of KLK7 was purified using cation-exchange chromatography. The HPLC fractions with higher UV observation were collected and measured with KLK7 ELISA, including fraction 3, 10, 11, 16, 22, 23 and 25-30. I found that “fraction 28 and 29” contained the highest concentrations of KLK7 and further analyzed. The purity of KLK7 was confirmed by coomassie blue staining, western blotting and mass spectrometry (data not shown). Three bands were observed on the reduced SDS-PAGE corresponding to molecular masses of ~30 kDa (I), 28 kDa (II) and 20 kDa (III), respectively. Mass spectrometry analysis validated the presence of KLK7 in all three bands, as also confirmed by western blots (Fig. 2.3A-B). The molecular mass of full- length mature KLK7 is predicted to be approximately 25 kDa, and KLK7 has one predicted glycosylation site at N246DT, which is located away from the KLK7 catalytic triad and substrate binding pocket (Fig. 2.3C). PNGase F treatment suggested the ~30 kDa band was the glycosylated form of KLK7, while the ~28 kDa and ~20 kDa bands were non-glycosylated forms (Fig. 2.3 B). The N-terminal sequence of 30 kDa and 28 kDa bands was found to be I30IDGA, which was in agreement with the N-terminal sequence of active, full-length KLK7; N-terminal sequence of the 20 kDa band was S105TQTHV, indicating that this isoform of KLK7 was formed via a proteolytic cleavage at Tyr104 (likely auto-cleavage).

Fig 2.3. Characterization of purified recombinant active of protein KLK7. Panel A, left. Coomassie-stained reduced SDS-PAGE of purified mat-KLK7 (mat=mature). Two chromatographic fractions are shown, fraction 28 (lanes 1, 3, 5) and 29 (lanes 2, 4, 6). Three bands are visible, I (30 kDa), II (28 kDa) and III (20 kDa). All three bands were found to contain KLK7 sequences by mass spectrometry and reacted with anti- KLK7 antibodies on western blots (middle panel). On gelatin zymography (right panel) only bands I and II had enzymatic activity. Panel B. Coomassie-staining of reduced SDS-PAGE (left) and western blotting analysis (right) of purified mature KLK7 (FPLC fraction 28 and 29) with or without PNGase F treatment. * refers to PNGase F band, while arrow refers to 20 kDa band of KLK7. Panel C. Amino acid sequence of active-form of KLK7. Catalytic amino acids are shown in brackets. The N-terminal sequence of bands I and II were identical (IIDGAP) and in accordance with the sequence of active mature KLK7. The N-terminal sequence of band III was STQTHV, indicating that this form of KLK7 originated via proteolytic (likely autoproteolytic) cleavage after Tyr 104. Boxed amino acids indicate data from N-terminal sequencing and underlined amino acid N 246 is the potential glycosylation site of KLK7.

2.3.3 Characterization of recombinant mature form of KLK7

To test the enzymatic activity of purified KLK7, gelatin zymography was performed and indicated that both 30 kDa and 28 kDa bands of KLK7 had enzymatic activity, while the ~20kDa band was not observed on the gelatin gel (Fig. 2.3). Furthermore, N-terminal sequencing of the bottom band (MW: 20KD) represented a truncated form missing H70, which is a major part of the classical catalytic triad of KLK7 (H70, D112, S205), and therefore, this most probably represented an inactive form of KLK7.

To further test the enzymatic activity and substrate specificity of purified KLK7, we employed a panel of AMC fluorogenic substrates. As shown in Table 2.1, recombinant KLK7 clearly showed Y>F preference at the P1 site. No activity was detected against substrates with K or A in P1 sites, while minimum activity (<10% activity compared to trypsin-like KLKs) was observed against substrates with R in the P1 site.

Since environmental factors such as pH can regulate KLKs activity, I performed enzymatic activity assay of KLK7 at different pH buffers (pH 5.5-10). KLK7 exhibited the higher activity at pH values (pH 7.5-9), while decreased KLK7 activity was observed at the acidic pH (5.5-7) and higher basic pH (pH10), with optimal KLK7 chymotrypsin activity observed at pH 8.5 (Fig. 2.4).

The effect of different cations on LLVY-AMC hydrolytic activity of KLK7 was also tested. PMSF was severed as a positive control, and as shown in Fig 2.5, 1mM PMSF completed inhibited the enzymatic reaction. I found that Na+ (final concentration: 10mM), Mg2+ (final: 10mM) and Ca2+ (final: 10mM) exhibited stimulatory effects on KLK7 activity (Fig. 2.5), respectively, although significantly increased KLK7 activity was not observed using low concentrations of these metal cations (final concentration: 1 mM). Interestingly, Zn2+ was found to inhibit KLK7 activity in a dose-dependent manner. I observed that 100 µM ZnCl2 reduced around half of the KLK7 activity, while 1 mM ZnCl2 completely inhibited the activity of KLK7 (Fig. 2.5).

Table 2.1 Kinetic parameters for the hydrolysis of synthetic AMC substrates by KLK7. Chymotrypsin- Km Kcat Kcat/Km Normalized like Vmax (FU/min) (mM) (min-1) (mM-1min-1) Activity LLVY-AMC 8176.5 1.18 59.3 325.8 100% AAPF-AMC 3053.8 0.18 9.6 55 16.9% Trypsin-like Boc-VPR-AMC 1408 0.4 4.1 10.7 Boc-GGR-AMC NR NR Tos-GPK-AMC NR NR Boc-EKK-AMC NR NR Negative Control AAPV-AMC NR NR

NR = no reaction was detected

1.2

0.8

0.6

0.4 Relatedactivity

0.2

0 5.5 6.5 7 7.5 8 8.5 9 10 pH

Figure 2.4 Regulation of the activity of KLK7 by pH. KLK7 was incubated with substrate LLVY-AMC at different pH buffers. The highest activity of KLK7 at pH 8.5 was considered as 1.

1.4

1.2

0.8

0.6

0.4 Relatedactivity

0.2

0 Control 10mM MgCl2 10mM NaCl2 10mM CaCl2 1mM ZnCl2 1mM PMSF -0.2

Figure 2.5 Regulation of the activity of KLK7 by different cations.

Active KLK7 were incubated in the absence (control) or presence of various salt cation solutions (Na, Mg2+, Ca2+ and Zn2+) with AMC-LLVY (final concentration: 0.25 mM). The activity of positive control (without cation) was considered as 1, while PMSF (final concentration: 1mM) was used as an inhibitor control.

2.3.4 Analysis of KLK7 expression pattern in human tissues

Publicly available microarray data profiles from normal human tissues and 8 cancer types were download and analyzed. As shown in Fig 2.6, elevated expression of KLK7 was found in two types of cancer tissues compared to normal tissues, including ovarian cancer tissue and pancreatic cancer tissues. Lower KLK7 expression was found in the rest 5 types of cancer tissues (i.e.. melanoma, kidney cancer, lung cancer, liver cancer and breast cancer tissues) compared to corresponded normal tissues. A slightly different expression pattern was observed between gastric normal tissues and gastric cancer tissues.

Figure 2.6 Expression pattern of KLK7 in 8 types of human normal and cancer tissues. The data were analyzed from publicly available microarray database. The tissues types are ordered as following: ovary, pancreas, lung, skin, kidney, liver, breast and gastric tissue.

2.4 Discussion

As one of the first discovered desquamation KLKs, KLK7 has attracted a lot of attention due to its functions in skin physiology and pathobiology. Plenty of work has been done to explore its physiological targets, inhibitors and (patho)-biological roles. Since KLK3 and KLK9 are not detected in skin, KLK7 solely represents all the chymotryptic-activity of serine proteases in skin (121). All KLKs were secreted as a zymogen form, which contains a pro-peptide to maintain their latent status. In vivo, this pro-form of peptide was cleaved by many proteases or undergo self-active process (121). I expressed the active-form KLK7 in a yeast expression system, and various biochemical assays confirmed its enzymatic activity.

According to previous literature, there was controversy of the P1 specificity of KLK7. One study had shown that recombinant KLK7 preferred Y>F at P1 position (71), while data from MEROPS suggested KLK7 cleaved peptides after amino acid F more than Y. Our result indicated KLK7 clearly had Y>F preference at the P1 site, as we did not detect any activity of KLK7 against substrates with K or A in P1 sites, and minimum activity (<10% activity compared to trypsin- like KLKs) was observed against substrates with R in the P1 site.

KLKs activities can be tightly regulated by many environmental factors, such as pH and cations. pH is a critical factor in maintaining the normal barrier function of skin. The outer layer of normal skin exhibits acidic pH (4.5-5.5), while the inner layers (such as stratum granulosum) have neutral pH. However, in patients with skin disease (i.e. atopic dermatitis), increased pH level at stratum corneum altered the activity of skin proteases (i.e. enhance KLKs and lipid processing enzyme activity), leading to the disrupted skin barrier homeostasis (121,223). Similar to other skin KLKs, KLK7 has an optimal pH at basic condition and less activity at acidic pH (67,73). Of note, we found that KLK7 was still active at acidic pH 5.5, which is in vivo pH of human skin, although with much less activity compared to pH 8.5. All these data suggested hyperactivity of KLK7 in skin inflammatory diseases may due to the elevated pH of skin.

Moreover, pervious reports indicated cations could regulate KLKs activity (10). Na+ , Mg2+ , Ca2+ and K+ were reported to increase KLKs activity (10), while Zn2+ and Cu2+ were found to inhibit KLKs activity. We confirmed that Na+, Mg2+, or Ca2+ indeed enhanced the activity of KLK7, while Zn2+ inhibited it. Our result was consistent with previous crystal structure studies

56 of KLK7, which revealed znic as an inhibitor of KLK7 (16). All the tested ions are known to be present at higher concentrations in skin stratum corneum. For instance, Zn was detected in human sweat and skin with millimolar range (224). Thus, it is very likely that these ions act as important regulators of KLK7 activity in skin.

Beside skin, KLK7 is found to be expressed in many other human tissues and its expression is aberrant in human tumor samples (39,43,49,54,61,225); however, a lot of tissue microarray data were not well analyzed. Here, we analyzed the public available microarray data, and summarized KLK7 expression in 8 different types of cancer tissues compared to normal tissues. Consistent with previous report, elevated expression of KLK7 was found in ovarian cancer tissue and pancreatic cancer tissue compared to normal control (39,56). To date, lower level of KLK7 was only reported in lung cancer and prostate cancer tissues (59-61). Here, we found that the expression of KLK7 was decreased in many other types of cancer as well. Based on our summarized data, we think KLK7 may serve as distinct function regulator in different disease settings.

In summary, recombinant protein KLK7 was generated and its enzymatic activity was validated by various biochemical assays. The active-form recombinant KLK7 protein will used be in the Chapter 3, to search for novel biological substrates for KLK7.

2.5 Author Contributions

YY designed, performed, and analyzed the experiments and wrote the paper. IP helped design and analyze the experiments in Chapter 3 (Figure 3.4-3.5). AD assisted with microarray profile experiments. EPD contributed to the conception and design of the study, and helped draft and revise the manuscript.

Chapter 3 Novel Biological Substrates of Kallikrein-related peptidase 7 Identified Through Degradomics

The work presented in this chapter is published, in the Journal of Biological Chemistry:

Yijing Yu, Ioannis Prassas, Apostolos Dimitromanolakis, Eleftherios P. Diamandis. “Novel Biological Substrates of Human Kallikrein 7 Identified Through Degradomics.” J Biol Chem, 2015, Jul 17; 290(29): 17762-17775. Copyright permission has been granted.

3 Novel Biological Substrates of Kallikrein-related peptidase 7 Identified Through Degradomics 3.1 Introduction

Kallikrein-related peptidases (KLKs) are a group of 15 homologous serine proteases. The KLK genes are co-localized on the long arm of human chromosome 19q13.3-13.4 (13). There is a varying degree of sequence homology among the different KLKs (ranging from 40-80%) (96). KLKs are secreted as inactive zymogens, subsequently activated via proteolytic cleavage of their pro-signal peptides, either by autoactivation or by other proteases. The majority of KLKs display trypsin-like activity, except for KLKs 3, 7 and 9, which are chymotrypsin-like enzymes (96). There is a wide distribution of KLK expression in human tissues and fluids (13). Among the different tissues, significant attention has been brought to the roles of KLKs in skin, central nervous system, kidneys, tooth and the reproductive system (96). Especially in the skin, KLKs are known to participate in a well-characterized proteolytic cascade which regulates skin desquamation, skin barrier function and innate immunity (102,114,121,181). Aberrant regulation of certain skin KLKs (eg KLK5, 7) have been linked to severe skin pathophysiology (i.e. psoriasis, atopic dermatitis, rosacea, Netherton Syndrome and melanoma) (107,109,112,226,227). Despite these recent developments in our understanding of the (patho)physiological roles of KLK7, little is known regarding its endogenous substrates (215).

Several strategies have been previously followed, including yeast two-hybrid screening, combinatorial scanning of peptide libraries, phage display and matrix substrate library, which collectively identified some putative KLK substrates (e.g. e-cadherin, fibronectin, laminin and insulin-like growth factor binding protein 3 (IGFBP3)) (215). However, a common limitation of these methods is their inability to provide direct insights as to how KLKs interact with their substrates in the biological environment. Recently, the emergence of powerful mass spectrometry (MS)-based technologies, such as cell surface degradomics, terminal amine isotopic labeling of substrates (TAILS), protein topography and migration analysis platform (PROTOMAP), combined fractional diagonal chromatography (COFRADIC), and proteomic identification of protease cleavage sites (PICS), have enabled the systematic discovery of endogenous protease substrates (197). In this study, we employed a KLK substrate identification approach, which combines degradomics with sequence-based substrate specificity analysis to

59 identify endogenous KLK7 substrates. Our approach revealed both known (e.g. fibronectin) and new substrates of KLK7 (e.g. midkine (MDK), tenascin-C (TNC), and cysteine rich angiogenic inducer 61 (CYR61)). To validate our findings, we used selected reaction monitoring (SRM) and in vitro degradation assays. We found that KLK7 preferentially cleaves midkine in the presence of other candidate substrates (tenascin-C and CYR61). We further demonstrated that midkine is a substrate for KLK7, but not for other KLKs (i.e. KLK5, 8, 13 and 14). To test whether KLK7- mediated cleavage of new substrates has any effects on their biological function, we used midkine as an example, and found that the cleavage of midkine by KLK7 reduced the pro- proliferative effects and cell migration that were mediated by full-length midkine. Collectively, our data provide further insights into the physiological roles of KLK7 and set the ground for the development of similar degradomics approaches for the substrate profiling of other proteases.

3.2 Materials and Methods

3.2.1 Reagents, cells and antibodies

The melanoma cell lines WM35 (RGP), WM902 (VGP) and WM9 (MET), which represent different melanoma phases (from primary to metastasis), were purchased from the Coriell Institute for Medical Research (Camden, NJ). All melanoma cells were grown in RPMI 1640 media (Gibco, Burlington, Ontario, Canada) supplemented with 10% fetal bovine serum (FBS) (Sigma Aldrich). Spontaneously immortalized human keratinocytes (HaCaT) cultured in DMEM media, were purchased from ATCC and used as previously described (67). The synthetic fluorogenic 7-amino-4-methylcoumarin (AMC) peptides were purchased from Bachem Bioscience (King of Prussia, PA). Human midkine and CYR61 recombinant proteins were purchased from PEPROTECH (Dollard des Ormeaux, Quebec, Canada). Human tenascin-C was purchased from EMD Millipore (Etobicoke, Ontario, Canada). Anti-KLK7 antibodies were generated in-house as previously reported (33). All synthetic peptides with heavy labeled C- terminal arginine or lysine were ordered from JPT Peptide Technologies (Berlin, Germany).

3.2.2 Preparation of cell culture media for proteomic analysis

Melanoma cell lines WM35, WM 902 and WM9 were cultured in media supplemented with 10% FBS in T175 flasks until ~70% confluence. Following wash (with PBS), cells were grown in serum-free medium for 1 day, and then switched to media with or without (vehicle control) active-form of KLK7 (3 µg per flask) for 30 min at 37°C. Media was collected and cell debris removed by centrifugation. The media were concentrated 20x using Amicon® Ultra centrifugal filters (EMD Millipore), and buffer exchanged with 50 mM ammonium bicarbonate. The total protein concentration was measured using the Coomassie Blue protein assay. Proteins were reduced with 10 mM DTT at 50°C for 30 minutes and alkylated with 20mM iodoacetamide at room temperature for 1 hour and were digested overnight at 37°C using trypsin (Sigma) at 1:50 ratio of trypsin: total protein. Formic acid was then added to a final 0.5% (V/V), and the samples centrifuged at 15000g for 10min. The supernatant was transferred into a new tube and stored at - 20°C. For the screening experiment, all samples (100 µg per condition, with or without KLK7 treatment) were fractioned with strong cation exchange high performance liquid chromatography

(SCX-HPLC) (analytic column, PolyLC INC) with Agilent 1100 system. A 60-minute fractionation method was used at a flow rate of 0.2 ml/min (Buffer A: 0.26M formic acid in 5 % acetonitrile, pH2-3, Buffer B: 0.26M formic acid in 5 % acetonitrile with 1M ammonium formate; gradient elution: 10 min buffer A, followed by linear gradient of buffer B 0 % to 20 % in 20 min, then 20-100 % in 15 min. Seventeen fractions per sample were collected and stored for further use.

With regards to the HaCaT cell lines, the experiment was performed as previously reported by Becker-Pauly et al (67). Briefly, HaCaT cells were initially cultured to 70% confluence, and were then switched to serum free medium with (or without) spiked active KLK7 (1:100 ratio of protease/secretome (W/W)). Condition medium was collected 48 hours post-treatment and was subjected to sample preparation prior to MS-analysis.

LC-MS/MS mass spectrometry-The fractionated samples were processed using OMIX-Mini Bed 96 C18 pipette tips (Agilent Technologies). Peptides were washed with buffer containing 95 % water in 0.02 % TFA buffer, and then eluted in 5 µl of buffer B (64.5 %ACN, 35.4 % water, 0.1 % formic acid). For the screening experiment, 80 µl of buffer A (0.1 % FA in H2O) were added to each sample and 40 µl were loaded onto a 2 cm trap C18 column using an EASY-nLC nano-flow pump. The peptides were initially eluted from the trap column onto a 5cm C18 column. The liquid chromatography setup was connected to a Thermo LTQ Orbitrap XL mass spectrometer with a nanoelectrospray ionization source (Proxeon Biosystems, Odense, Denmark). Analysis of the eluted peptides was done by tandem mass spectrometry in positive-ion mode.

Data analysis of mass spectrometry- RAW MS files were generated with the use of the XCalibur software (Thermo Fisher). Files were subsequently analyzed (to DAT files) using the Mascot Daemon software (Matrix Science, London, UK, version 2.2.07). Protein searches were performed against the International Protein Index (IPI) human database (version 3.71), with the following parameters: non-enzyme search, 7 ppm precursor ion mass tolerance, 0.4 Da fragment ion mass tolerance, allowance of one missed cleavage and fixed modification of carbamidomethylation of cysteines. Variable modification included oxidation of methionines. DAT files were further combined using the Scaffold software (Proteome Software Inc., V. 4.3.2), and searched with a Scaffold built in X!Tandem search engine (Global Proteome Machine

Manager, version 2010.12.01.1). Scaffold was used to adjust the levels of false discovery rates (FDRs). With Scaffold analysis, the Decoy FDR at peptide level was 0.19 %, and 0.9 % at the protein level (in the WM9 cell line). In WM902 cell line, the Decoy FDR at peptide level was 0.1 %, and 0.3 % at the protein level. In WM35 cell line, the Decoy FDR at peptide level was 0.11 % and 0.9 % was found at the protein level. Lastly, in the HaCaT cell line, the Decoy FDR at peptide level was 0.09 %, and 0.7 % at the protein level. The cellular location of each protein was analyzed by filtering with UniProtKB database and ProteinCenter (Proxeon Biosystems).

3.2.3 Selected reaction monitoring assay

Method development: peptide sequences were obtained from the scanning results of the LTQ Orbitrap XL mass spectrometer. In the first step of method development, 1 pmole of each synthetic heavy peptide was loaded and analyzed by a triple-quadrupole mass spectrometer (Quantiva, Thermo Scientific). In-silico digestion and fragmentation of each peptide was performed with the PinPoint software (Thermo), and 7-8 transitions per peptide were selected and run in a non-scheduled mode with 1.2s scan time per cycle. The three most intense and selective transitions of each peptide were chosen for the final method (19).The final method contains information for both light and heavy peptide: m/z, retention time, scan time of precursor and fragments. Heavy peptides (100 fmol) were spiked into 10 µg of trypsin-digested sample (before ZipTip processing). The 5 µl eluted samples from the ZipTip were mixed with 60 µl of buffer A (0.1% FA in H2O) and 18 µl was injected.

Sample preparation for monitoring proteolytic reaction: 0.25mM of each recombinant protein or mixtures (midkine, TNC and CYR61) were incubated at 37 °C with active KLK7 (at 1:100 ratio) in 0.1 M sodium phosphate, pH 8.5 buffer. At various time points (5, 15, 30, 45, 60, 90, 120min), the reaction was stopped by adding DTT (final 10mM in 50 mM ammonium bicarbonate) and incubation at 60 °C for 30 min. Iodoacetamide (final 20mM) was then added followed by incubation at room temperature for 1 hour. Proteins were digested overnight at 37°C using trypsin (Sigma) at 1:50 ratio (trypsin: total protein). To confirm the linear range of heavy/light peptide ratio, serial dilutions of heavy peptides (1fmole - 10pmole) were spiked into a set of digested combined recombinant proteins.

3.2.4 In vitro recombinant protein digestion assay

Proteolysis of substrate proteins was performed in KLK7 activity buffer (100mM sodium phosphate buffer, 0.01% Tween-20, pH 8.5). Enzyme and substrate were mixed at ratio of 1:100. The mixtures were incubated at 37 °C and collected at different time points (0, 30min, 1h, 2h, and 3h). The reaction was stopped by boiling with 4x SDS-PAGE sample buffer (Bio-Rad) and analyzed by 4–12% gradient SDS-polyacrylamide gel electrophoresis and silver-stained using the PlusOne Silver Staining kit (GE Healthcare).

3.2.5 Cell proliferation assay

WM9 melanoma cells (104 cells per well) were seeded in 96 well plates and cultured in media (RPMI 1640 with 10% FBS) for 24 h, followed by serum starvation for 24 h. The cells were then treated either with i) full-length midkine (100 ng/ml), or ii) KLK7-fragmented midkine or iii) KLK7 (1 ng/ml) iv) without midkine (vehicle control) for 3 days. After treatment, the cells were mixed with alamarBlue® reagent (Invitrogen) and the proliferation rates were measured according to the manufacturer‟s instruction.

3.2.6 Migration and invasion assay

Cell migration was measured using the cell migration assay kit (CULTREX®, USA). The cells were trypsinized, counted, and then resuspended in medium supplemented with or without treatment (as described in proliferation assay) to 106 cells/ml. Cells (5*104 / well) in each treatment were loaded into the top chamber of a 96 well plate containing a 8-micron polyethylene terephthalate membrane. After 24 hours incubation, free cells were washed off. The migrated cells were dissociated using the cell dissociation buffer containing Calcein-AM. The Calcein fluorescence released by the intracellular esterases was measured by Wallac Envision Multilabel Reader (PerkinElmer Life Science) at 485 nm excitation, 520nm emission, to quantitate the number of migrated cells.

Cell invasion assay was performed according to the instructions of the cell invasion assay kit (CULTREX®). The top chamber of the invasion plate was coated with basement membrane extract (BME) one day before the assay. Cells (106 cells/ml) were starved one day before the assay and resuspended in medium supplemented with or without treatment (as described in proliferation assay) on the experiment day. Cells (5*104 / well) in each treatment were loaded into the BME coated top chamber of a 96 well plate. After 24 hours incubation, free cells were washed off. The invaded cells were dissociated using the cell dissociation buffer containing Calcein-AM. The Calcein fluorescence released by the intracellular esterases was measured as described above.

3.2.7 Microarray profiles

As described in Chapter 2.2.3, briefly, we downloaded publicly available microarray data profiles of cancer and normal human from the NCBI GEO repository of microarray experiments (220). All data were loaded in a common array platform (Affymetrix HG-U133 Plus2) to generate raw CEL files for further processing. The CEL files were further normalized using R 2.15.2 and the Bioconductor platform v2.8 (221). Simpleaffy was used to quality control metrics each array (average background, RNA degradation, scale factors, percent present). After quality control, gcRMA array normalization was used to normalize expression data(222).

3.2.8 Data analysis

All cell proliferation, migration and invasion assay were analyzed by one-way ANOVA (GraphPad Prism Software). Results comparing different treatments were considered significant if the p-value was less than 0.05. All data are expressed as means ± standard error of the mean (SEM).

3.3 Results

3.3.1 Preparation of cell culture media for proteomic analysis

To systemically search for novel physiological substrates of KLK7, we conducted comparative proteomic analysis of peptides in the supernatants of cell cultures (cell lines melanoma WM9, WM902, WM35 and HaCaT) treated with KLK7 or untreated as a control (the experimental outline is shown in Fig. 3.1). Our KLK7 ELISA revealed that endogenous KLK7 expression levels are negligible in these cell lines (data not shown). This approach could enable identification of KLK7 substrates expressed on the cell surface or released into the extracellular space. The latter proteins include actively secreted intracellular proteins, shed cell surface proteins, or non-specifically released proteins from damaged or dead cells. It is well-known that KLK7 is secreted as an inactive zymogen, which is subsequently activated in the extracellular space; thus, the peptides released from KLK7 proteolytic cleavage, should theoretically originate from cell surface or extracellular proteins. Notably, the peptides generated by KLK7 are predicted to contain a chymotryptic P1 cleavage site (P1: phenylalanine or tyrosine). Peptides with P1: Arg or Lys was also found in both culture supernatants, since all samples were digested with trypsin prior to mass spectrometry.

Our MS analysis revealed 2081 unique peptides in the KLK7-treated WM9 cell culture supernatant (Fig. 3.2). One hundred and sixty chymotryptic peptides (corresponding to 109 proteins) were found exclusively in the KLK7-treated samples and were selected for further scrutiny as originating from candidates of KLK7 substrates. Next, gene ontology was used to exclude intracellular proteins, which resulted in a total of 15 proteins as putative KLK7 substrates (Fig. 3.2). To further validate our approach, a similar strategy was used to study putative KLK7 substrates in other cell lines, such as the melanoma cell lines WM902 (vertical growth phase) and WM35 (rapid growth phase). As shown in Fig. 3, we identified 126 putative KLK7 substrates peptide (corresponding to 88 proteins) in the WM902 cell line and 208 putative KLK7 substrates peptide (corresponding to 138 proteins) in the WM35 cell line. Filtering for extracellular and/or membrane proteins resulted in 8 and 10 candidate KLK7 substrates in the WM902 and WM35 cells, respectively. We also followed this strategy with the human normal keratinocyte cell secretome (HaCaT)(Fig. A1).

Figure 3.1 Workflow of our degradomics approach and criteria for candidate selection. Two groups of cell conditioned media are prepared: one treated with active KLK7 (KLK7 digestion) and one without KLK7 treatment (control). The cell supernatants are then trypsinized for MS preparation. The processed cell supernatant can therefore contain peptides with KLK7/KLK7, KLK7/trypsin, trypsin/KLK7, and trypsin/trypsin cleavage sites. Selection criteria were applied to the list of proteins identified by Mascot and included: unique peptides in KLK7 treatment samples, peptides which have at least one KLK7 cleavage site and from proteins with extracellular or cell surface protein localization.

Figure 3.2 The number of proteins identified after applying the selection criteria in three cell lines samples. Melanoma WM9, WM902 and WM35 cell lines were processed as described in FIG. 2. Venn diagrams show the number of peptides identified in control and KLK7-treated cells. For specific peptide selection criteria see text and legend of figure 3.1.

Among the identified candidate substrates, there are proteins, such as fibronectin and cadherin 1 which have previously been reported as putative substrates of KLK7, further supporting the validity of our approach (215). Interestingly, novel KLK7 candidate substrates (e.g. midkine, tenascin-C) appeared as putative hits in all three different melanoma cell lines and in normal human skin keratinocyte secretome (Fig. 3.2).

Based on the results of the three skin cell lines, we generated a list of novel putative candidate substrates. Previously reported KLK7 substrates such as fibronectin, collagen and laminin were excluded (215). Proteins with unknown functions in UniProtKB database were also removed, including Inter-alpha (Globulin) inhibitor H2, isoform CRAa. Also, protein such as albumin and KLK7 were also excluded, as they were added to the culture. Lastly, proteins with major function in the cytoplasm were removed, including actinin and casein kinase II subunit beta. A list of novel candidate substrates is presented in Table 3.1.

3.3.2 Validation of candidates with an SRM assay in WM9 cultures

To further validate the candidate KLK7 substrates, we performed SRM assays targeting the specific peptides generated by KLK7 cleavage. Cell supernatants were digested by trypsin following incubation with or without KLK7. Since KLK7 has substrate specificity different from trypsin, KLK7-cleaved peptides can be identified as having KLK7/trypsin or KLK7/KLK7 P1 sites. Tryptic only peptides (trypsin/trypsin) were used to confirm that the targeted proteins are detected in the samples. The candidates analyzed by SRM assay were selected based either on their detection frequency in different cell lines or reported as KLK substrates in the literature. The endogenous peptides from cultured cells were light peptides, while synthetic peptides with heavy isotope-labeled Arg/Lys (13C6, 15N4-Arg; 13C6, 15N2-Lys) were spiked into the samples and light/ heavy area ratios were calculated for each peptide. We were able to detect tryptic peptides of many proteins, such as midkine, tenascin-C, CYR61, CTGF, TGF-β1 and insulin-like growth factor binding protein 3 (IGFBP3), in both KLK7-treated and untreated WM9 samples, indicating that these proteins can be detected in the culture supernatants independently of KLK7 treatment. However, we were able to detect the KLK7- specific peptides of midkine, tenascin-C and CYR61 only in KLK7-treated cultures (Table 3.2).

Table 3.1 Candidate substrates of KLK7 based on the screening results from melanoma and HaCaT cell lines.

Protein ID Specific peptide sequence Found in cell lines

WM9,WM902,WM35, Midkine (MDK) Y-NAQCQETIR-V HaCaT

R-AQSEGRPCEY-N, R- Protein CYR61 (CYR61) WM9 ICEVRPCGQPVY-S

Y-TVTLHGEVR-G WM9, WM902,WM35

F-TTIGLLYPFPK-D WM9, HaCaT

Y-SLADLSPSTHYTAK-I WM902,WM35 Tenascin-C (TNC) Y- LSGLAPSIR-T WM35

Y-TGEKVPEITR-A WM35

R-TAHISGLPPSTDFIVY-L WM9

Connective tissue growth Y-RLEDTFGPDPTMIR-A, K- WM9 factor (CTGF) TCACHYNCPGDNDIFESLY-Y

Isoform 1 of Extracellular Y-HVGLGDAAQPR-N WM902, WM35 sulfatase Sulf-2 (SULF2)

Fructose-bisphosphate K-CPLLKPWALTF-S, WM9, WM35, HaCaT

aldolase A (ALDOA) F-LSGGQSEEEASINLNAINK-C WM902

Y-KTDLEKDIISDTSGDFRK-L, WM9, WM902 Isoform 1 of Annexin A2 (ANXA2) Y-TNFDAERDALNIETAIK-T WM9, HaCaT

Isoform 1 of Complement C1q tumor necrosis factor- F-AGFLLFETK WM9, WM35 related protein 3 (C1QTNF3)

Spondin-2 (SPON2) F-SAPAVPSGTGQTSAELEVQR-R WM902

Complement C4-A (C4A) F-LSCCQFAESLR-K WM9

Insulin-like growth factor Y-KVDYESQSTDTQNF-S WM902, HaCaT binding protein-3 (IGFBP3)

Table 3.2 Selected reaction monitoring assay validation of specific peptides in cell culture models. Total of five biological samples per condition (control or KLK7 treated samples) were tested in this study.

Protein ID Peptides type Peptides sequence WM9

Control KLK7 treated

TNC Tryptic peptide ITAQGQYELR 5/5* 5/5

KLK7 peptide TVTLHGEVR 0/5 5/5

MDK Tryptic peptide YNAQCQETIR 5/5 5/5

KLK7 peptide NAQCAETIR 0/5 5/5

CYR61 Tryptic peptide LPVFGMEPR 2/5 5/5

KLK7 peptide ICEVRPCGQPVY 0/5 5/5

CTGF Tryptic peptide LPSPDCPFPR 5/5 5/5

KLK7 peptide RLEDTFGPDPTMIR 0/5 0/5

IGFBP3 Tryptic peptide YGQPLPGYTTK 5/5 5/5

KLK7 peptide KVDYESQSTDTQNF 0/5 0/5

*x/y = number of times peptide was detected/total number of independent biological samples. For full name of protein see Table 1.

3.3.3 Validation of midkine, tenascin-C and CYR61 as novel substrates for KLK7 by in vitro recombinant protein digestion assay

In vitro recombinant protein digestion assay is a common approach to validate enzymatic substrates (215). We first tested some known substrates of KLK enzymes, such as fibronectin, laminin and IGFBP3, and found that active KLK7 was able to cleave these substrates (data not shown). The recombinant midkine, CYR61 and tenascin-C proteins were then tested in a time- course digestion assay using KLK7 at enzyme: substrate ratio of 1:100 (w/w). The digest was analyzed by SDS-PAGE, silver or coomassie blue staining. The enzyme: substrate ratio of in vitro recombinant protein digestion was selected based on previous studies (161). Fragmentation of midkine was visible after 30 min incubation with KLK7 and complete digestion was observed after 3 hours (Fig. 3.3A). Midkine alone did not degrade after overnight incubation at 37oC, but degraded in the presence of KLK7 (data not shown). A similar KLK7 digestion pattern was observed for CYR61 and tenascin-C (Fig. 3.3B-C). Thus, the in vitro digestion assays further demonstrated that these three novel candidate proteins were potential substrates for active KLK7.

3.3.4 In vitro digestion of midkine, CYR61 and tenascin-C by other skin KLKs

To test whether the aforementioned three new candidate substrates were KLK7-specific, these recombinant proteins were incubated with active KLK 5, 8, 13 and 14 at a ratio of 1:100 (w/w), respectively. Midkine appeared to be a KLK7-specific substrate, since other tested KLKs were not able to cleave it (Fig. 3.4A). In contrast, KLK14, but not the other KLKs, were able to digest CYR61, while tenascin-C also appeared to be cleaved by KLK5 and KLK8 (as indicated by degradation of top band) (Fig. 3.4B-C). Interestingly no signs of self-degradation was observed after 3h incubation of all tested KLKs at 37oC (Fig 3.4D). These results suggest a possible redundancy in KLK substrate cleavage.

Figure 3.3 In vitro proteolytic processing of substrate candidates by KLK7. Briefly, 100 ng of recombinant MDK (750nM; A), CYR61 (250nM; B) and 2.5 µg of TNC (330nM; C) were incubated with active KLK7 for different times (0, 30 min, 1 hour, 3 hours and overnight; O/N). The reaction was stopped and the cleaved products were then separated by 4-12% SDS-PAGE followed by silver staining or coomassie blue staining. The arrows indicate the changes of band intensity over time.

Figure 3.4 processing of In vitro proteolytic substrate candidates by other KLKs. KLK5, 7, 8, 13 or KLK14 was tested (all produced in-house as recombinant proteins). Briefly, 100 ng of recombinant MDK (A), CYR61 (B) and 2.5 µg of TNC (C) were incubated with active KLK5, 7, 8, 13 or 14 for 3 hours, respectively. KLK5 (2.5 nM), KLK7 (3.3 nM), KLK8 (3.1 nM), KLK13 (3.1 nM) or KLK14 (3.2 nM) was used to digest midkine and CYR61. For TNC, KLK5 (62.5 nM), KLK8 (83nM), KLK8 (83 nM), KLK13 (83 nM) or KLK14 (83 nM) was added into the reaction. (D) KLK5, 7, 8, 13 or 14 alone was incubated at 37 °C for 3 hours or O/N, respectively. The reaction was stopped and the cleaved products were then separated by 4-12% SDS-PAGE followed by silver staining or coomassie blue staining. The arrows indicate the cleaved products.

3.3.5 In vitro monitoring of proteolytic processing of substrates by KLK7

To study the activity of KLK7 against the novel substrates, we monitored the KLK7-mediated digestion of the recombinant proteins (i.e. midkine, CYR61 and tenascin-C) over time using the SRM assay with KLK7-specific released peptides. Heavy peptides served as relative quantification standards and tryptic peptides served as internal loading controls. The amounts of KLK7-specific released peptides were quantitated as a ratio of light-to-heavy peptides. As shown in Fig. 3.5A, the KLK7-cleaved peptide Y-NAQCQETIR-V (midkine) was not detected after 0 min incubation (control) but appeared after 5 min incubation and the signal nearly saturated after 2 hour-incubation. The associated tryptic peptide for this substrate R-YNAQCQETIR-V, was also monitored and was found to be at the highest level at 0 min, followed by decline over time, as expected (Fig. 3.5A). Similar digestion kinetic patterns were observed for tenascin-C but not CYR61 (Fig. 3.5B-C). To identify which novel candidate was the most preferred substrate, we incubated all three substrate proteins (same molar amount) with KLK7 and observed the preferential cleavage of midkine, compared to the other two substrates (Fig. 3.5D). These results suggest that midkine was the most favorable substrate of KLK7 among these three novel substrates under the tested conditions.

3.3.6 Biological effects of KLK-mediated cleavage of new substrates

To test whether KLK7-mediated digestion of new substrates has biological significance at the cellular level, we selected midkine as an example. Since midkine is a hairpin-binding growth factor, which enhances cell survival and migration, while inhibiting apoptosis in many cancers (228), we wondered whether KLK7-mediated midkine digestion decreased the potential biological effects of full-length midkine in melanoma. We cultured the melanoma cells with recombinant midkine, in the presence or absence of KLK7. We found enhanced proliferation of WM9 cells was stimulated by full-length midkine, but not by the KLK7-cleaved midkine or by KLK7 alone (Fig. 3.6A, P<0.05). We also observed that the cleavage of midkine by KLK7 decreased the melanoma cell migration that was stimulated by intact midkine (Fig. 3.6B, P<0.05). Furthermore, the cell invasion assay showed that intact midkine, but not KLK7-cleaved midkine, tended to enhance WM9 cell invasion, although the results were not statistically significant (Fig.

3.6C). We also observed that KLK7 alone did not significantly affect the migration and invasion of WM9 cells (data not shown). These results suggest that KLK7-mediated digestion of substrates could have a significant effect on their biological function.

The biological relevance of KLK7-mediated substrate cleavage also depends on the co- expression/colocalization in time/space under physiological/pathophysiological conditions of KLK7 and the putative substrate. To this end, we analyzed gene expression patterns of some of the candidate substrates (Table 3.1) in eight normal and cancer tissues (Fig. A2). As shown in Chapter 2.3.4, KLK7 was highly expressed in several cancer types (Fig. 2.5). Consistent with previous studies (228), we found that the expression of midkine was elevated in all types of cancers, compared to normal tissues. Interestingly, we observed that the expression levels of midkine and KLK7 seemed to be negatively associated, when comparing melanoma with normal skin samples (Fig. 3.7A-B).

Figure 3.5. SRM based time course analysis of KLK7-mediated protein cleavage. - Briefly, 0.25 µM of recombinant MDK (A), CYR61 (B) and TNC (C) were incubated with KLK7 (enzyme: substrate ratio of 1:100 (w/w)) at various time points at 37 °C. The reaction was stopped and the peptides monitored using the SRM assay. (D), 0.25 µM of each substrate were mixed and incubated with KLK7 (enzyme: substrate ratio of 1:100 (w/w)). The cleaved products were monitored at several time points. Digestion rates of TNC, MDK and CYR61 cleaved by KLK7 were compared. The Y-axis represents peptide abundance, calculated as the ratio of light/heavy peptide area; for (A-C) the highest ratio observed in each experiment was considered as 1. Peptide abundances were fit to a well-established pseudo-fist-order kinetic equation (A/A0= −(kcat∕Km*Eo*t) 1- e with A/A0: relative abundance and E0: enzyme concentration) to determine the activity pattern for each substrate(229). HD = hairpin binding domain.

Figure 3.6. Functional studies of MDK and KLK7-treated MDK (MDKT) in the WM9 cell line. (A) Proliferation of WM9 cells stimulated with MDK or MDKT. WM9 cells were cultured for 72 hours in RPMI1640 medium without additive (RPMI), with 100ng/mL midkine (RPMI+MDK), or with KLK7-treated midkine (RPMI+MDKT), or 1ng/mL KLK7 alone. MDKT truncated form was generated by pre-incubation of MDK with KLK7 (100:1 ratio) at 37 °C for 3 hours. The proliferation rate of cells in RPMI alone was considered as 100%. The data are presented as means ± SEM (n=6, P<0.0001). (B) The WM9 cell migration assay. WM9 cells were cultured in RPMI serum-free medium without additive, with 100ng/ml MDK, or with 100ng/ml MDKT. The migration rate of cells in RPMI alone was considered as 100%. P=0.0114. (C) The WM9 cell invasion assay. WM9 cell were cultured in RPMI serum free medium without additive, with 100ng/ml MDK, or with 100ng/ml MDKT. The invasion rate of cells in RPMI alone was considered as 100%. P=0.4175. *P<0.05.

Figure 3.7. A, expression pattern of MDK in 8 types of human normal and cancer tissues were analyzed from publicly available microarray database. The tissues types are ordered as following: ovary, pancreas, lung, skin, kidney, liver, breast and gastric tissue. B, expression pattern of MDK and KLK7 in normal skin and melanoma. For more details see text.

3.4 Discussion

In this study, we report a strategy to systematically search for KLK substrates. This approach, which combines degradomics and substrate specificity analysis, successfully identified both known (e.g. fibronectin) and new substrates for KLK7. Using SRM and in vitro recombinant protein degradation assays, we validated the specificity of several new substrates (i.e. midkine, tenascin-C and CYR61). We also found that the cleavage of midkine by KLK7 reduced the pro- proliferative effects and cell migration that were mediated by full-length midkine. These results indicated that novel KLK substrates identified through our screening approach were biologically relevant, and that the KLK-mediated cleavage of these substrates has significant effects on their biological function.

The KLK family, comprised of 15 members, is the largest serine protease family in humans. Secreted KLKs are inactive zymogens, which are activated extracellularly by cleavage of their pro-peptides. Through a constant interplay with others proteases and endogenous inhibitors, KLKs participate in many physiological and pathophysiological processes, such as skin desquamation, semen liquefaction, neuron degeneration, tissue remodeling, wound healing and tumor metastasis (13,96). A deeper understanding of the physiological roles of KLKs is currently hampered by the absence of knowledge regarding the physiological substrates of these enzymes.

In the past, several efforts have been made for the identification of KLK substrates. For example, using the yeast two-hybrid (Y2H) system, it has been found that KLK4 and KLK14, but not KLK3 and KLK7, can interact with sex hormone-binding globulin (SHBG) (230,231). In addition to Y2H, chemical or biological peptide library scanning has also been used to investigate substrate cleavage sites for several KLKs. For example, substrates for KLK1, KLK2, KLK3-7, KLK10-11 and KLK14 have been determined with the use of chemical peptide library scanning (169,172,199), while specificity and potential biological targets for KLK1, KLK2, KLK4, KLK6 and KLK14 have been scanned by phage display screening (158,198,202). Some of the limitations of these approaches include detection of interaction but not activity per se, use of short peptides (not whole proteins) and detection of substrates under non-native conditions.

New strategies based on mass spectrometry, such as degradomics, TAILS and PROTOMAP, can systemically identify and quantify enzymatic substrates in biological samples; but most of these

81 advanced approaches have been as yet not utilized in the KLK field (215). To date, among the 15 KLKs, only KLK12 has been screened for its physiological substrates by a degradomics approach (161). In the latter study, tryptic peptides were compared between KLK12-treated and non-treated samples. Since KLK12 is a trypsin-like enzyme, the cleaved peptides generated from KLK12 or trypsin digestion could not be distinguished and the detection of KLK12 substrates depended on identification of peptides only present in KLK12-treated samples. To overcome this weakness, we improved on this traditional degradomics approach by integrating cleavage site sequence analysis, which could efficiently filter out non-specific peptides. Our approach can be applied in other serine proteases by replacing trypsin with other enzymes, during sample preparation.

To test our screening strategy, we used KLK7 as an example, since it is one of the most abundantly expressed KLKs in many tissues (121). Theoretically, any cell line could be used as a natural source of candidate substrates. We selected several cells lines that have previously been used to study KLK7. KLK7 is highly expressed and it is active in skin (121). It may therefore be expected that some biological KLK7 substrates derive from cells in the skin. Furthermore, as one of the most aggressive skin cancers, melanoma was shown to have better prognosis in patients with high expression level of KLK7 (125). WM9 and WM902 cell lines have been used in melanoma progression studies, including KLK interaction with proteinase-activated receptors (PARs) (184,232). In addition, HaCaT was one of the cell lines used to study the functional role of KLKs in both normal skin and skin inflammatory diseases (67,233). Furthermore, HaCaT cells had been used to identify substrates for metalloproteases (234).

Our screening approach identified several known substrates for KLK7, such as fibronectin and cadherin 1. We also identified new KLK7 substrates, such as midkine, CYR61 and tenascin-C. To validate these putative substrates, we employed both SRM and in vitro digestion assays. Midkine, a candidate substrate for KLK7, has been previously investigated as a source of antimicrobial peptides, and as a heparin-binding growth factor (235,236). Midkine was found to be expressed in various cell types and promoted cell growth, survival, migration and angiogenesis (235,236). In normal skin, midkine has been detected in eccrine sweat glands, root sheath, and the upper to middle layer of the epidermis (237). In diseased skin, midkine was found to be highly expressed in keratinocyte-proliferating layers of fungal dermatitis, psoriasis,

82 lichen planus, and wart and molluscum contagiosum (237,238). Enhanced expression of midkine was also observed in various cancers, such as neuroblastomas, astrocytomas, pancreatic cancers and gastrointestinal stromal tumors, and promoted cancer metastasis (228,239). Furthermore, midkine was associated with an anti-apoptotic role in oral squamous cell carcinoma, leukoplakia and meningiomas, which promoted cancer cell metastasis (240,241). Based on our bioinformatic data, we found that the expression of midkine was elevated in melanoma, compared to normal skin tissues. An inverse association pattern was observed between KLK7 and midkine expression (Fig. 3.7B). High levels of KLK7 expression have been linked to good prognosis in melanoma patients (125) and thus, the inverse relationship between KLK7 and midkine expression suggest that a potential protective role of KLK7 in melanoma may be mediated through its direct cleavage of midkine. Midkine cleavage appeared to be KLK7-specific, since other tested KLKs (e.g. KLK 5, 8, 13 and 14) were not able to cleave it. Here, we also found KLK7 could cleave tenascin-C and CYR61. Tenascin-C is an ECM component, which plays an essential role in wound healing, nerve regeneration, tissue remodeling, tumorgenesis and metastasis (242). CYR61 belongs to the CYR61, CTGF, (nephroblastoma overexpressed) NOV (CCN) family, which were reported as putative substrates for KLK12 and KLK14 (161), and contribute to cancer progression through interaction with various cytokines (i.e. VEGF). Here for the first time, we report that KLK7 could proteolyze CYR61. Another member of NOV family-CTGF (connective tissue growth factor), was also identified as a putative KLK7 substrate during our screening experiment. We also tested the efficacy of KLK7 to cleave recombinant CTGF protein in vitro and indeed we found a clear cleavage of this substrate by KLK7. However, we failed to reproduce this finding in the validation phase (using SRM assays), which may be due to sample preparation or technical problems during SRM assay (i.e. masking effect, sample loss). IGFBP3- specific peptide was also not detected in the SRM validation of WM9 cell lines, which confirmed the initially shotgun results that the peptide was found in WM902 and HaCat cells, but not in WM9 cell lines.

With the development of high-throughput proteomic techniques, more and more novel substrates are being discovered at an increasing rate for a number of proteases. However, the identification of substrates using in vitro biochemical experiments does not necessarily imply that these are the actual functional biological substrates in the tissues (243). For instance, the concentration of the spiked enzyme in an in vitro setting might not reflect the actual physiological amount of this

83 enzyme in vivo. In general, the in vivo setting is a very dynamic environment in which each protein displays a unique time- and spatial distribution of interaction with its endogenous partners. With these limitations in mind, we tested whether KLK7-mediated cleavage of substrates may affect their proposed biological function (using midkine as an example). Based on our results, the KLK7 cleavage site (LKKARY90-NAQCQETIR) was located in the long loop of midkine, which was predicted to be important for hairpin binding and midkine dimerization (244). Thus, cleavage of midkine by KLK7 may disrupt the binding ability to midkine receptors, which mediate its biological functions. Indeed, we found that the cleavage of midkine by KLK7 reduced the cell proliferation and migration that were mediated by full-length midkine in cell lines. These data suggested that KLK7-mediated digestion of new substrates has significant effects on their biological function. But further studies to test whether KLK7-mediated midkine cleavage affects the function of midkine in more physiological conditions (e.g. in vivo animal models) are warranted.

In summary, our degradomics approach combined with sequence-based specificity analysis, identified known and novel substrates for KLK7. Some of the novel substrates (i.e. midkine, CYR61, tenascin-C) were subsequently validated by in vitro digestion and SRM assays. Furthermore, we demonstrated that cleavage of midkine by KLK7 reduced the pro-proliferative effects and cell migration that were mediated by full-length midkine. Our approach is generally applicable to identify new substrates for other KLKs and non-KLK enzymes. Identification of such substrates will help to understand the roles of KLKs and non-KLK enzymes in various physiological and pathophysiological processes.

3.5 Author Contributions

YY designed, performed, and analyzed the experiments and wrote the paper. IP helped design and analyze the experiments in Figures 3.4-3.5. AD assisted with microarray profile experiments. EPD contributed to the conception and design of the study, and helped draft and revise the manuscript.

Chapter 4

Proteomic and peptidomic analysis of human sweat with

emphasis on proteolysis

A manuscript based on the work presented in this chapter is accepted by Journal of Proteomics:

Yijing Yu, Ioannis Prassas, Carla M. J. Muytjens, Eleftherios P. Diamandis. “Proteomic and peptidomic analysis of human sweat with emphasis on proteolysis.” (accepted by Journal of Proteomics)

4 Proteomic and peptidomic analysis of human sweat with emphasis on proteolysis 4.1 Introduction

Sweat is a slightly acidic (pH 4.0-6.8) biological fluid secreted by skin eccrine and apocrine glands upon stimuli of the sympathetic nervous system. It consists of water (99%), amino acids, urea, metal/non-metal ions and metabolites. The two established roles of sweat are thermo- regulation and host infection defense (245). From a biochemical point of view, human sweat represents a natural skin „wash-off‟ that reflects the composition and activity of human skin.

The view of human sweat as a promising source of disease biomarkers is not new. Early proteomic studies have shown that the protein composition of human sweat is highly dynamic and can alter significantly in various skin and other disorders, such as ectodermal dysplasia, atopic dermatitis, and even schizophrenia (246-248). Compared to other diagnostic biological fluids (e.g. blood, cerebrospinal fluid, ascites) sweat has the extra advantages of non- invasiveness and easy retrieval. From a clinical perspective, the best illustration of the use of sweat as a diagnostic fluid is the development of the „sweat test‟ for newborn cystic fibrosis screening (CF) (249). This test works by measuring the concentration of sweat salt as a reliable way to diagnose CF. More recently, additional candidate diagnostic applications of sweat analysis have been reported, including screening for lung cancer or drugs of abuse biomarkers (250,251).

Despite these applications, the full potential of using sweat for diagnostics has yet to be realized. To some extent, this can be attributed to technological limitations that until recently prevented the full delineation of its proteomic composition. Notably, all previous proteomic analyses of human sweat have revealed less than 600 proteins (246,247,252-254). For instance, the first proteomic analysis of human sweat by Park et al. identified 115 sweat proteins, followed by a study by Raiszadeh et al. which identified 213 proteins (with some of them being the same in the two studies) (247,252). Furthermore, Burian et al. identified 544 proteins from skin wash samples2. Additionally, Csősz et al identified 95 sweat proteins from healthy people (253). More recently, Adewole et al. identified 117 proteins from the sweat of healthy and lung diseases‟ patients (i.e. tuberculosis) (254).

Besides proteomics, peptidomics is another MS-based technology, which was not explored in the context of human sweat (216). Most peptides in biological systems derive from proteolytic cleavage, mediated by endogenous peptidases. Therefore, imbalances in the activities of endogenous proteases and/or their inhibitors may result in distinct peptide patterns that can be captured by MS-based proteomic technologies. The last decade, a plethora of peptidomics studies of several biofluids (i.e. serum, plasma, urine, cerebrospinal fluid and saliva) have been reported, but not for sweat (216-218).

To address these unmet needs, we performed deep proteomic and peptidomic analysis of sweat from healthy humans. In total, total 861 unique proteins were identified during our proteomic analysis and an additional 32,818 endogenous peptides (corresponding to 1,067 proteins) from our peptidomics workflow.

4.2 Materials and Methods

4.2.1 Sample collection and preparation

Sweat samples were collected from 10 healthy (5 male & 5 female) volunteers using absorbing tissue pads, following strenuous physical exercise. All individuals were 20-60 year of age. A standard protocol was used, as follows (255). Volunteers showered the night before sample collection and were instructed not to apply any skin cream or other skin product until sweat was collected the following day. As shown in Fig. 4.1, the sweat samples were collected from the forehead, neck, chest, arms and back of the volunteers. The two sweat pools (one from the 5 males and one from the 5 females) were prepared by mixing 2 mL of sweat sample from each volunteer (final pool volume of 10mL). Both pools were then centrifuged at 4,000g for 10 min at 4°C, followed by sample ultrafiltration using 10kDa cutoff Amicon® Ultra centrifugal filters (Millipore). The high molecular weight retentate (around 0.5mL) was used for proteomic analysis, whereas the lower molecular weight flow-through (filtrate; around 9.5mL) was used for peptidomic analysis. Protein concentration was determined with a coomassie protein assay (Thermo Scientific). As shown in Fig. 4.1, the sample preparation methods for the two protocols (proteomics and peptidomics) were performed in parallel, as briefly described below: i) Proteomic analysis: Samples were buffer-exchanged with 50 mM ammonium bicarbonate using 3kDa Amicon® Ultra centrifugal filters (Millipore). Proteins were then reduced and alkylated (with 10 mM dithiothreitol and 20 mM iodoacetamide) and digested overnight at 37°C using trypsin (Sigma). Following overnight incubation, formic acid was added to the samples at a final concentration of 0.5% (v/v). The samples were centrifuged at 15,000g for 10 min and stored at -20°C until further analysis. ii) Peptidomic analysis: Unlike proteomic analysis, samples were prepared without enzyme digestion. Native peptides were desalted after sweat was ultracentrifuged with 10KDa cutoff filters and concentrated with a hydrophilic-lipophilic-balanced reversed-phase cartridge (Oasis HLB), as previously described (217,218). Briefly, the cartridge was pre-equilibrated with 1mL of 90 % equilibration buffer (acetonitrile (ACN), 0.1 % formic acid (FA) and 0.02 % trifluoroacetic acid (TFA)), and was further washed with 15mL of buffer A (5 % ACN, 0.1 % FA and 0.02 % TFA) prior to sample loading. The sweat samples were acidified to pH 4 with formic acid and

88 then passed through the cartridge. The cartridge was washed with 5mL buffer A, and peptides eluted with 700µL of elution buffer (60 % ACN, 0.1 % FA and 0.02 % TFA). The eluted samples were concentrated 5-fold with a speed vacuum concentrator, and stored at -20°C until further analysis.

Figure 4.1. Workflow for sweat proteomics and peptidomics sample preparation.

4.2.2 Strong cation exchange chromatography

Both proteomics and peptidomics samples (300 µg per pool) were subjected to further fractionation by strong cation-exchange high performance liquid chromatography (SCX-HPLC) (The Nest Group Inc. Southborough, MA) on an Agilent 1100 system. A 60-minute fractionation method was used at a flow rate of 0.2 mL/min (Buffer A: 0.26M formic acid in 5% acetonitrile, pH 2-3, Buffer B: 0.26M formic acid in 5 % acetonitrile with 1M ammonium formate; gradient elution: 10 min buffer A), followed by linear gradient (0 % to 20 % in 20 min, then 20-100 % in 15 min) of buffer B. Twenty-six fractions per sample were collected and were subjected to MS analysis as described below.

4.2.3 Mass spectrometry and data analysis

Both proteomic and peptidomic MS analyses were performed on a Q Exactive™ plus Hybrid Quadrupole-Orbitrap Mass Spectrometer, following our established protocols for quantitative shotgun proteomic analysis (162). In brief, the fractionated samples were first desalted using OMIX-Mini Bed 96 C18 pipette tips (Agilent Technologies). Peptides were washed with buffer

A (0.1 % FA in H2O), and eluted in 5µL of buffer B (64.5 % ACN, 35.4 % water, 0.1 % formic acid). Sixty µL of buffer A were added to each sample and 18 µL were loaded onto a 2cm trap C18 column using a Proxeon EASY-nLC nano-flow pump. Peptides were eluted from the trap column onto a 15cm C18 column. The liquid chromatography setup was connected to the Q Exactive Mass Spectrometer. RAW MS files were generated with the use of the XCalibur software (Thermo Fisher). Files were subsequently analyzed (to DAT files) using the Mascot Daemon software (Matrix Science, London, UK, version 2.2.07), as previously described (256). Protein searches were performed against the International Protein Index (IPI) human database (version 3.71), using the following parameters: tryptic search for sweat proteome, non-enzyme search for sweat peptidome, 7ppm precursor ion mass tolerance, 0.4 Da fragment ion mass tolerance, allowance of two missed cleavages and fixed modification of carbamidomethylation of cysteines. Variable modification included oxidation of methionines. DAT files were further combined using the Scaffold software (Proteome Software Inc., V. 4.3.2), and searched with a Scaffold built-in X!Tandem search engine (Global Proteome Machine Manager, version

2010.12.01.1). For the proteomics experiment, the false discovery rate (FDR) was set at 0.4 % at the protein level and at 0.1% at the peptide level. Similarly, for the peptidomics analysis the FDR was set at 0.9 % at the protein level and at 0.02 % at the peptide level.

4.2.4 Bioinformatic analysis

Gene ontology analysis was performed using the publically available PANTHER database (http://www.pantherdb.org/). The predicted antimicrobial peptides were analyzed with the CAMP (collection of anti-microbial peptides) database (http://www.camp.bicnirrh.res.in/)(257). Tissue-specific protein lists were downloaded from human protein atlas (http://www.proteinatlas.org/) and were matched with the proteins identified in this study. By definition, tissue-enriched genes were classified as genes whose mRNA levels were more than 5 times higher in a particular tissue compared to all other tissues. Grouping and analysis of peptide signals were performed using the PepEx program (https://github.com/eparker05/PeptideExtractor).

4.2.5 SDS-PAGE and Western blotting-SDS-PAGE

The Western Blotting was performed as previously described in Chapter 2. Briefly, 12 µg of sweat samples were loaded into the SDS-PAGE (except 3µg of sweat was loaded for KLK11),. All Primary anti-KLK antibodies were produced and purified from immunized rabbit, except KLK13 (monoclonal antibody). Primary antibodies were diluted at 1:1000 TBST buffer containing 1% milk.

4.2.6 Immunodetection of KLKs in sweat (ELISA)

All ELISAs performed in this study were followed followed the protocols previously described (33). Briefly, the 96 well plates were coated with monoclonal antibodies (500 ng/well in 100 µL coating buffer) overnight. On the next day, the plates were washed and incubated with recombinant protein (as standard protein and positive control) and sweat samples for 2 hours. In

92 this part of the study, we collected sweat samples from 5 individuals (different from Chapter 4.2.1), and mixed them to make a pool of sample. After 6 times wash, biotinylated monoclonal antibodies were added into the plates and incubated for 1 hour. Finally, the substrate (alkaline phosphatase-conjugated streptavidin) was added into each well. The reaction was stopped and measured with a plate reader.

4.2.7 In vitro recombinant protein digestion assay

The assay was performed similar to Chapter 3. Briefly, recombinant protein CTGF and active KLK7 were mixed at ratio of 1:100. The mixtures were incubated at 37 °C and collected at different time points (0, 30min, 1h, 2h, and 3h). The reaction was stopped by boiling with 4x SDS-PAGE sample buffer (Bio-Rad) and analyzed by 4–12% gradient SDS-polyacrylamide gel electrophoresis and silver-stained using the PlusOne Silver Staining kit (GE Healthcare).

4.3 Results

4.3.1 Characterization of sweat proteome

To enrich for protein abundance and to account for possible sex-specific differences, the sweat sample pools from male and female individuals (5 per group) were analyzed separately. The two pools displayed similar total protein concentrations. Starting with the same amount of total protein (300 μg), 861 unique proteins were collectively identified from the two pools. As shown in Fig. 4.2A, 562 proteins were common between the male and female samples, while another 198 and 101 proteins were identified as unique proteins in the male and female pools, respectively. To illustrate the depth of our analysis, we compared our data with 5 sweat proteomic reports (247,252),(246). As shown in Figure 4.2B-F, our analysis not only identified the majority of previously known sweat proteins, but also identified a plethora of new sweat- associated proteins.

In order to classify all identified sweat proteins according to their predicted biological activity, we performed gene ontology analysis using the publicly available classification system PANTHER (Protein Analysis Through Evolutionary Relationships). Of note, 858 (out of the 861) proteins fit the criteria for classification using the PANTHER system. Similar patterns of gene ontology analysis, concerning molecular function, cellular components and biological activity were observed between male and female sweat (Fig. 4.3). About 40% of the sweat proteins were classified as „involved in catalytic activity‟, which is consistent with previous reports of intense catalytic activity in human epidermis (258).

Our study validated the presence of almost all previously reported sweat proteases, including the major skin cysteine (e.g. caspase-14, cathepsin L), aspartate (e.g. cathepsin D), metallo- (matrix metalloprotease 9 (MMP9)) and serine (KLK 5 and 7) proteases (Table 4.1). In addition, we revealed a plethora of previously unknown sweat-associated proteases, such as the cathepsins B, D, Z, F, S, L2, MMP8 and several members of the major skin desquamatory family of kallikrein- related peptidases (such as KLK1, KLK6-11, KLK13)(96) (Table 4.1 and Table A1). The importance of protease activity in human skin is reflected also by the presence of several endogenous protease inhibitors. Our proteomic approach identified several known skin protease inhibitors in human sweat(10) (Table 4.1 and Table A2).

Characterization of tissue-specific protein expression patterns in ex-vivo models can potentially improve our understanding of the (patho)physiological mechanisms and has been proposed as a promising strategy for future biomarker discovery efforts (259). In pursuit of skin-specific proteins, we mapped our identified sweat proteins to the skin-specific proteome (as defined by the Human Protein Atlas, http://www.proteinatlas.org/). This comparison revealed 28 sweat proteins that are considered as „skin-enriched‟ (Figure 4.4). Interestingly, other than skin, several esophagus-associated proteins were also identified in our lists. This finding further supports previous reports of general homology between the proteomes and gene expression profiles between skin and esophagus (http://www.proteinatlas.org/).

Figure 4.2. (A)Venn diagrams of total proteins identified from female and male sweat pools. (B-

F) Overlap of our sweat proteome with previously published sweat proteomic datasets.

Figure 4.3. Gene ontology classification of proteins identified by sweat proteomics. Left: data from male sample; Right: data from female pool. (The definitions of the above classes can be found in the PANTHER database. For example, cell part (GO:0044464): any constituent part of a cell, the basic structural and functional unit of all organisms

(http://www.pantherdb.org/panther/category.do?categoryAcc=GO:0044464))

Figure 4.4. Tissue specificity of proteins identified in sweat proteomics study.

Table 4.1 Selected list of proteases and protease inhibitors identified in sweat proteome

Gene Proteases Class Name Protein Name MEROPS ID

Serine (26)

KLK7 Kallikrein-related peptidase 7 S01.300

KLK6 Kallikrein-related peptidase 6 S01.236

KLK11 Kallikrein-related peptidase 11 S01.257

LTF Lactotransferrin S60.001

DPP7 Dipeptidyl peptidase 2 S28.002

Aspartic (3)

CTSD Cathepsin D A01.009

SPPL2A Signal Peptide Peptidase-Like 2A A22.007

PGC Gastricsin A01.003

Cysteine(14)

CASP14 Caspase-14 C14.018

GGH Gamma-Glutamyl Hydrolase C26.001

CTSV Cathepsin L2 C01.009

Metalloproteinase (14)

CPE Carboxypeptidase E M14.006

CPA4 Carboxypeptidase A4 M14.017

CPM Carboxypeptidase M M14.005

Protease inhibitors (38)

SPINK5 Serine protease inhibitor Kazal-type 5 I01.013

A2ML1 Alpha-2-macroglobulin-like protein 1 I39.007

C3 Complement C3 I39.950

1. Example of different classes of proteases and protease inhibitors (selected based on total

spectral counts) are shown as in above table. The number of proteins in each class, found

in this study, is shown in brackets. For a full list, pleases see Table A1-2.

100

4.3.2 Characterization of human sweat peptidome

Peptidomic analysis of the two human sweat pools resulted in the identification of 32,818 unique but overlapping endogenous peptides, originating from a total of 1,651 proteins (Fig. 4.5 A, B). To our knowledge, this is the first reported human sweat peptidomics study. To explore the overlap of protein identification between our proteomics and peptidomics study, we compared the sweat-associated proteins found in these two experiments. A total of 584 proteins were found in both studies, representing a possible core sweat proteome. Additionally, another 1067 from peptidomics and 277 from proteomics proteins were uniquely identified in each of the experiments (Fig. 4.5 C). Taken together, a total of 1,928 sweat proteins were collectively identified. Similar to the proteomics approach, the 1,651 proteins identified from peptidomics were uploaded to the PANTHER server for gene ontology analyses. As expected, very similar clustering of proteins based on their molecular function and tissue-specificity was observed between the peptodomics and the proteomics studies (Fig. 4.6-7). Although some of the proteins identified from our peptidomics approach have been previously reported in sweat, the majority represent novel findings. Among them, of special interest is the identification of skin antimicrobial peptides, cystatins and corneodesmosomal proteins, which, as discussed below, represent major components of human skin physiology.

101

Figure 4.5. Venn diagram of peptides (A) and proteins (B) identified from female and male sweat pools. (C) Venn diagram of proteins found in the sweat proteome and peptidome.

102

Figure 4.6. Gene ontology analysis for proteins identified in sweat peptidomics samples. A, molecular functions analysis; B, cellular components analysis; C biological processes analysis.

103

Figure 4.7. Tissue specification of proteins identified in sweat peptidomics study.

104

4.3.3 Sweat antimicrobial peptides

Apart from its apparent skin barrier roles, human skin represents an effective biochemical defense system against environmental pathogens (i.e. bacteria, viruses, fungi). To exert its innate immune functions, skin produces a plethora of anti-microbial peptides (AMPs) that concurrently regulate the growth of normal skin flora and fight the pathogenic accumulation of foreign microorganisms (260). These peptides (e.g. cathelicidin and β-defensins) are produced either as a result of skin inflammatory signals or are constitutively produced by the skin in the form of a natural antibiotic (e.g. dermcidin) (260). Consistent with previous studies, our peptidomic analysis revealed the presence of several dermcidin peptides in human sweat. Unlike dermcidin, neither cathelicidin nor defensin peptides were identified in our sweat samples, even though they are known to be expressed in human skin. The precursor of dermcidin (DCD) contains 110 amino acids, with an N-terminal signal peptide sequence (1-19 amino acids). Fourteen different kinds of DCD-derived peptides have been previously detected in healthy human sweat, but not all have been shown to display antimicrobial activity (261,262). In our study we were able to detect 815 different but overlapping DCD peptides (both of isoform 1 and 2), eight of which have been identified in a previous study. Overlapping peptides have been identified in peptidomics studies, first originating from endoproteolysis, followed by exoprotesolysis(263). In our peptidomics data, we have also found series of overlapping peptides. For example, series of overlapping peptides of DCD are summarized in Table A3. Collectively, our DCD peptides correspond to 82.7% protein coverage of the intact DCD protein sequence, suggesting that DCD is one of the most abundant sweat proteins (Fig. 4.8). The great majority of the previously reported antimicrobial DCD peptides are mapped to the C-terminal region of the protein. Some N-terminal DCD peptides have also been recently identified, and exhibit neurogenic and neuroprotective effects (i.e. non-antimicrobial) (264). To test if these peptides are also antimicrobial, we used the PepEx software to map the distribution of the DCD peptides to their full protein sequence. PepEx is a peptidomics visualization tool that allows the representation of quantitative alignments. As shown in Fig. 4.8, the C-terminal part of DCD appeared as the major location for the mapping of the sweat DCD peptides. Future studies on the candidate antimicrobial peptides of this extended repertoire of DCD peptides in human sweat are expected to uncover novel insights into the role and organization of this protein in skin innate immunity.

105

Furthermore, we examined all endogenous sweat peptides, to predict their antimicrobial activity, by overlapping them with an AMP prediction database (CAMP). A total 32818 peptides were submitted to the database. A list of 1,541 predicted antimicrobial peptides was generated with an SVM (Support vector machine) algorithm score of 0.8 or higher. Additionally, we analyzed the protease cleavage sites of those predicted AMPs. Serine proteases seem to contribute almost half of the predicted AMPs, as 324 tryptic peptides and 403 chymotryptic peptides were found in the predicted AMPs.

106

Figure 4.8. Protein sequence coverage map and endogenous proteolytic map of dermcidin (DCD). Top part: The map was generated by Scaffold. Amino acid sequences with bold font depicts the peptides whose sequence was identified by MS. Amino acid sequence with * on the top illustrates amino acid modifications identified by MS. Bottom part: the graph was generated by PepEx. The height of each column represents the total ion current (TIC) of the corresponding peptides. Underline (top figure) highlights the peptide sequences with relatively higher TIC (AA 63-109).

107

4.3.4 Detection of KLKs in human sweat

As mentioned above, we were able to detect 9 of the 15 human KLKs in this proteomics study (Table 4.2). To confirm the existence of these KLKs in sweat, we performed 6 KLKs (KLK5, 6, 7, 8, 11 and 13) specific ELISA (Table 4.2). Based on ELISA results, KLK7 seems to be the most abundant KLKs in sweat while KLK8 and 11 were less abundant, and KLK5, 6 and 13 were much less detectable. We also performed western blotting of sweat samples using KLK- specific antibodies, in order to detect different isoforms of KLKs. We were able to detect full- length form of KLK 7, 8, 11 and 13 in sweat (Fig. 4.9), which indicated that sweat may be a favorite sample to study the in vivo functional roles of skin KLKs.

108

Fig. 4.9 Western blotting analysis of KLKs in human sweat. S refers concentrated sweat sample (1 pooled sample from 5 different individuals). rK7=recombinant KLK7 proteins, rK8=recombinant KLK8 proteins, rK11=recombinant KLK11 proteins, rK13=recombinant KLK13 proteins. All recombinant proteins are active form proteins produced and purified from yeast expression system (SDS-PAGE result see Figure 7). 12 µg of sweat samples were loaded into the lanes (except 3µg of sweat was loaded for KLK11 WB), while 500 ng of recombinant KLK proteins were loaded as positive control. All Primary anti-KLK antibodies were produced and purified from immunized rabbit, except KLK13 (monoclonal antibody).

109

4.3.5 Detection of putative KLK7 substrate peptides

As mentioned in Chapter 3, a list of putative KLKs substrates and their responding peptides were reported (162). We matched those peptide sequences with our sweat peptidomics data and interestingly, we were able to find the peptide “Y-RLEDTFGPDPTMIR-A” (originated from connective tissue growth factor (CTGF)), Y-KTDLEKDIISDTSGDFRK-L (Annexin A2), Y- TNFDAERDALNIETAIK-T (Annexin A2) and F-LSGGQSEEEASINLNAINK-C (Fructose- bisphosphate aldolase A) as endogenous sweat peptides. The peptides were summarized in Table 4.3. Thus, these specific peptides might be generated by in vivo KLK7 proteolysis. To further validate CTGF as a potential substrate of KLK7, the recombinant CTGF protein was incubated with KLK7 in a time-course digestion assay, and we found that CTGF was cleaved by KLK7 after 30min incubation (Fig. 4.10). We have summarized a table of putative KLKs substrate proteins detected in human sweat (Table 4.4).

110

Table 4.2 KLKs identified in this proteomics study.

Unique Protein name Biological sample Unique SP Total SP Sequence coverage Concentration PC

Male 2 2 2 0.169 KLK1 - Female 2 2 4 0.138

Male 3 4 11 0.0683 KLK5 1.6 mg/g Female 3 4 12 0.0683

Male 9 12 25 0.389 KLK6 0.33mg/g Female 5 7 17 0.234

Male 21 45 84 0.482 KLK7 6.33mg/g Female 15 29 53 0.482

Male 5 6 12 0.19 KLK8 2.13mg/g Female 7 10 17 0.256

KLK9 Female 1 2 2 0.048 -

Male 1 1 2 0.0399 KLK10 - Female 2 2 3 0.0652

KLK11 Male 10 12 24 0.38 2.16mg/g

Male 5 6 7 0.173 KLK13 0.8mg/g Female 5 5 8 0.173

PC means peptide count, and SP means spectrum count. Concentrations of KLKs were measured by ELISA.1 pooled sample (5 mixed sweat samples) by specific KLKs ELISA. Concentration unit is presented as KLKs/ total protein of sweat samples. – means ELISA system was not available in this study.

111

Table 4.3 Previous known putative KLKs substrates were identified in human sweat.

Gene Name Protein name Peptides

CTGF Connective tissue growth factor Y-RLEDTFGPDPTMIR-A

ANXA2 Annexin A2 Y-KTDLEKDIISDTSGDFRK-L,

Y-TNFDAERDALNIETAIK-T

ALDOA Fructose-bisphosphate aldolase F-LSGGQSEEEASINLNAINK-C A

112

Table 4.4. KLK7 substrates peptides found in substrate screening experiment (Chapter 3) were detected in sweat samples.

KLKs Substrates

KLK1 -*

KLK5 DSG1, DSC1, CDSN, ProKLK7, cathelicidin, fibronectin,

laminin, collagen, semenogelins, profilaggrin, IGFBP2, 7

KLK6 fibronectin, collagen type 1, laminin, DSC1, E-cadherin

KLK7 DSC1, CDSN, cathelicidin, fibronectin, E-cadherin, CTGF, TNC**, ANXA2

KLK8 Fibronectin

KLK9 -

KLK10 -

KLK11 -

KLK12 CTGF

KLK13 Fibronectin, collagen type I, Laminin, DSG1

- No substrate report

-* known substrates not detected

** Protein detected in one individual sweat sample

113

Figure 4.10. In vitro proteolytic processing of CTGF by KLK7. Briefly, 100ng of recombinant CTGF were incubated with active KLK7 (3.3nM), and the reaction was stopped at different time points (1, 30 min and 3 hours), respectively. The arrows indicate the degradation of CTGF.

114

4.4 Discussion

Human sweat is a skin-associated biological fluid, secreted by eccrine and apocrine glands. From an evolutionary standpoint, the raison d'etre of eccrine sweat is to allow the human body to adapt to conditions associated with increased heat (e.g. hot weather, vigorous exercise, emotional stress e.t.c). The initial notion that sweat is merely a water and salt solution was challenged by early biochemical analyses of human sweat, which revealed the presence of several sweat- associated proteins and peptides. This realization has triggered numerous efforts towards the elucidation of the composition of human sweat. Recent advances in proteomic technologies brought to light an ever increasing list of sweat associated proteins, many of which are seen as candidate protein biomarkers for a plethora of skin pathologies. It is noteworthy, that it is really difficult to know whether the proteins identified in human sweat samples truly originate in human sweat, or if they simply end up in sweat from proximal tissues (e.g. skin). While this knowledge may be important from a functional perspective, we believe that it is of limited importance when it comes to possible translational applications of human sweat analysis. Any molecule that can be assayed and quantified in human sweat (original or not) may carry important clinical information.

From a functional perspective, the analysis of sweat proteome provided some new insights into the putative roles of this biological fluid. For example, it is now evident that sweat is a fluid with intense proteolytic activity. All previous sweat proteomic studies agree on the presence of several skin proteases and skin protease inhibitors. This observation is in accord with the intense proteolytic activity seen in human skin. Several classes of skin enzymes are found in human sweat, including serine proteases (i.e. KLKs), cysteine proteases (i.e. caspase-14), aspartate proteases (i.e. cathepsin D), and metallo-proteases (i.e. MMP9) (67,246,247,265,266). Given that these proteases collectively regulate major skin pathophysiological functions, such as barrier formation, desquamation and innate immunity, direct analysis of the activity of these enzymes in human sweat might lead to the identification of novel biomarkers and/or therapeutic targets for several skin conditions (96).

Among the sweat-identified proteases, considerable attention has been paid to KLKs, since they represent the most prominent skin-associated serine proteases. KLKs account for most of the

115

serine protease activity seen in human skin (96). Several putative roles have been proposed for these enzymes in skin pathophysiology, among which skin desquamation remains the most established (101). It is now clear that a group of skin KLKs are co-expressed in human skin, where they participate in enzymatic cascades, through which each KLK can activate (via a feedback mechanism) more members of the KLK family in downstream (in a way similar to the blood coagulation cascade) (267). The established roles of these skin enzymes, include skin desquamation, regulation of innate immunity and maintenance of skin barrier function (96). The notion that skin harbors intense serine protease activity is not new. Early biochemical studies have led to the cloning and characterization of KLK5 and KLK7 as the major serine proteases in human skin. More recently, our group used a KLK8-specific immunoassay (ELISA) and found that KLK8 is another major KLK of human sweat and showed that KLK8 is catalytically active, implicating new roles for this protease in the maintenance of normal skin homeostasis (67). Our current study broadens our view of sweat-associated skin KLKs, by identifying several new KLKs, such as KLK1, KLK9, KLK10, KLK11 and KLK13 in human sweat (Table 4.3). Of note, our in-house specific KLK ELISAs further validated our MS-based identification of multiple KLKs in skin. Moreover, this result was supported by our lab‟s study using SRM assays in sweat (268). Furthermore, with western blot, full length and truncated form of KLKs were found in sweat, indicating that active KLKs are possibly present in sweat. Previous studies have indicated that trypsin-like and chymotrypsin-like activities were detected in SC and sweat. Using immune- based activity assay, it reveals that several (i.e. KLK8) are catalytically active in human sweat ((67). Since the specific KLK7 antibodies did not perform well with the activity assay, further studies are needed to explore the activity of KLK7 in sweat. Of note, the protein level of KLK9 was the lowest among 9 sweat KLKs (based on sweat proteomics data). However, truncated KLK9 was detected with the most abundance among 9 sweat KLKs according to sweat peptidomics results. One possibility would be degradation of KLK9 is essential to maintain the skin healthy. Thus, future elucidation of the unique and overlapping activities of these enzymes is expected to lay the foundation for a deeper appreciation of the pathophysiological roles of KLKs in human skin.

Desquamation refers to the shedding of the outermost layer of human skin epidermis during the constant process of skin renewal. Normally, skin desquamation occurs when keratinocytes are

116

individually shed unnoticeably, following their 14 days differentiation journey from the deeper layers of the epidermis to the outermost layers of the stratum corneum (SC) (121). The latter consists of a 10-micron thick layer, which accounts for about 10% of the human epidermis and is comprised of layers of fully-differentiated keratinocytes (called corneocytes) that are embedded in a glue-like matrix. During desquamation, skin proteases (mainly skin KLKs) cleave the cell- junction structures, called corneodesmosomes (comprising the proteins desmogleins desmocollins and corneodesmosins), that hold the corneocytes together and enable the shedding of the dead outer cells (96). Imbalanced corneodesmosome degradation due to abnormal skin protease activity can lead to severe skin pathophysiologies, as best exemplified by the Netherton Syndrome, in which a genetic defect of an endogenous KLK inhibitor leads to unopposed skin desquamation (96). On the contrary, reduced desquamation rate can also lead to severe skin diseases, such as ichthyosis vulgaris and soap-induced xerosis (269). Previous studies reported the identification of a few corneodesmosomal proteins (i.e. DSG2) proximal to sweat glands (deeper in epidermis) (270). Here, for the first time, we identified a plethora of corneodesmosome-related peptides in human sweat (Table A4). The frequent degradation of these proteins from the top layer of the skin and their subsequent release into sweat allows a relatively straightforward identification of these peptides in human sweat by skin peptidomics. We envisage that similar peptidomic analyses of human sweat may serve in the near future as convenient tools for the real-time monitoring of skin barrier function, which might enable personalized interventions for both therapeutic and cosmetic applications.

Other than proteases, sweat contains large amounts of endogenous protease inhibitors. For instance, Serine protease inhibitor Kazal-type 5 (SPINK5), a gene that encodes the serine protease inhibitor Lymphoepithelial Kazal-type-related inhibitor (LEKTI), was detected as the most abundant protease inhibitor in our study (based on total MS spectral counts). LEKTI is known to be highly expressed in the spinous and granular layer of the epidermis (72), while SPINK5 mRNA has also been detected in sweat glands (97). The rare genetic skin disease known as Netherton syndrome is caused by germline SPINK5 mutations (271-273). Interestingly, besides SPINK5, SPINK7 and SPINK9 were also detected in our study. SPINK9 has been previously reported to be expressed in the palmo-plantar epidermis, lichen simplex chronicus and the lesions of actinic keratosis, as well as in squamous cell carcinoma (274,275). Almost no

117

information was available regarding the expression of SPINK7 and SPINK 9 in human sweat. Here, for first time we report their proteomic detection in human sweat, a finding that raises more questions regarding their potential functional relevance in skin-associated processes. Other than SPINKs, cystatins are well-established endogenous inhibitors of a major group of skin proteases, called cysteine proteases (i.e. cathepsins) (276). In total, there are 3 major types of cystatins (comprising a 12-member super-family) and almost all of them have been previously detected in human skin. Cystatins are known for their vital roles in the maintenance of normal skin barrier function, mainly via direct inhibition of specific target enzymes (i.e. cysteine proteases from pathogens). Dysregulated activities of type I and II cystatins have been directly associated with the development of skin inflammatory diseases, such as atopic dermatitis and psoriasis (276). Furthermore, new antimicrobial roles for the cystatins have also been recently suggested. For instance, N-terminal peptides of cystatin C have been shown to exhibit anti- leishmanial activity, while cystatin S peptides have been reported to exert potent antibacterial activity (277-279). Previously, the full-length form of cystatin A has been detected in sweat samples using in-gel tryptic digestion followed by mass spectrometry analysis (266). Here, for the first time, we report the expression of additional 7 different cystatins (i.e. cystatin A, B, C, M, S, SN, SA) in human sweat, identified independently by both our proteomics and peptidomics approach (Table A5). Cystatin D was also found in our proteomics dataset, but not in the peptidomics one. This could either be due to an overall increased resistance of cystatin D to endogenous proteolytic degradation, or decreased ability of cystatin D peptides to be identified by MS.

Apart from desquamation, skin represents the first-line barrier of innate immunity against infection. Several antimicrobial peptides have been previously identified in human skin, with proposed roles in skin immunity. Cathelicidins and β-defensins are the two major types of antimicrobial peptides expressed by human skin keratinocytes, in response to inflammation or injury (280). We were able to detect the low-abundance cathelicidin in our sweat proteomics sample but not in our peptidomics sample. To our knowledge this is the first time that cathelicidin peptides get identified in sweat using shotgun proteomics. It is likely that the overall low abundance of these peptides in this biofluid prevented its identification by previous sweat analyses. Cathelicidin has been shown to exert potent antibacterial activities against a wide range

118

of environmental pathogens and efforts have been reported for the development of cathelicidin- based natural antibiotics. No β-defensin isoforms were identified in our sweat analysis. This could be attributed to sensitivity issues or to a possible reduced ability of certain defensin peptides to be ionized. Elafin (also known as skin-derived antileukoproteinase) and dermcidin, both molecules with established antimicrobial functions in human skin, were also abundantly identified in our sweat proteome (281). In addition, we were able to predict 1,541 potential AMPs through matching with the CAMP database, which include some known AMP proteins, such as dermcidin and cystatin-S. It is well-known that the activity of AMPs can be regulated by various proteases, i.e. protease proteolysis of AMP precursors into active forms, or degradation of active forms into small non-active peptides (102). Accordingly, we analyzed those AMPs based on their cleavage sites and found that serine proteases seem to contribute half of predicted AMPs. This finding suggests that serine proteases may play an essential role in sweat AMP generation, which is consistent with previous human sweat studies that serine proteases could control the antimicrobial function in skin (102).

Direct analysis of the sweat peptidome can be used for validation of predicted endogenous substrates of skin proteases. For instance, from our previous study, a list of putative KLK7 substrates was elucidated from a cell culture-based degradome (162). Among the substrate candidates, we were able to identify several peptides, which were predicted to be generated by KLK7 proteolysis. For example, we found that KLK7 may cleave connective tissue growth factor (CTGF) and generate the peptide “Y-RLEDTFGPDPTMIR-A” (162). We were able to find the same peptide in our sweat sample. Similar observations were made for AnnexinA2 (Y- KTDLEKDIISDTSGDFRK-L, Y-TNFDAERDALNIETAIK-T) and Fructose-bisphosphate aldolase A (F-LSGGQSEEEASINLNAINK-C). Thus, peptidomics data can add more validation to previous degradomics analysis.

Taken together, our study represents the largest proteomic and peptidomic analysis of human eccrine sweat. Our analysis not only validated the presence of almost all previously identified sweat proteins, but also significantly expanded our view of its protein composition. For the first time, we provide an in-depth characterization of endogenous peptides seen in human sweat. Collectively, these findings set the basis for future mechanistic studies towards a better

119

understanding of the complex web of proteases that dictate the (patho)physiology of human skin and may act as a platform for the future discovery of novel skin biomarkers.

4.5 Author Contributions

YY designed, performed and analyzed the experiments and wrote the paper. IP designed and analyzed the experiments and wrote the paper. CM helped sweat proteomic samples preparation and data analysis, draft and revise manuscript revision. EPD contributed to the conception and design of the study, and helped draft and revise the manuscript.

120

Chapter 5

Summary and Future Directions

121

5 Summary and Future Directions 5.1 Summary

A combined degradomics with sequence-based substrate specificity analysis approach was employed in this thesis to identify novel physiological substrates for KLK7, and various in vitro assays were subsequently used to validate some of the putative substrates. Furthermore, endogenous peptides identified from human sweat served as additional information to the KLKs substrate finding. Midkine, CYR61, CTGF and tenascin were identified as novel putative substrates of KLK7. This innovative substrate identification strategy has the potential of being applied for the identification of putative substrates of other proteases.

Key findings:

1. Recombinant active-form KLK7 protein:

a) Glycosylated form KLK7 and self-degradation fragment of KLK7 were found in the purified recombinant KLK7 protein

b) Active KLK7 was found to have preference of P1 at Try than Phe

c) In vitro optimal pH of KLK7 activity was basic (pH 8.5), although KLK7 was still active under acidic conditions

2. Putative substrates of KLK7:

a) A list of novel putative KLK7 substrate proteins were identified using degradomics combined with sequence-based substrate specificity analysis approach

b) Midkine was a preferred substrate for KLK7, compared to other novel substrates

c) Midkine was a KLK7-specific substrate among the tested skin KLKs in vitro

d) KLK7 may regulate the biological function of midkine in the diseased skin conditions (e.g. skin melanoma) by proteolysis

122

e) CTGF, CYR61 and tenascin were found to be cleaved by KLK7 as well in vitro

f) CTGF specific peptide was also detected in human sweat, suggesting that KLK7 may cleave endogenous CTGF in vivo.

3. Human sweat proteome and peptidome:

a) 861 unique proteins were identified during our proteomic analysis and an additional 32,818 endogenous peptides (corresponding to 1,067 proteins) from our peptidomics workflow.

b) Gene ontology analysis revealed that skin proteases and endogenous protease inhibitors were abundantly expressed in human sweat, highlighting the intense proteolytic activity of the human skin.

c) A list of 1,541 predicted antimicrobial peptides was generated with the algorithm of CAMP. The presence of these predicted together with previously reported antimicrobial peptides in sweat supports the notion that this fluid is involved in the host defense and innate immunity of skin.

d) 9 KLKs (i.e. KLK1 and 5-13) were detected in human sweat using shot-gut proteomics and the majority of them in sweat were detectable by ELISA and WB.

123

5.2 Future direction

The experimental data presented in this thesis has expanded the understanding of KLK7 and its biological substrates in skin (patho-)physiological processes. A combined degradomics with sequence-based substrate specificity analysis approach was utilized and a large number of known and novel substrates were identified. Several novel putative substrates (i.e. MDK, CTGF, TNC and CYR61) were validated via various in vitro biochemical assays. The endogenous KLK7 substrates (e.g. CTGF) revealed by human sweat proteomics and peptidomics further corroborated this innovative substrate finding approach. Similar degradomic approaches are being used for the substrate profiling of other proteases (i.e. KLK9) in our lab (Panagiota S. Filippou, Sofia Farkona, Davor Brinc, Yijing Yu, Ioannis Prassas and Eleftherios P. Diamandis. Manuscript in submission). Comparing to these traditional scanning techniques (i.e. combinatorial peptide libraries scanning), our approach has advantages such as it is roust and can represent the protein-protein interaction within living cells. Additionally, our strategy covered the limitation of previous cell surface-based degradomics approach by combination of substrate specificity analysis (161,162). Furthermore, compared to other MS-based substrate finding approaches (i.e. TALIS), our techniques is less time consuming and costs less. Even through, there are some limitations of our approach. For instance, unlike TAILS, the degradomic approach used in this thesis was label-free quantification, thus there is possible of a statistical limitation for low abundant proteins. Furthermore, any variation in sample preparation, LC-MS reproducibility, ionization efficiency can lead to increased error. To resolve this problem, biological and technological replicates can reduce data variation to some extent, however, a standard control or isotopic labeling used in the samples will be a great key, which should be considered in further experiment design.

We have to acknowledge that our cell culture model for KLK7 substrate finding cannot fully mimic the in vivo dynamic environment. For instance, four epidermal layers contain distinct components (i.e. keratinocytes, melanocytes, Langerhans cells, fibrocytes) and microenvironment (i.e. pH), which functionally cooperate with each other (121). Since KLK7 is abundantly expressed in skin, proteomic analysis of human sweat may reveal endogenous substrates for KLK7 and other KLKs. Early proteomic studies have shown that the protein composition of human sweat is highly dynamic and can alter significantly in various skin and

124

other disorders, such as ectodermal dysplasia, atopic dermatitis, and even schizophrenia (246- 248). Notably, all previous proteomic analyses of human sweat have revealed less than 600 proteins (246,247,252-254). Here, we performed deep proteomic and peptidomic analysis of sweat from healthy humans, and almost 2,000 proteins and 32,818 unique but overlapping endogenous peptides were identified, providing the most comprehensive proteomic description of human sweat to-date. Since protein composition of human sweat can change dramatically in various skin diseases, it will be interesting to compare the sweat proteomics and peptidomics between healthy people and patients with skin disorders, which may provide potential skin disease biomarker and therapeutic targets.

KLK7 participates in various skin patho(biological) processes through interaction with its putative substrates. Through N-terminal sequencing and mass spectrometry analysis, we predicted that KLK7 cleavage might disrupt the binding ability of midkine to midkine receptors, which is needed to mediate its biological functions. Indeed, we found that the cleavage of midkine by KLK7 reduced the cell proliferation and migration in vitro. However, additional experiments (Co-Immunoprecipitation) should be performed to explore the potential interaction between truncated midkine and its receptor. Moreover, although our preliminary data suggest the cleavage of midkine by KLK7 may affect its apoptotic function, more experiments are needed to further confirm it. Furthermore, the biological function studies with midkine were performed in melanoma cell line in vitro, further studies to test whether KLK7-mediated midkine cleavage affects the function of midkine in more physiological conditions (e.g. in vivo animal models) are warranted. Since both midkine and KLK7 were detected in various cells and tissues, it will be also interesting to test this interaction in other cell systems. Lastly, it has been suggested that midkine is a putative substrate for KLK9 in tumorigenic tongue squamous cell culture based on degradomics (Panagiota S. Filippou, Sofia Farkona, Davor Brinc, Yijing Yu, Ioannis Prassas and Eleftherios P. Diamandis. Manuscript in submission), and thus further studies are needed to investigate whether various enzymes are utilized to proteolyze midkine under different conditions.

Two putative KLK7 substrates: CYR61 and CTGF, belong to the CCN family, and both were previously reported as putative substrates for KLK12 and KLK14 (161). In our study, KLK7 could proteolyze CYR61 and CTGF, and CTGF chymotryptic-peptide “Y- 125

RLEDTFGPDPTMIR-A” was detected in human sweat. However, further studies are needed to determine whether CTGF and CYR61 harbor similar functional roles in skin diseases setting. Furthermore, CYR61 and CTGF were reported to contribute to lung cancer progression through interaction with various cytokines (i.e. VEGF). Thus it will be interesting to test whether KLK7 is involved in the regulation of CTGF and CYR61 in other pathophysiological processes (e.g. lung cancer).

Antimicrobial peptides are part of the first defense line of human immunity. The activities of antimicrobial peptides (i.e. cathelicidin) were reported to be regulated by skin enzymes (i.e. KLK5 and KLK7). In this study, we were able to detect some known sweat antimicrobial peptides and some novel “predicted” antimicrobial peptides. By sorting the cleavage sites of these predicted peptides, we discovered a list of putative (chymo-)tryptic antimicrobial peptides in human sweat. However, additional experiments are needed to investigate the functions of these predicted antimicrobial peptides in biological settings.

KLKs including KLK7 are highly expressed in various tissues. Besides skin, gut tissues (small intestine and colon) also express the high level of KLKs. Aberrant proteases activities were found in gastrointestinal tract diseases, and thus it raises the possibility to target these proteases with specific inhibitors (282). It will be interesting to investigate whether KLKs are involved in the gastrointestinal tract diseases.

Gene deletion is absolutely a key way to investigate the function of genes of interests, and knockout mice are widely used genetic models. In KLKs family, knockout mice were generated for KLK1, 4 and 8 (25,26,283). In addition, knockdown of KLK5 and KLK7 by siRNA was tested as an in vitro therapeutic approach in an eczema model (233); however, there are several limitations of siRNA including variant efficiency, incompleteness of knockdown and off-target effects (284). Clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 is a newly emerged powerful gene-editing methods that currently impact extremely in scientific research (285), and it has lots of advantages such as low cost, reduced off-target effects, less time-consuming and relatively easy handling, compared to other gene-editing tools (e.g. transcription activator-like effector nuclease or TALEN) (285). To date, there is no report of KLKs knockout generated by CRISPR in either cell lines or animal models. We believe that

126

CRISPR/Cas9 will greatly expand our current knowledge about KLK biology in the next few years.

127

References:

1. Puente, X. S., Sanchez, L. M., Gutierrez-Fernandez, A., Velasco, G., and Lopez-Otin, C. (2005) A genomic view of the complexity of mammalian proteolytic systems. Biochem Soc Trans 33, 331-334

2. Rawlings, N. D., Waller, M., Barrett, A. J., and Bateman, A. (2014) MEROPS: the database of proteolytic enzymes, their substrates and inhibitors. Nucleic Acids Res 42, D503-509

3. Kraut, J. (1977) Serine proteases: structure and mechanism of catalysis. Annu Rev Biochem 46, 331-358

4. Hedstrom, L. (2002) Serine protease mechanism and specificity. Chem Rev 102, 4501- 4524

5. Laskar, A., Rodger, E. J., Chatterjee, A., and Mandal, C. (2012) Modeling and structural analysis of PA clan serine proteases. BMC Res Notes 5, 256

6. Di Cera, E. (2009) Serine proteases. IUBMB Life 61, 510-515

7. Page, M. J., and Di Cera, E. (2008) Serine peptidases: classification, structure and function. Cell Mol Life Sci 65, 1220-1236

8. Schechter, I., and Berger, A. (2012) On the size of the active site in proteases. I. Papain. 1967. Biochem Biophys Res Commun 425, 497-502

9. Yousef, G. M., and Diamandis, E. P. (2001) The new human tissue kallikrein gene family: structure, function, and association to disease. Endocr Rev 22, 184-204

10. Goettig, P., Magdolen, V., and Brandstetter, H. (2010) Natural and synthetic inhibitors of kallikrein-related peptidases (KLKs). Biochimie 92, 1546-1567

11. Krowarsch, D., Cierpicki, T., Jelen, F., and Otlewski, J. (2003) Canonical protein inhibitors of serine proteases. Cell Mol Life Sci 60, 2427-2444

12. Otlewski, J., Krowarsch, D., and Apostoluk, W. (1999) Protein inhibitors of serine proteinases. Acta Biochim Pol 46, 531-565

13. Borgono, C. A., and Diamandis, E. P. (2004) The emerging roles of human tissue kallikreins in cancer. Nat Rev Cancer 4, 876-890

14. Bernett, M. J., Blaber, S. I., Scarisbrick, I. A., Dhanarajan, P., Thompson, S. M., and Blaber, M. (2002) Crystal structure and biochemical characterization of human kallikrein 6 reveals that a trypsin-like kallikrein is expressed in the central nervous system. J Biol Chem 277, 24562-24570

128

15. Carvalho, A. L., Sanz, L., Barettino, D., Romero, A., Calvete, J. J., and Romao, M. J. (2002) Crystal structure of a prostate kallikrein isolated from stallion seminal plasma: a homologue of human PSA. J Mol Biol 322, 325-337

16. Debela, M., Magdolen, V., Grimminger, V., Sommerhoff, C., Messerschmidt, A., Huber, R., Friedrich, R., Bode, W., and Goettig, P. (2006) Crystal structures of human tissue kallikrein 4: activity modulation by a specific zinc binding site. J Mol Biol 362, 1094- 1107

17. Debela, M., Goettig, P., Magdolen, V., Huber, R., Schechter, N. M., and Bode, W. (2007) Structural basis of the zinc inhibition of human tissue kallikrein 5. J Mol Biol 373, 1017- 1031

18. Katz, B. A., Liu, B., Barnes, M., and Springman, E. B. (1998) Crystal structure of recombinant human tissue kallikrein at 2.0 A resolution. Protein Sci 7, 875-885

19. Marceau, F., and Regoli, D. (2004) Bradykinin receptor ligands: therapeutic perspectives. Nat Rev Drug Discov 3, 845-852

20. Bhoola, K. D., Misso, N. L., Naran, A., and Thompson, P. J. (2007) Current status of tissue kallikrein inhibitors: importance in cancer. Curr Opin Investig Drugs 8, 462-468

21. Ghosh, M. C., Grass, L., Soosaipillai, A., Sotiropoulou, G., and Diamandis, E. P. (2004) Human kallikrein 6 degrades extracellular matrix proteins and may enhance the metastatic potential of tumour cells. Tumour Biol 25, 193-199

22. Oikonomopoulou, K., Hansen, K. K., Saifeddine, M., Tea, I., Blaber, M., Blaber, S. I., Scarisbrick, I., Andrade-Gordon, P., Cottrell, G. S., Bunnett, N. W., Diamandis, E. P., and Hollenberg, M. D. (2006) Proteinase-activated receptors, targets for kallikrein signaling. J Biol Chem 281, 32095-32112

23. Yamakoshi, Y., Simmer, J. P., Bartlett, J. D., Karakida, T., and Oida, S. (2013) MMP20 and KLK4 activation and inactivation interactions in vitro. Arch Oral Biol 58, 1569-1577

24. Hansson, L., Backman, A., Ny, A., Edlund, M., Ekholm, E., Ekstrand Hammarstrom, B., Tornell, J., Wallbrandt, P., Wennbo, H., and Egelrud, T. (2002) Epidermal overexpression of stratum corneum chymotryptic enzyme in mice: a model for chronic itchy dermatitis. J Invest Dermatol 118, 444-449

25. Simmer, J. P., Hu, Y., Lertlam, R., Yamakoshi, Y., and Hu, J. C. (2009) Hypomaturation enamel defects in Klk4 knockout/LacZ knockin mice. J Biol Chem 284, 19110-19121

26. Yoshida, S. (2010) Klk8, a multifunctional protease in the brain and skin: analysis of knockout mice. Biol Chem 391, 375-380

27. Fernandez, I. S., Standker, L., Magert, H. J., Forssmann, W. G., Gimenez-Gallego, G., and Romero, A. (2008) Crystal structure of human epidermal kallikrein 7 (hK7)

129

synthesized directly in its native state in E. coli: insights into the atomic basis of its inhibition by LEKTI domain 6 (LD6). J Mol Biol 377, 1488-1497

28. Lundstrom, A., and Egelrud, T. (1991) Stratum corneum chymotryptic enzyme: a proteinase which may be generally present in the stratum corneum and with a possible involvement in desquamation. Acta Derm Venereol 71, 471-474

29. Egelrud, T., and Lundstrom, A. (1991) A chymotrypsin-like proteinase that may be involved in desquamation in plantar stratum corneum. Arch Dermatol Res 283, 108-112

30. Hansson, L., Stromqvist, M., Backman, A., Wallbrandt, P., Carlstein, A., and Egelrud, T. (1994) Cloning, expression, and characterization of stratum corneum chymotryptic enzyme. A skin-specific human serine proteinase. J Biol Chem 269, 19420-19426

31. Yousef, G. M., Scorilas, A., Magklara, A., Soosaipillai, A., and Diamandis, E. P. (2000) The KLK7 (PRSS6) gene, encoding for the stratum corneum chymotryptic enzyme is a new member of the human kallikrein gene family - genomic characterization, mapping, tissue expression and hormonal regulation. Gene 254, 119-128

32. Debela, M., Beaufort, N., Magdolen, V., Schechter, N. M., Craik, C. S., Schmitt, M., Bode, W., and Goettig, P. (2008) Structures and specificity of the human kallikrein- related peptidases KLK 4, 5, 6, and 7. Biol Chem 389, 623-632

33. Shaw, J. L., and Diamandis, E. P. (2007) Distribution of 15 human kallikreins in tissues and biological fluids. Clin Chem 53, 1423-1432

34. Thomas, A. C., Tattersall, D., Norgett, E. E., O'Toole, E. A., and Kelsell, D. P. (2009) Premature terminal differentiation and a reduction in specific proteases associated with loss of ABCA12 in Harlequin ichthyosis. Am J Pathol 174, 970-978

35. Ishida-Yamamoto, A., Deraison, C., Bonnart, C., Bitoun, E., Robinson, R., O'Brien, T. J., Wakamatsu, K., Ohtsubo, S., Takahashi, H., Hashimoto, Y., Dopping-Hepenstal, P. J., McGrath, J. A., Iizuka, H., Richard, G., and Hovnanian, A. (2005) LEKTI is localized in lamellar granules, separated from KLK5 and KLK7, and is secreted in the extracellular spaces of the superficial stratum granulosum. J Invest Dermatol 124, 360-366

36. Tanimoto, H., Underwood, L. J., Shigemasa, K., Yan Yan, M. S., Clarke, J., Parmley, T. H., and O'Brien, T. J. (1999) The stratum corneum chymotryptic enzyme that mediates shedding and desquamation of skin cells is highly overexpressed in ovarian tumor cells. Cancer 86, 2074-2082

37. Dong, Y., Kaushal, A., Brattsand, M., Nicklin, J., and Clements, J. A. (2003) Differential splicing of KLK5 and KLK7 in epithelial ovarian cancer produces novel variants with potential as cancer biomarkers. Clin Cancer Res 9, 1710-1720

38. Prezas, P., Arlt, M. J., Viktorov, P., Soosaipillai, A., Holzscheiter, L., Schmitt, M., Talieri, M., Diamandis, E. P., Kruger, A., and Magdolen, V. (2006) Overexpression of

130

the human tissue kallikrein genes KLK4, 5, 6, and 7 increases the malignant phenotype of ovarian cancer cells. Biol Chem 387, 807-811

39. Shan, S. J., Scorilas, A., Katsaros, D., Rigault de la Longrais, I., Massobrio, M., and Diamandis, E. P. (2006) Unfavorable prognostic value of human kallikrein 7 quantified by ELISA in ovarian cancer cytosols. Clin Chem 52, 1879-1886

40. Kyriakopoulou, L. G., Yousef, G. M., Scorilas, A., Katsaros, D., Massobrio, M., Fracchioli, S., and Diamandis, E. P. (2003) Prognostic value of quantitatively assessed KLK7 expression in ovarian cancer. Clin Biochem 36, 135-143

41. Santin, A. D., Cane, S., Bellone, S., Bignotti, E., Palmieri, M., De Las Casas, L. E., Roman, J. J., Anfossi, S., O'Brien, T., and Pecorelli, S. (2004) The serine protease stratum corneum chymotryptic enzyme (kallikrein 7) is highly overexpressed in squamous cervical cancer cells. Gynecol Oncol 94, 283-288

42. Tian, X., Shigemasa, K., Hirata, E., Gu, L., Uebaba, Y., Nagai, N., O'Brien, T. J., and Ohama, K. (2004) Expression of human kallikrein 7 (hK7/SCCE) and its inhibitor antileukoprotease (ALP/SLPI) in uterine endocervical glands and in cervical adenocarcinomas. Oncol Rep 12, 1001-1006

43. Li, W., Zhao, Y., Ren, L., and Wu, X. (2014) Serum human kallikrein 7 represents a new marker for cervical cancer. Med Oncol 31, 208

44. Talieri, M., Diamandis, E. P., Gourgiotis, D., Mathioudaki, K., and Scorilas, A. (2004) Expression analysis of the human kallikrein 7 (KLK7) in breast tumors: a new potential biomarker for prognosis of breast carcinoma. Thromb Haemost 91, 180-186

45. Holzscheiter, L., Biermann, J. C., Kotzsch, M., Prezas, P., Farthmann, J., Baretton, G., Luther, T., Tjan-Heijnen, V. C., Talieri, M., Schmitt, M., Sweep, F. C., Span, P. N., and Magdolen, V. (2006) Quantitative reverse transcription-PCR assay for detection of mRNA encoding full-length human tissue kallikrein 7: prognostic relevance of KLK7 mRNA expression in breast cancer. Clin Chem 52, 1070-1079

46. Li, X., Liu, J., Wang, Y., Zhang, L., Ning, L., and Feng, Y. (2009) Parallel underexpression of kallikrein 5 and kallikrein 7 mRNA in breast malignancies. Cancer Sci 100, 601-607

47. Talieri, M., Devetzi, M., Scorilas, A., Prezas, P., Ardavanis, A., Apostolaki, A., and Karameris, A. (2011) Evaluation of kallikrein-related peptidase 5 expression and its significance for breast cancer patients: association with kallikrein-related peptidase 7 expression. Anticancer Res 31, 3093-3100

48. Dorn, J., Gkazepis, A., Kotzsch, M., Kremer, M., Propping, C., Mayer, K., Mengele, K., Diamandis, E. P., Kiechle, M., Magdolen, V., and Schmitt, M. (2014) Clinical value of protein expression of kallikrein-related peptidase 7 (KLK7) in ovarian cancer. Biol Chem 395, 95-107

131

49. Dorn, J., Bronger, H., Kates, R., Slotta-Huspenina, J., Schmalfeldt, B., Kiechle, M., Diamandis, E. P., Soosaipillai, A., Schmitt, M., and Harbeck, N. (2015) OVSCORE - a validated score to identify ovarian cancer patients not suitable for primary surgery. Oncol Lett 9, 418-424

50. Talieri, M., Mathioudaki, K., Prezas, P., Alexopoulou, D. K., Diamandis, E. P., Xynopoulos, D., Ardavanis, A., Arnogiannaki, N., and Scorilas, A. (2009) Clinical significance of kallikrein-related peptidase 7 (KLK7) in colorectal cancer. Thromb Haemost 101, 741-747

51. Talieri, M., Li, L., Zheng, Y., Alexopoulou, D. K., Soosaipillai, A., Scorilas, A., Xynopoulos, D., and Diamandis, E. P. (2009) The use of kallikrein-related peptidases as adjuvant prognostic markers in colorectal cancer. Br J Cancer 100, 1659-1665

52. Devetzi, M., Trangas, T., Scorilas, A., Xynopoulos, D., and Talieri, M. (2013) Parallel overexpression and clinical significance of kallikrein-related peptidases 7 and 14 (KLK7KLK14) in colon cancer. Thromb Haemost 109, 716-725

53. Inoue, Y., Yokobori, T., Yokoe, T., Toiyama, Y., Miki, C., Mimori, K., Mori, M., and Kusunoki, M. (2010) Clinical significance of human kallikrein7 gene expression in colorectal cancer. Ann Surg Oncol 17, 3037-3042

54. Walker, F., Nicole, P., Jallane, A., Soosaipillai, A., Mosbach, V., Oikonomopoulou, K., Diamandis, E. P., Magdolen, V., and Darmoul, D. (2014) Kallikrein-related peptidase 7 (KLK7) is a proliferative factor that is aberrantly expressed in human colon cancer. Biol Chem 395, 1075-1086

55. Johnson, S. K., Ramani, V. C., Hennings, L., and Haun, R. S. (2007) Kallikrein 7 enhances pancreatic cancer cell invasion by shedding E-cadherin. Cancer 109, 1811-1820

56. Ramani, V. C., Hennings, L., and Haun, R. S. (2008) Desmoglein 2 is a substrate of kallikrein 7 in pancreatic cancer. BMC Cancer 8, 373

57. Ramani, V. C., and Haun, R. S. (2008) Expression of kallikrein 7 diminishes pancreatic cancer cell adhesion to vitronectin and enhances urokinase-type plasminogen activator receptor shedding. Pancreas 37, 399-404

58. Pettus, J. R., Johnson, J. J., Shi, Z., Davis, J. W., Koblinski, J., Ghosh, S., Liu, Y., Ravosa, M. J., Frazier, S., and Stack, M. S. (2009) Multiple kallikrein (KLK 5, 7, 8, and 10) expression in squamous cell carcinoma of the oral cavity. Histol Histopathol 24, 197- 207

59. Xuan, Q., Yang, X., Mo, L., Huang, F., Pang, Y., Qin, M., Chen, Z., He, M., Wang, Q., and Mo, Z. N. (2008) Expression of the serine protease kallikrein 7 and its inhibitor antileukoprotease is decreased in prostate cancer. Arch Pathol Lab Med 132, 1796-1801

60. Jamaspishvili, T., Scorilas, A., Kral, M., Khomeriki, I., Kurfurstova, D., Kolar, Z., and Bouchal, J. (2011) Immunohistochemical localization and analysis of kallikrein-related 132

peptidase 7 and 11 expression in paired cancer and benign foci in prostate cancer patients. Neoplasma 58, 298-303

61. Zhang, C. Y., Zhu, Y., Rui, W. B., Dai, J., and Shen, Z. J. (2015) Expression of kallikrein-related peptidase 7 is decreased in prostate cancer. Asian J Androl 17, 106-110

62. Leusink, F. K., van Diest, P. J., Frank, M. H., Broekhuizen, R., Braunius, W., van Hooff, S. R., Willems, S. M., and Koole, R. (2015) The Co-Expression of Kallikrein 5 and Kallikrein 7 Associates with Poor Survival in Non-HPV Oral Squamous-Cell Carcinoma. Pathobiology 82, 58-67

63. Shaw, J. L., and Diamandis, E. P. (2008) Regulation of human tissue kallikrein-related peptidase expression by steroid hormones in 32 cell lines. Biol Chem 389, 1409-1419

64. Morizane, S., Yamasaki, K., Kajita, A., Ikeda, K., Zhan, M., Aoyama, Y., Gallo, R. L., and Iwatsuki, K. (2012) TH2 cytokines increase kallikrein 7 expression and function in patients with atopic dermatitis. J Allergy Clin Immunol 130, 259-261 e251

65. Morizane, S., Yamasaki, K., Kabigting, F. D., and Gallo, R. L. (2010) Kallikrein expression and cathelicidin processing are independently controlled in keratinocytes by calcium, vitamin D(3), and retinoic acid. J Invest Dermatol 130, 1297-1306

66. Yoon, H., Blaber, S. I., Li, W., Scarisbrick, I. A., and Blaber, M. (2013) Activation profiles of human kallikrein-related peptidases by matrix metalloproteinases. Biol Chem 394, 137-147

67. Eissa, A., Amodeo, V., Smith, C. R., and Diamandis, E. P. (2011) Kallikrein-related peptidase-8 (KLK8) is an active serine protease in human epidermis and sweat and is involved in a skin barrier proteolytic cascade. J Biol Chem 286, 687-706

68. Brattsand, M., Stefansson, K., Lundh, C., Haasum, Y., and Egelrud, T. (2005) A proteolytic cascade of kallikreins in the stratum corneum. J Invest Dermatol 124, 198-203

69. Voegeli, R., Rawlings, A. V., Breternitz, M., Doppler, S., Schreier, T., and Fluhr, J. W. (2009) Increased stratum corneum serine protease activity in acute eczematous atopic skin. Br J Dermatol 161, 70-77

70. Voegeli, R., Doppler, S., Joller, P., Breternitz, M., Fluhr, J. W., and Rawlings, A. V. (2011) Increased mass levels of certain serine proteases in the stratum corneum in acute eczematous atopic skin. Int J Cosmet Sci 33, 560-565

71. Debela, M., Hess, P., Magdolen, V., Schechter, N. M., Steiner, T., Huber, R., Bode, W., and Goettig, P. (2007) Chymotryptic specificity determinants in the 1.0 A structure of the zinc-inhibited human tissue kallikrein 7. Proc Natl Acad Sci U S A 104, 16086-16091

72. Yang, T., Liang, D., Koch, P. J., Hohl, D., Kheradmand, F., and Overbeek, P. A. (2004) Epidermal detachment, desmosomal dissociation, and destabilization of corneodesmosin in Spink5-/- mice. Genes Dev 18, 2354-2358 133

73. Deraison, C., Bonnart, C., Lopez, F., Besson, C., Robinson, R., Jayakumar, A., Wagberg, F., Brattsand, M., Hachem, J. P., Leonardsson, G., and Hovnanian, A. (2007) LEKTI fragments specifically inhibit KLK5, KLK7, and KLK14 and control desquamation through a pH-dependent interaction. Mol Biol Cell 18, 3607-3619

74. Fischer, J., and Meyer-Hoffert, U. (2013) Regulation of kallikrein-related peptidases in the skin - from physiology to diseases to therapeutic options. Thromb Haemost 110, 442- 449

75. Silverman, G. A., Bird, P. I., Carrell, R. W., Church, F. C., Coughlin, P. B., Gettins, P. G., Irving, J. A., Lomas, D. A., Luke, C. J., Moyer, R. W., Pemberton, P. A., Remold- O'Donnell, E., Salvesen, G. S., Travis, J., and Whisstock, J. C. (2001) The serpins are an expanding superfamily of structurally similar but functionally diverse proteins. Evolution, mechanism of inhibition, novel functions, and a revised nomenclature. J Biol Chem 276, 33293-33296

76. Farshchian, M., Kivisaari, A., Ala-Aho, R., Riihila, P., Kallajoki, M., Grenman, R., Peltonen, J., Pihlajaniemi, T., Heljasvaara, R., and Kahari, V. M. (2011) Serpin peptidase inhibitor clade A member 1 (SerpinA1) is a novel biomarker for progression of cutaneous squamous cell carcinoma. Am J Pathol 179, 1110-1119

77. Hoffmann, D. C., Textoris, C., Oehme, F., Klaassen, T., Goppelt, A., Romer, A., Fugmann, B., Davidson, J. M., Werner, S., Krieg, T., and Eming, S. A. (2011) Pivotal role for alpha1-antichymotrypsin in skin repair. J Biol Chem 286, 28889-28901

78. Jendrny, C., and Beck-Sickinger, A. G. (2016) Inhibition of Kallikrein-Related Peptidases 7 and 5 by Grafting Serpin Reactive-Center Loop Sequences onto Sunflower Trypsin Inhibitor-1 (SFTI-1). Chembiochem 17, 719-726

79. de Veer, S. J., Furio, L., Swedberg, J. E., Munro, C. A., Brattsand, M., Clements, J. A., Hovnanian, A., and Harris, J. M. (2017) Selective Substrates and Inhibitors for Kallikrein-Related Peptidase 7 (KLK7) Shed Light on KLK Proteolytic Activity in the Stratum Corneum. J Invest Dermatol 137, 430-439

80. Franzke, C. W., Baici, A., Bartels, J., Christophers, E., and Wiedow, O. (1996) Antileukoprotease inhibits stratum corneum chymotryptic enzyme. Evidence for a regulative function in desquamation. J Biol Chem 271, 21886-21890

81. Tan, X., Furio, L., Reboud-Ravaux, M., Villoutreix, B. O., Hovnanian, A., and El Amri, C. (2013) 1,2,4-Triazole derivatives as transient inactivators of kallikreins involved in skin diseases. Bioorg Med Chem Lett 23, 4547-4551

82. Tan, X., Bertonati, C., Qin, L., Furio, L., El Amri, C., Hovnanian, A., Reboud-Ravaux, M., and Villoutreix, B. O. (2013) Identification by in silico and in vitro screenings of small organic molecules acting as reversible inhibitors of kallikreins. Eur J Med Chem 70, 661-668

134

83. Flohr, S., Randl, S. A., Ostermann, N., Hassiepen, U., Berst, F., Bodendorf, U., Gerhartz, B., Marzinzik, A., Ehrhardt, C., and Meingassner, J. G. (2008) Kallikrein 7 modulators. Google Patents

84. Janeway, C. A., Travers, P., Walport, M., and Shlomchik, M. J. (1997) Immunobiology: the immune system in health and disease, Current Biology

85. Leroy, M., Lafleur, M., Auger, M., Laroche, G., and Pouliot, R. (2013) Characterization of the structure of human skin substitutes by infrared microspectroscopy. Anal Bioanal Chem 405, 8709-8718

86. Fore, J. (2006) A review of skin and the effects of aging on skin structure and function. Ostomy Wound Manage 52, 24-35; quiz 36-27

87. Candi, E., Schmidt, R., and Melino, G. (2005) The cornified envelope: a model of cell death in the skin. Nat Rev Mol Cell Biol 6, 328-340

88. Milstone, L. M. (2004) Epidermal desquamation. J Dermatol Sci 36, 131-140

89. Eckhart, L., Lippens, S., Tschachler, E., and Declercq, W. (2013) Cell death by cornification. Biochim Biophys Acta 1833, 3471-3480

90. Garrod, D., and Chidgey, M. (2008) Desmosome structure, composition and function. Biochim Biophys Acta 1778, 572-587

91. Al-Amoudi, A., Castano-Diez, D., Devos, D. P., Russell, R. B., Johnson, G. T., and Frangakis, A. S. (2011) The three-dimensional molecular structure of the desmosomal plaque. Proc Natl Acad Sci U S A 108, 6480-6485

92. Delva, E., Tucker, D. K., and Kowalczyk, A. P. (2009) The desmosome. Cold Spring Harb Perspect Biol 1, a002543

93. Egelrud, T. (2000) Desquamation in the stratum corneum. Acta Derm Venereol Suppl (Stockh) 208, 44-45

94. Kowalczyk, A. P., and Green, K. J. (2013) Structure, function, and regulation of desmosomes. Prog Mol Biol Transl Sci 116, 95-118

95. Rawlings, A. V., and Voegeli, R. (2013) Stratum corneum proteases and dry skin conditions. Cell Tissue Res 351, 217-235

96. Prassas, I., Eissa, A., Poda, G., and Diamandis, E. P. (2015) Unleashing the therapeutic potential of human kallikrein-related serine proteases. Nat Rev Drug Discov 14, 183-202

97. Komatsu, N., Takata, M., Otsuki, N., Toyama, T., Ohka, R., Takehara, K., and Saijoh, K. (2003) Expression and localization of tissue kallikrein mRNAs in human epidermis and appendages. J Invest Dermatol 121, 542-549

135

98. Komatsu, N., Tsai, B., Sidiropoulos, M., Saijoh, K., Levesque, M. A., Takehara, K., and Diamandis, E. P. (2006) Quantification of eight tissue kallikreins in the stratum corneum and sweat. J Invest Dermatol 126, 925-929

99. Brattsand, M., and Egelrud, T. (1999) Purification, molecular cloning, and expression of a human stratum corneum trypsin-like serine protease with possible function in desquamation. J Biol Chem 274, 30033-30040

100. Caubet, C., Jonca, N., Brattsand, M., Guerrin, M., Bernard, D., Schmidt, R., Egelrud, T., Simon, M., and Serre, G. (2004) Degradation of corneodesmosome proteins by two serine proteases of the kallikrein family, SCTE/KLK5/hK5 and SCCE/KLK7/hK7. J Invest Dermatol 122, 1235-1244

101. Borgono, C. A., Michael, I. P., Komatsu, N., Jayakumar, A., Kapadia, R., Clayman, G. L., Sotiropoulou, G., and Diamandis, E. P. (2007) A potential role for multiple tissue kallikrein serine proteases in epidermal desquamation. J Biol Chem 282, 3640-3652

102. Yamasaki, K., Schauber, J., Coda, A., Lin, H., Dorschner, R. A., Schechter, N. M., Bonnart, C., Descargues, P., Hovnanian, A., and Gallo, R. L. (2006) Kallikrein-mediated proteolysis regulates the antimicrobial effects of cathelicidins in skin. FASEB J 20, 2068- 2080

103. Shaw, J. L., Petraki, C., Watson, C., Bocking, A., and Diamandis, E. P. (2008) Role of tissue kallikrein-related peptidases in cervical mucus remodeling and host defense. Biol Chem 389, 1513-1522

104. Yamasaki, K., Di Nardo, A., Bardan, A., Murakami, M., Ohtake, T., Coda, A., Dorschner, R. A., Bonnart, C., Descargues, P., Hovnanian, A., Morhenn, V. B., and Gallo, R. L. (2007) Increased serine protease activity and cathelicidin promotes skin inflammation in rosacea. Nat Med 13, 975-980

105. Hachem, J. P., Man, M. Q., Crumrine, D., Uchida, Y., Brown, B. E., Rogiers, V., Roseeuw, D., Feingold, K. R., and Elias, P. M. (2005) Sustained serine proteases activity by prolonged increase in pH leads to degradation of lipid processing enzymes and profound alterations of barrier function and stratum corneum integrity. J Invest Dermatol 125, 510-520

106. Hachem, J. P., Houben, E., Crumrine, D., Man, M. Q., Schurer, N., Roelandt, T., Choi, E. H., Uchida, Y., Brown, B. E., Feingold, K. R., and Elias, P. M. (2006) Serine protease signaling of epidermal permeability barrier homeostasis. J Invest Dermatol 126, 2074- 2086

107. Komatsu, N., Saijoh, K., Kuk, C., Liu, A. C., Khan, S., Shirasaki, F., Takehara, K., and Diamandis, E. P. (2007) Human tissue kallikrein expression in the stratum corneum and serum of atopic dermatitis patients. Exp Dermatol 16, 513-519

108. Jang, H., Matsuda, A., Jung, K., Karasawa, K., Matsuda, K., Oida, K., Ishizaka, S., Ahn, G., Amagai, Y., Moon, C., Kim, S. H., Arkwright, P. D., Takamori, K., Matsuda, H., and 136

Tanaka, A. (2016) Skin pH Is the Master Switch of Kallikrein 5-Mediated Skin Barrier Destruction in a Murine Atopic Dermatitis Model. J Invest Dermatol 136, 127-135

109. Komatsu, N., Saijoh, K., Kuk, C., Shirasaki, F., Takehara, K., and Diamandis, E. P. (2007) Aberrant human tissue kallikrein levels in the stratum corneum and serum of patients with psoriasis: dependence on phenotype, severity and therapy. Br J Dermatol 156, 875-883

110. Bitoun, E., Chavanas, S., Irvine, A. D., Lonie, L., Bodemer, C., Paradisi, M., Hamel- Teillac, D., Ansai, S., Mitsuhashi, Y., Taieb, A., de Prost, Y., Zambruno, G., Harper, J. I., and Hovnanian, A. (2002) Netherton syndrome: disease expression and spectrum of SPINK5 mutations in 21 families. J Invest Dermatol 118, 352-361

111. Mitsudo, K., Jayakumar, A., Henderson, Y., Frederick, M. J., Kang, Y., Wang, M., El- Naggar, A. K., and Clayman, G. L. (2003) Inhibition of serine proteinases plasmin, trypsin, subtilisin A, cathepsin G, and elastase by LEKTI: a kinetic analysis. Biochemistry 42, 3874-3881

112. Descargues, P., Deraison, C., Bonnart, C., Kreft, M., Kishibe, M., Ishida-Yamamoto, A., Elias, P., Barrandon, Y., Zambruno, G., Sonnenberg, A., and Hovnanian, A. (2005) Spink5-deficient mice mimic Netherton syndrome through degradation of desmoglein 1 by epidermal protease hyperactivity. Nat Genet 37, 56-65

113. Descargues, P., Deraison, C., Prost, C., Fraitag, S., Mazereeuw-Hautier, J., D'Alessio, M., Ishida-Yamamoto, A., Bodemer, C., Zambruno, G., and Hovnanian, A. (2006) Corneodesmosomal cadherins are preferential targets of stratum corneum trypsin- and chymotrypsin-like hyperactivity in Netherton syndrome. J Invest Dermatol 126, 1622- 1632

114. Wang, S., Olt, S., Schoefmann, N., Stuetz, A., Winiski, A., and Wolff-Winiski, B. (2014) SPINK5 knockdown in organotypic human skin culture as a model system for Netherton syndrome: effect of genetic inhibition of serine proteases kallikrein 5 and kallikrein 7. Exp Dermatol 23, 524-526

115. Furio, L., Pampalakis, G., Michael, I. P., Nagy, A., Sotiropoulou, G., and Hovnanian, A. (2015) KLK5 Inactivation Reverses Cutaneous Hallmarks of Netherton Syndrome. PLoS Genet 11, e1005389

116. Furio, L., de Veer, S., Jaillet, M., Briot, A., Robin, A., Deraison, C., and Hovnanian, A. (2014) Transgenic kallikrein 5 mice reproduce major cutaneous and systemic hallmarks of Netherton syndrome. J Exp Med 211, 499-513

117. Kasparek, P., Ileninova, Z., Zbodakova, O., Kanchev, I., Benada, O., Chalupsky, K., Brattsand, M., Beck, I. M., and Sedlacek, R. (2017) KLK5 and KLK7 Ablation Fully Rescues Lethality of Netherton Syndrome-Like Phenotype. PLoS Genet 13, e1006566

137

118. Kirihara, T., Matsumoto-Miyai, K., Nakamura, Y., Sadayama, T., Yoshida, S., and Shiosaka, S. (2003) Prolonged recovery of ultraviolet B-irradiated skin in neuropsin (KLK8)-deficient mice. Br J Dermatol 149, 700-706

119. Kitayoshi, H., Inoue, N., Kuwae, K., Chen, Z. L., Sato, H., Ohta, T., Hosokawa, K., Itami, S., Yoshikawa, K., Yoshida, S., and Shiosaka, S. (1999) Effect of 12-O- tetradecanoyl-phorbol ester and incisional wounding on neuropsin mRNA and its protein expression in murine skin. Arch Dermatol Res 291, 333-338

120. Kishibe, M., Bando, Y., Terayama, R., Namikawa, K., Takahashi, H., Hashimoto, Y., Ishida-Yamamoto, A., Jiang, Y. P., Mitrovic, B., Perez, D., Iizuka, H., and Yoshida, S. (2007) Kallikrein 8 is involved in skin desquamation in cooperation with other kallikreins. J Biol Chem 282, 5834-5841

121. Eissa, A., and Diamandis, E. P. (2008) Human tissue kallikreins as promiscuous modulators of homeostatic skin barrier functions. Biol Chem 389, 669-680

122. Haass, N. K., Smalley, K. S., Li, L., and Herlyn, M. (2005) Adhesion, migration and communication in melanocytes and melanoma. Pigment Cell Res 18, 150-159

123. Winnepenninckx, V., Lazar, V., Michiels, S., Dessen, P., Stas, M., Alonso, S. R., Avril, M. F., Ortiz Romero, P. L., Robert, T., Balacescu, O., Eggermont, A. M., Lenoir, G., Sarasin, A., Tursz, T., van den Oord, J. J., Spatz, A., Melanoma Group of the European Organization for, R., and Treatment of, C. (2006) Gene expression profiling of primary cutaneous melanoma and clinical outcome. J Natl Cancer Inst 98, 472-482

124. Rezze, G. G., Fregnani, J. H., Duprat, J., and Landman, G. (2011) Cell adhesion and communication proteins are differentially expressed in melanoma progression model. Hum Pathol 42, 409-418

125. Martins, W. K., Esteves, G. H., Almeida, O. M., Rezze, G. G., Landman, G., Marques, S. M., Carvalho, A. F., LF, L. R., Duprat, J. P., and Stolf, B. S. (2011) Gene network analyses point to the importance of human tissue kallikreins in melanoma progression. BMC Med Genomics 4, 76

126. Lee, N. R., Yoon, N. Y., Jung, M., Kim, J. Y., Seo, S. J., Wang, H. Y., Lee, H., Sohn, Y. B., and Choi, E. H. (2016) Skin Barrier Function Is Not Impaired and Kallikrein 7 Gene Polymorphism Is Frequently Observed in Korean X-linked Ichthyosis Patients Diagnosed by Fluorescence in Situ Hybridization and Array Comparative Genomic Hybridization. J Korean Med Sci 31, 1307-1318

127. Chittock, J., Cooke, A., Lavender, T., Brown, K., Wigley, A., Victor, S., Cork, M. J., and Danby, S. G. (2016) Development of stratum corneum chymotrypsin-like protease activity and natural moisturizing factors from birth to 4 weeks of age compared with adults. Br J Dermatol

138

128. McGovern, J. A., Meinert, C., de Veer, S. J., Hollier, B. G., Parker, T. J., and Upton, Z. (2016) Attenuated kallikrein-related peptidase activity disrupts desquamation and leads to stratum corneum thickening in human skin equivalent models. Br J Dermatol

129. Chao, J., Bledsoe, G., Yin, H., and Chao, L. (2006) The tissue kallikrein-kinin system protects against cardiovascular and renal diseases and ischemic stroke independently of blood pressure reduction. Biol Chem 387, 665-675

130. Lima, A. R., Alves, F. M., Angelo, P. F., Andrade, D., Blaber, S. I., Blaber, M., Juliano, L., and Juliano, M. A. (2008) S(1)' and S(2)' subsite specificities of human plasma kallikrein and tissue kallikrein 1 for the hydrolysis of peptides derived from the bradykinin domain of human kininogen. Biol Chem 389, 1487-1494

131. Andrade, D., Assis, D. M., Santos, J. A., Alves, F. M., Hirata, I. Y., Araujo, M. S., Blaber, S. I., Blaber, M., Juliano, M. A., and Juliano, L. (2011) Substrate specificity of kallikrein-related peptidase 13 activated by salts or glycosaminoglycans and a search for natural substrate candidates. Biochimie 93, 1701-1709

132. Deperthes, D., Marceau, F., Frenette, G., Lazure, C., Tremblay, R. R., and Dube, J. Y. (1997) Human kallikrein hK2 has low kininogenase activity while prostate-specific antigen (hK3) has none. Biochim Biophys Acta 1343, 102-106

133. Bourgeois, L., Brillard-Bourdet, M., Deperthes, D., Juliano, M. A., Juliano, L., Tremblay, R. R., Dube, J. Y., and Gauthier, F. (1997) Serpin-derived peptide substrates for investigating the substrate specificity of human tissue kallikreins hK1 and hK2. J Biol Chem 272, 29590-29595

134. Veveris-Lowe, T. L., Kruger, S. J., Walsh, T., Gardiner, R. A., and Clements, J. A. (2007) Seminal fluid characterization for male fertility and prostate cancer: kallikrein- related serine proteases and whole proteome approaches. Semin Thromb Hemost 33, 87- 99

135. Lilja, H., Oldbring, J., Rannevik, G., and Laurell, C. B. (1987) Seminal vesicle-secreted proteins and their reactions during gelation and liquefaction of human semen. J Clin Invest 80, 281-285

136. Webber, M. M., Waghray, A., and Bello, D. (1995) Prostate-specific antigen, a serine protease, facilitates human prostate cancer cell invasion. Clin Cancer Res 1, 1089-1094

137. Deperthes, D., Frenette, G., Brillard-Bourdet, M., Bourgeois, L., Gauthier, F., Tremblay, R. R., and Dube, J. Y. (1996) Potential involvement of kallikrein hK2 in the hydrolysis of the human seminal vesicle proteins after ejaculation. J Androl 17, 659-665

138. Michael, I. P., Sotiropoulou, G., Pampalakis, G., Magklara, A., Ghosh, M., Wasney, G., and Diamandis, E. P. (2005) Biochemical and enzymatic characterization of human kallikrein 5 (hK5), a novel serine protease potentially involved in cancer progression. J Biol Chem 280, 14628-14635

139

139. Obiezu, C. V., Michael, I. P., Levesque, M. A., and Diamandis, E. P. (2006) Human kallikrein 4: enzymatic activity, inhibition, and degradation of extracellular matrix proteins. Biol Chem 387, 749-759

140. Ramani, V. C., and Haun, R. S. (2008) The extracellular matrix protein fibronectin is a substrate for kallikrein 7. Biochem Biophys Res Commun 369, 1169-1173

141. Sher, Y. P., Chou, C. C., Chou, R. H., Wu, H. M., Wayne Chang, W. S., Chen, C. H., Yang, P. C., Wu, C. W., Yu, C. L., and Peck, K. (2006) Human kallikrein 8 protease confers a favorable clinical outcome in non-small cell lung cancer by suppressing tumor cell invasiveness. Cancer Res 66, 11763-11770

142. Malm, J., Hellman, J., Hogg, P., and Lilja, H. (2000) Enzymatic action of prostate- specific antigen (PSA or hK3): substrate specificity and regulation by Zn(2+), a tight- binding inhibitor. Prostate 45, 132-139

143. Robert, M., and Gagnon, C. (1996) Purification and characterization of the active precursor of a human sperm motility inhibitor secreted by the seminal vesicles: identity with semenogelin. Biol Reprod 55, 813-821

144. Lovgren, J., Airas, K., and Lilja, H. (1999) Enzymatic action of human glandular kallikrein 2 (hK2). Substrate specificity and regulation by Zn2+ and extracellular protease inhibitors. Eur J Biochem 262, 781-789

145. Emami, N., Deperthes, D., Malm, J., and Diamandis, E. P. (2008) Major role of human KLK14 in seminal clot liquefaction. J Biol Chem 283, 19561-19569

146. Fortugno, P., Bresciani, A., Paolini, C., Pazzagli, C., El Hachem, M., D'Alessio, M., and Zambruno, G. (2011) Proteolytic activation cascade of the Netherton syndrome-defective protein, LEKTI, in the epidermis: implications for skin homeostasis. J Invest Dermatol 131, 2223-2232

147. Hachem, J. P., Wagberg, F., Schmuth, M., Crumrine, D., Lissens, W., Jayakumar, A., Houben, E., Mauro, T. M., Leonardsson, G., Brattsand, M., Egelrud, T., Roseeuw, D., Clayman, G. L., Feingold, K. R., Williams, M. L., and Elias, P. M. (2006) Serine protease activity and residual LEKTI expression determine phenotype in Netherton syndrome. J Invest Dermatol 126, 1609-1621

148. Sakabe, J., Yamamoto, M., Hirakawa, S., Motoyama, A., Ohta, I., Tatsuno, K., Ito, T., Kabashima, K., Hibino, T., and Tokura, Y. (2013) Kallikrein-related peptidase 5 functions in proteolytic processing of profilaggrin in cultured human keratinocytes. J Biol Chem 288, 17179-17189

149. Mitsui, S., Okui, A., Uemura, H., Mizuno, T., Yamada, T., Yamamura, Y., and Yamaguchi, N. (2002) Decreased cerebrospinal fluid levels of neurosin (KLK6), an aging-related protease, as a possible new risk factor for Alzheimer's disease. Ann N Y Acad Sci 977, 216-223

140

150. Ogawa, K., Yamada, T., Tsujioka, Y., Taguchi, J., Takahashi, M., Tsuboi, Y., Fujino, Y., Nakajima, M., Yamamoto, T., Akatsu, H., Mitsui, S., and Yamaguchi, N. (2000) Localization of a novel type trypsin-like serine protease, neurosin, in brain tissues of Alzheimer's disease and Parkinson's disease. Psychiatry Clin Neurosci 54, 419-426

151. Scarisbrick, I. A., Sabharwal, P., Cruz, H., Larsen, N., Vandell, A. G., Blaber, S. I., Ameenuddin, S., Papke, L. M., Fehlings, M. G., Reeves, R. K., Blaber, M., Windebank, A. J., and Rodriguez, M. (2006) Dynamic role of kallikrein 6 in traumatic spinal cord injury. Eur J Neurosci 24, 1457-1469

152. Scarisbrick, I. A., Towner, M. D., and Isackson, P. J. (1997) Nervous system-specific expression of a novel serine protease: regulation in the adult rat spinal cord by excitotoxic injury. J Neurosci 17, 8156-8168

153. Blaber, S. I., Ciric, B., Christophi, G. P., Bernett, M. J., Blaber, M., Rodriguez, M., and Scarisbrick, I. A. (2004) Targeting kallikrein 6 proteolysis attenuates CNS inflammatory disease. FASEB J 18, 920-922

154. Spencer, B., Michael, S., Shen, J., Kosberg, K., Rockenstein, E., Patrick, C., Adame, A., and Masliah, E. (2013) Lentivirus mediated delivery of neurosin promotes clearance of wild-type alpha-synuclein and reduces the pathology in an alpha-synuclein model of LBD. Mol Ther 21, 31-41

155. Magklara, A., Mellati, A. A., Wasney, G. A., Little, S. P., Sotiropoulou, G., Becker, G. W., and Diamandis, E. P. (2003) Characterization of the enzymatic activity of human kallikrein 6: Autoactivation, substrate specificity, and regulation by inhibitors. Biochem Biophys Res Commun 307, 948-955

156. He, X. P., Shiosaka, S., and Yoshida, S. (2001) Expression of neuropsin in oligodendrocytes after injury to the CNS. Neurosci Res 39, 455-462

157. Attwood, B. K., Bourgognon, J. M., Patel, S., Mucha, M., Schiavon, E., Skrzypiec, A. E., Young, K. W., Shiosaka, S., Korostynski, M., Piechota, M., Przewlocki, R., and Pawlak, R. (2011) Neuropsin cleaves EphB2 in the amygdala to control anxiety. Nature 473, 372- 375

158. Felber, L. M., Borgono, C. A., Cloutier, S. M., Kundig, C., Kishi, T., Ribeiro Chagas, J., Jichlinski, P., Gygi, C. M., Leisinger, H. J., Diamandis, E. P., and Deperthes, D. (2005) Enzymatic profiling of human kallikrein 14 using phage-display substrate technology. Biol Chem 386, 291-298

159. Kapadia, C., Ghosh, M. C., Grass, L., and Diamandis, E. P. (2004) Human kallikrein 13 involvement in extracellular matrix degradation. Biochem Biophys Res Commun 323, 1084-1090

160. Cloutier, S. M., Chagas, J. R., Mach, J. P., Gygi, C. M., Leisinger, H. J., and Deperthes, D. (2002) Substrate specificity of human kallikrein 2 (hK2) as determined by phage display technology. Eur J Biochem 269, 2747-2754 141

161. Guillon-Munos, A., Oikonomopoulou, K., Michel, N., Smith, C. R., Petit-Courty, A., Canepa, S., Reverdiau, P., Heuze-Vourc'h, N., Diamandis, E. P., and Courty, Y. (2011) Kallikrein-related peptidase 12 hydrolyzes matricellular proteins of the CCN family and modifies interactions of CCN1 and CCN5 with growth factors. J Biol Chem 286, 25505- 25518

162. Yu, Y., Prassas, I., Dimitromanolakis, A., and Diamandis, E. P. (2015) Novel Biological Substrates of Human Kallikrein 7 Identified through Degradomics. J Biol Chem 290, 17762-17775

163. Smith, H. W., and Marshall, C. J. (2010) Regulation of cell signalling by uPAR. Nat Rev Mol Cell Biol 11, 23-36

164. Takayama, T. K., McMullen, B. A., Nelson, P. S., Matsumura, M., and Fujikawa, K. (2001) Characterization of hK4 (prostase), a prostate-specific serine protease: activation of the precursor of prostate specific antigen (pro-PSA) and single-chain urokinase-type plasminogen activator and degradation of prostatic acid phosphatase. Biochemistry 40, 15341-15348

165. Takayama, T. K., Fujikawa, K., and Davie, E. W. (1997) Characterization of the precursor of prostate-specific antigen. Activation by trypsin and by human glandular kallikrein. J Biol Chem 272, 21582-21588

166. Beaufort, N., Debela, M., Creutzburg, S., Kellermann, J., Bode, W., Schmitt, M., Pidard, D., and Magdolen, V. (2006) Interplay of human tissue kallikrein 4 (hK4) with the plasminogen activation system: hK4 regulates the structure and functions of the urokinase-type plasminogen activator receptor (uPAR). Biol Chem 387, 217-222

167. Klucky, B., Mueller, R., Vogt, I., Teurich, S., Hartenstein, B., Breuhahn, K., Flechtenmacher, C., Angel, P., and Hess, J. (2007) Kallikrein 6 induces E-cadherin shedding and promotes cell proliferation, migration, and invasion. Cancer Res 67, 8198- 8206

168. Yoi, O. O., Seldin, D. C., Spragg, J., Pinkus, G. S., and Austen, K. F. (1979) Sequential cleavage of proinsulin by human pancreatic kallikrein and a human pancreatic kininase. Proc Natl Acad Sci U S A 76, 3612-3616

169. Borgono, C. A., Michael, I. P., Shaw, J. L., Luo, L. Y., Ghosh, M. C., Soosaipillai, A., Grass, L., Katsaros, D., and Diamandis, E. P. (2007) Expression and functional characterization of the cancer-related serine protease, human tissue kallikrein 14. J Biol Chem 282, 2405-2422

170. Michael, I. P., Pampalakis, G., Mikolajczyk, S. D., Malm, J., Sotiropoulou, G., and Diamandis, E. P. (2006) Human tissue kallikrein 5 is a member of a proteolytic cascade pathway involved in seminal clot liquefaction and potentially in prostate cancer progression. J Biol Chem 281, 12743-12750

142

171. Rehault, S., Monget, P., Mazerbourg, S., Tremblay, R., Gutman, N., Gauthier, F., and Moreau, T. (2001) Insulin-like growth factor binding proteins (IGFBPs) as potential physiological substrates for human kallikreins hK2 and hK3. Eur J Biochem 268, 2960- 2968

172. Matsumura, M., Bhatt, A. S., Andress, D., Clegg, N., Takayama, T. K., Craik, C. S., and Nelson, P. S. (2005) Substrates of the prostate-specific serine protease prostase/KLK4 defined by positional-scanning peptide libraries. Prostate 62, 1-13

173. Kasza, A. (2013) IL-1 and EGF regulate expression of genes important in inflammation and cancer. Cytokine 62, 22-33

174. Nylander-Lundqvist, E., Back, O., and Egelrud, T. (1996) IL-1 beta activation in human epidermis. J Immunol 157, 1699-1704

175. Nylander-Lundqvist, E., and Egelrud, T. (1997) Formation of active IL-1 beta from pro- IL-1 beta catalyzed by stratum corneum chymotryptic enzyme in vitro. Acta Derm Venereol 77, 203-206

176. Killian, C. S., Corral, D. A., Kawinski, E., and Constantine, R. I. (1993) Mitogenic response of osteoblast cells to prostate-specific antigen suggests an activation of latent TGF-beta and a proteolytic modulation of cell adhesion receptors. Biochem Biophys Res Commun 192, 940-947

177. Shahinian, H., Loessner, D., Biniossek, M. L., Kizhakkedathu, J. N., Clements, J. A., Magdolen, V., and Schilling, O. (2014) Secretome and degradome profiling shows that Kallikrein-related peptidases 4, 5, 6, and 7 induce TGFbeta-1 signaling in ovarian cancer cells. Mol Oncol 8, 68-82

178. Caliendo, G., Santagada, V., Perissutti, E., Severino, B., Fiorino, F., Frecentese, F., and Juliano, L. (2012) Kallikrein protease activated receptor (PAR) axis: an attractive target for drug development. J Med Chem 55, 6669-6686

179. Ramachandran, R., Noorbakhsh, F., Defea, K., and Hollenberg, M. D. (2012) Targeting proteinase-activated receptors: therapeutic potential and challenges. Nat Rev Drug Discov 11, 69-86

180. Hollenberg, M. D., Oikonomopoulou, K., Hansen, K. K., Saifeddine, M., Ramachandran, R., and Diamandis, E. P. (2008) Kallikreins and proteinase-mediated signaling: proteinase-activated receptors (PARs) and the pathophysiology of inflammatory diseases and cancer. Biol Chem 389, 643-651

181. Oikonomopoulou, K., Diamandis, E. P., and Hollenberg, M. D. (2010) Kallikrein-related peptidases: proteolysis and signaling in cancer, the new frontier. Biol Chem 391, 299-310

182. Stefansson, K., Brattsand, M., Roosterman, D., Kempkes, C., Bocheva, G., Steinhoff, M., and Egelrud, T. (2008) Activation of proteinase-activated receptor-2 by human kallikrein- related peptidases. J Invest Dermatol 128, 18-25 143

183. Mize, G. J., Wang, W., and Takayama, T. K. (2008) Prostate-specific kallikreins-2 and -4 enhance the proliferation of DU-145 prostate cancer cells through protease-activated receptors-1 and -2. Mol Cancer Res 6, 1043-1051

184. Krenzer, S., Peterziel, H., Mauch, C., Blaber, S. I., Blaber, M., Angel, P., and Hess, J. (2011) Expression and function of the kallikrein-related peptidase 6 in the human melanoma microenvironment. J Invest Dermatol 131, 2281-2288

185. Noorbakhsh, F., Tsutsui, S., Vergnolle, N., Boven, L. A., Shariat, N., Vodjgani, M., Warren, K. G., Andrade-Gordon, P., Hollenberg, M. D., and Power, C. (2006) Proteinase-activated receptor 2 modulates neuroinflammation in experimental autoimmune encephalomyelitis and multiple sclerosis. J Exp Med 203, 425-435

186. Briot, A., Deraison, C., Lacroix, M., Bonnart, C., Robin, A., Besson, C., Dubus, P., and Hovnanian, A. (2009) Kallikrein 5 induces atopic dermatitis-like lesions through PAR2- mediated thymic stromal lymphopoietin expression in Netherton syndrome. J Exp Med 206, 1135-1147

187. Lu, Y., Papagerakis, P., Yamakoshi, Y., Hu, J. C., Bartlett, J. D., and Simmer, J. P. (2008) Functions of KLK4 and MMP-20 in dental enamel formation. Biol Chem 389, 695-700

188. Oikonomopoulou, K., DeAngelis, R. A., Chen, H., Diamandis, E. P., Hollenberg, M. D., Ricklin, D., and Lambris, J. D. (2013) Induction of complement C3a receptor responses by kallikrein-related peptidase 14. J Immunol 191, 3858-3866

189. Yamamoto, M., Miyai, M., Matsumoto, Y., Tsuboi, R., and Hibino, T. (2012) Kallikrein- related peptidase-7 regulates caspase-14 maturation during keratinocyte terminal differentiation by generating an intermediate form. J Biol Chem 287, 32825-32834

190. Sueiras-Diaz, J., Jones, D. M., Ashworth, D., Horton, J., Evans, D. M., and Szelke, M. (1994) Cleavage of human kininogen fragments at Met-Lys by human tissue kallikrein. Braz J Med Biol Res 27, 1935-1942

191. Yamakoshi, Y., Hu, J. C., Fukae, M., Yamakoshi, F., and Simmer, J. P. (2006) How do enamelysin and kallikrein 4 process the 32-kDa enamelin? Eur J Oral Sci 114 Suppl 1, 45-51; discussion 93-45, 379-380

192. Ramani, V. C., Kaushal, G. P., and Haun, R. S. (2011) Proteolytic action of kallikrein- related peptidase 7 produces unique active matrix metalloproteinase-9 lacking the C- terminal hemopexin domains. Biochim Biophys Acta 1813, 1525-1531

193. Oikonomopoulou, K., Hansen, K. K., Saifeddine, M., Vergnolle, N., Tea, I., Blaber, M., Blaber, S. I., Scarisbrick, I., Diamandis, E. P., and Hollenberg, M. D. (2006) Kallikrein- mediated cell signalling: targeting proteinase-activated receptors (PARs). Biol Chem 387, 817-824

144

194. Yousef, G. M., Scorilas, A., Magklara, A., Memari, N., Ponzone, R., Sismondi, P., Biglia, N., Abd Ellatif, M., and Diamandis, E. P. (2002) The androgen-regulated gene human kallikrein 15 (KLK15) is an independent and favourable prognostic marker for breast cancer. Br J Cancer 87, 1294-1300

195. Kuzmanov, U., Jiang, N., Smith, C. R., Soosaipillai, A., and Diamandis, E. P. (2009) Differential N-glycosylation of kallikrein 6 derived from ovarian cancer cells or the central nervous system. Mol Cell Proteomics 8, 791-798

196. Hutchinson, S., Luo, L. Y., Yousef, G. M., Soosaipillai, A., and Diamandis, E. P. (2003) Purification of human kallikrein 6 from biological fluids and identification of its complex with alpha(1)-antichymotrypsin. Clin Chem 49, 746-751

197. Schilling, O., and Overall, C. M. (2007) Proteomic discovery of protease substrates. Curr Opin Chem Biol 11, 36-45

198. de Veer, S. J., Swedberg, J. E., Parker, E. A., and Harris, J. M. (2012) Non-combinatorial library screening reveals subsite cooperativity and identifies new high-efficiency substrates for kallikrein-related peptidase 14. Biol Chem 393, 331-341

199. Debela, M., Magdolen, V., Schechter, N., Valachova, M., Lottspeich, F., Craik, C. S., Choe, Y., Bode, W., and Goettig, P. (2006) Specificity profiling of seven human tissue kallikreins reveals individual subsite preferences. J Biol Chem 281, 25678-25688

200. Borgono, C. A., Gavigan, J. A., Alves, J., Bowles, B., Harris, J. L., Sotiropoulou, G., and Diamandis, E. P. (2007) Defining the extended substrate specificity of kallikrein 1-related peptidases. Biol Chem 388, 1215-1225

201. Coombs, G. S., Bergstrom, R. C., Pellequer, J. L., Baker, S. I., Navre, M., Smith, M. M., Tainer, J. A., Madison, E. L., and Corey, D. R. (1998) Substrate specificity of prostate- specific antigen (PSA). Chem Biol 5, 475-488

202. Li, H. X., Hwang, B. Y., Laxmikanthan, G., Blaber, S. I., Blaber, M., Golubkov, P. A., Ren, P., Iverson, B. L., and Georgiou, G. (2008) Substrate specificity of human kallikreins 1 and 6 determined by phage display. Protein Sci 17, 664-672

203. Sharma, N., Oikonomopoulou, K., Ito, K., Renaux, B., Diamandis, E. P., Hollenberg, M. D., and Rancourt, D. E. (2008) Substrate specificity determination of mouse implantation serine proteinase and human kallikrein-related peptidase 6 by phage display. Biol Chem 389, 1097-1105

204. Sotiropoulou, G., Rogakos, V., Tsetsenis, T., Pampalakis, G., Zafiropoulos, N., Simillides, G., Yiotakis, A., and Diamandis, E. P. (2003) Emerging interest in the kallikrein gene family for understanding and diagnosing cancer. Oncol Res 13, 381-391

205. Bredemeyer, A. J., Lewis, R. M., Malone, J. P., Davis, A. E., Gross, J., Townsend, R. R., and Ley, T. J. (2004) A proteomic approach for the discovery of protease substrates. Proc Natl Acad Sci U S A 101, 11785-11790 145

206. Staes, A., Impens, F., Van Damme, P., Ruttens, B., Goethals, M., Demol, H., Timmerman, E., Vandekerckhove, J., and Gevaert, K. (2011) Selecting protein N- terminal peptides by combined fractional diagonal chromatography. Nat Protoc 6, 1130- 1141

207. Schilling, O., and Overall, C. M. (2008) Proteome-derived, database-searchable peptide libraries for identifying protease cleavage sites. Nat Biotechnol 26, 685-694

208. Guo, S., Skala, W., Magdolen, V., Briza, P., Biniossek, M. L., Schilling, O., Kellermann, J., Brandstetter, H., and Goettig, P. (2016) A Single Glycan at the 99-Loop of Human Kallikrein-related Peptidase 2 Regulates Activation and Enzymatic Activity. J Biol Chem 291, 593-604

209. Dix, M. M., Simon, G. M., Wang, C., Okerberg, E., Patricelli, M. P., and Cravatt, B. F. (2012) Functional interplay between caspase cleavage and phosphorylation sculpts the apoptotic proteome. Cell 150, 426-440

210. Niessen, S., Hoover, H., and Gale, A. J. (2011) Proteomic analysis of the coagulation reaction in plasma and whole blood using PROTOMAP. Proteomics 11, 2377-2388

211. Shen, C., Yu, Y., Li, H., Yan, G., Liu, M., Shen, H., and Yang, P. (2012) Global profiling of proteolytically modified proteins in human metastatic hepatocellular carcinoma cell lines reveals CAPN2 centered network. Proteomics 12, 1917-1927

212. Fuhrman-Luck, R. A., Stansfield, S. H., Stephens, C. R., Loessner, D., and Clements, J. A. (2016) Prostate Cancer-Associated Kallikrein-Related Peptidase 4 Activates Matrix Metalloproteinase-1 and Thrombospondin-1. J Proteome Res 15, 2466-2478

213. Butler, G. S., Dean, R. A., Smith, D., and Overall, C. M. (2009) Membrane protease degradomics: proteomic identification and quantification of cell surface protease substrates. Methods Mol Biol 528, 159-176

214. Kleifeld, O., Doucet, A., auf dem Keller, U., Prudova, A., Schilling, O., Kainthan, R. K., Starr, A. E., Foster, L. J., Kizhakkedathu, J. N., and Overall, C. M. (2010) Isotopic labeling of terminal amines in complex samples identifies protein N-termini and protease cleavage products. Nat Biotechnol 28, 281-288

215. Yu, Y., Prassas, I., and Diamandis, E. P. (2014) Putative kallikrein substrates and their (patho)biological functions. Biol Chem 395, 931-943

216. Dallas, D. C., Guerrero, A., Parker, E. A., Robinson, R. C., Gan, J., German, J. B., Barile, D., and Lebrilla, C. B. (2015) Current peptidomics: applications, purification, identification, quantification, and functional analysis. Proteomics 15, 1026-1038

217. Bauca, J. M., Martinez-Morillo, E., and Diamandis, E. P. (2014) Peptidomics of urine and other biofluids for cancer diagnostics. Clin. Chem. 60, 1052-1061

146

218. Bery, A., Leung, F., Smith, C. R., Diamandis, E. P., and Kulasingam, V. (2014) Deciphering the ovarian cancer ascites fluid peptidome. Clin Proteomics 11, 13

219. Krastel, P., Liechty, B. M., Meingassner, J. G., Shmitt, E., and Schreiner, E. P. (2009) Cyclic Depsipeptides. Google Patents

220. Barrett, T., Wilhite, S. E., Ledoux, P., Evangelista, C., Kim, I. F., Tomashevsky, M., Marshall, K. A., Phillippy, K. H., Sherman, P. M., Holko, M., Yefanov, A., Lee, H., Zhang, N., Robertson, C. L., Serova, N., Davis, S., and Soboleva, A. (2013) NCBI GEO: archive for functional genomics data sets--update. Nucleic Acids Res 41, D991-995

221. Gentleman, R. C., Carey, V. J., Bates, D. M., Bolstad, B., Dettling, M., Dudoit, S., Ellis, B., Gautier, L., Ge, Y., Gentry, J., Hornik, K., Hothorn, T., Huber, W., Iacus, S., Irizarry, R., Leisch, F., Li, C., Maechler, M., Rossini, A. J., Sawitzki, G., Smith, C., Smyth, G., Tierney, L., Yang, J. Y., and Zhang, J. (2004) Bioconductor: open software development for computational biology and bioinformatics. Genome Biol 5, R80

222. Wu, Z., Irizarry, R. A., Gentleman, R., Martinez-Murillo, F., and Spencer, F. (2004) A model-based background adjustment for oligonucleotide expression arrays. Journal of the American statistical Association 99, 909-917

223. Tanaka, R. J., Ono, M., and Harrington, H. A. (2011) Skin barrier homeostasis in atopic dermatitis: feedback regulation of kallikrein activity. PLoS One 6, e19895

224. Nitzan, Y. B., Sekler, I., and Silverman, W. F. (2004) Histochemical and histofluorescence tracing of chelatable zinc in the developing mouse. J Histochem Cytochem 52, 529-539

225. Planque, C., de Monte, M., Guyetant, S., Rollin, J., Desmazes, C., Panel, V., Lemarie, E., and Courty, Y. (2005) KLK5 and KLK7, two members of the human tissue kallikrein family, are differentially expressed in lung cancer. Biochem Biophys Res Commun 329, 1260-1266

226. Coda, A. B., Hata, T., Miller, J., Audish, D., Kotol, P., Two, A., Shafiq, F., Yamasaki, K., Harper, J. C., Del Rosso, J. Q., and Gallo, R. L. (2013) Cathelicidin, kallikrein 5, and serine protease activity is inhibited during treatment of rosacea with azelaic acid 15% gel. J Am Acad Dermatol 69, 570-577

227. Meyer-Hoffert, U., and Schroder, J. M. (2011) Epidermal proteases in the pathogenesis of rosacea. J Investig Dermatol Symp Proc 15, 16-23

228. Kadomatsu, K., Kishida, S., and Tsubota, S. (2013) The heparin-binding growth factor midkine: the biological activities and candidate receptors. J Biochem 153, 511-521

229. Schlage, P., Egli, F. E., Nanni, P., Wang, L. W., Kizhakkedathu, J. N., Apte, S. S., and auf dem Keller, U. (2014) Time-resolved analysis of the matrix metalloproteinase 10 substrate degradome. Molecular & cellular proteomics : MCP 13, 580-593

147

230. Pope, S. N., and Lee, I. R. (2005) Yeast two-hybrid identification of prostatic proteins interacting with human sex hormone-binding globulin. J Steroid Biochem Mol Biol 94, 203-208

231. Sanchez, W. Y., de Veer, S. J., Swedberg, J. E., Hong, E. J., Reid, J. C., Walsh, T. P., Hooper, J. D., Hammond, G. L., Clements, J. A., and Harris, J. M. (2012) Selective cleavage of human sex hormone-binding globulin by kallikrein-related peptidases and effects on androgen action in LNCaP prostate cancer cells. Endocrinology 153, 3179- 3189

232. Kempkes, C., Rattenholl, A., Buddenkotte, J., Strozyk, E., Eberle, J., Hausser, A., Cevikbas, F., Schneider, S. W., and Steinhoff, M. (2012) Proteinase-activated receptors 1 and 2 regulate invasive behavior of human melanoma cells via activation of protein kinase D1. J Invest Dermatol 132, 375-384

233. Britland, S., and Hoyle, M. (2012) Transcriptional gene silencing of kallikrein 5 and kallikrein 7 using siRNA prevents epithelial cell detachment induced by alkaline shock in an in vitro model of eczema. Biotechnol Prog 28, 485-489

234. Becker-Pauly, C., Barre, O., Schilling, O., Auf dem Keller, U., Ohler, A., Broder, C., Schutte, A., Kappelhoff, R., Stocker, W., and Overall, C. M. (2011) Proteomic analyses reveal an acidic prime side specificity for the astacin metalloprotease family reflected by physiological substrates. Mol Cell Proteomics 10, M111 009233

235. Svensson, S. L., Pasupuleti, M., Walse, B., Malmsten, M., Morgelin, M., Sjogren, C., Olin, A. I., Collin, M., Schmidtchen, A., Palmer, R., and Egesten, A. (2010) Midkine and pleiotrophin have bactericidal properties: preserved antibacterial activity in a family of heparin-binding growth factors during evolution. J Biol Chem 285, 16105-16115

236. Frick, I. M., Nordin, S. L., Baumgarten, M., Morgelin, M., Sorensen, O. E., Olin, A. I., and Egesten, A. (2011) Constitutive and inflammation-dependent antimicrobial peptides produced by epithelium are differentially processed and inactivated by the commensal Finegoldia magna and the pathogen Streptococcus pyogenes. J Immunol 187, 4300-4309

237. Monma, F., Hozumi, Y., Ikematsu, S., Kawaguchi, M., Kadomatsu, K., and Suzuki, T. (2013) Expression of midkine in normal human skin, dermatitis and neoplasms: association with differentiation of keratinocytes. J Dermatol 40, 980-986

238. Nordin, S. L., Sonesson, A., Malmsten, M., Morgelin, M., and Egesten, A. (2012) The epithelium-produced growth factor midkine has fungicidal properties. J Antimicrob Chemother 67, 1927-1936

239. Cohen, S., Shoshana, O. Y., Zelman-Toister, E., Maharshak, N., Binsky-Ehrenreich, I., Gordin, M., Hazan-Halevy, I., Herishanu, Y., Shvidel, L., Haran, M., Leng, L., Bucala, R., Harroch, S., and Shachar, I. (2012) The cytokine midkine and its receptor RPTPzeta regulate B cell survival in a pathway induced by CD74. J Immunol 188, 259-269

148

240. Jham, B. C., Costa, N. L., Silva, J. M., de Miranda, A. C., Oliveira, J. C., Silva, T. A., and Batista, A. C. (2012) Midkine expression in oral squamous cell carcinoma and leukoplakia. J Oral Pathol Med 41, 21-26

241. Tong, Y., Mentlein, R., Buhl, R., Hugo, H. H., Krause, J., Mehdorn, H. M., and Held- Feindt, J. (2007) Overexpression of midkine contributes to anti-apoptotic effects in human meningiomas. J Neurochem 100, 1097-1107

242. Jones, F. S., and Jones, P. L. (2000) The tenascin family of ECM glycoproteins: structure, function, and regulation during embryonic development and tissue remodeling. Dev Dyn 218, 235-259

243. Rodriguez, D., Morrison, C. J., and Overall, C. M. (2010) Matrix metalloproteinases: what do they not do? New substrates and biological roles identified by murine models and proteomics. Biochim Biophys Acta 1803, 39-54

244. Iwasaki, W., Nagata, K., Hatanaka, H., Inui, T., Kimura, T., Muramatsu, T., Yoshida, K., Tasumi, M., and Inagaki, F. (1997) Solution structure of midkine, a new heparin-binding growth factor. EMBO J 16, 6936-6946

245. Murota, H., Matsui, S., Ono, E., Kijima, A., Kikuta, J., Ishii, M., and Katayama, I. (2015) Sweat, the driving force behind normal skin: an emerging perspective on functional biology and regulatory mechanisms. J Dermatol Sci 77, 3-10

246. Burian, M., Velic, A., Matic, K., Gunther, S., Kraft, B., Gonser, L., Forchhammer, S., Tiffert, Y., Naumer, C., Krohn, M., Berneburg, M., Yazdi, A. S., Macek, B., and Schittek, B. (2015) Quantitative proteomics of the human skin secretome reveal a reduction in immune defense mediators in ectodermal dysplasia patients. J Invest Dermatol 135, 759-767

247. Raiszadeh, M. M., Ross, M. M., Russo, P. S., Schaepper, M. A., Zhou, W., Deng, J., Ng, D., Dickson, A., Dickson, C., Strom, M., Osorio, C., Soeprono, T., Wulfkuhle, J. D., Petricoin, E. F., Liotta, L. A., and Kirsch, W. M. (2012) Proteomic analysis of eccrine sweat: implications for the discovery of schizophrenia biomarker proteins. J Proteome Res 11, 2127-2139

248. Mark, H., and Harding, C. R. (2013) Amino acid composition, including key derivatives of eccrine sweat: potential biomarkers of certain atopic skin conditions. Int J Cosmet Sci 35, 163-168

249. Wagener, J. S., Zemanick, E. T., and Sontag, M. K. (2012) Newborn screening for cystic fibrosis. Curr Opin Pediatr 24, 329-335

250. Calderon-Santiago, M., Priego-Capote, F., Turck, N., Robin, X., Jurado-Gamez, B., Sanchez, J. C., and Luque de Castro, M. D. (2015) Human sweat metabolomics for lung cancer screening. Anal Bioanal Chem 407, 5381-5392

149

251. De Giovanni, N., and Fucci, N. (2013) The current status of sweat testing for drugs of abuse: a review. Curr Med Chem 20, 545-561

252. Park, J. H., Park, G. T., Cho, I. H., Sim, S. M., Yang, J. M., and Lee, D. Y. (2011) An antimicrobial protein, lactoferrin exists in the sweat: proteomic analysis of sweat. Exp Dermatol 20, 369-371

253. Csosz, E., Emri, G., Kallo, G., Tsaprailis, G., and Tozser, J. (2015) Highly abundant defense proteins in human sweat as revealed by targeted proteomics and label-free quantification mass spectrometry. J Eur Acad Dermatol Venereol 29, 2024-2031

254. Adewole, O. O., Erhabor, G. E., Adewole, T. O., Ojo, A. O., Oshokoya, H., Wolfe, L. M., and Prenni, J. E. (2016) Proteomic profiling of eccrine sweat reveals its potential as a diagnostic biofluid for active tuberculosis. Proteomics Clin Appl 10, 547-553

255. Dai, X., Okazaki, H., Hanakawa, Y., Murakami, M., Tohyama, M., Shirakata, Y., and Sayama, K. (2013) Eccrine sweat contains IL-1alpha, IL-1beta and IL-31 and activates epidermal keratinocytes as a danger signal. PLoS One 8, e67666

256. Cretu, D., Prassas, I., Saraon, P., Batruch, I., Gandhi, R., Diamandis, E. P., and Chandran, V. (2014) Identification of psoriatic arthritis mediators in synovial fluid by quantitative mass spectrometry. Clin Proteomics 11, 27

257. Waghu, F. H., Gopi, L., Barai, R. S., Ramteke, P., Nizami, B., and Idicula-Thomas, S. (2014) CAMP: Collection of sequences and structures of antimicrobial peptides. Nucleic Acids Res 42, D1154-1158

258. de Veer, S. J., Furio, L., Harris, J. M., and Hovnanian, A. (2014) Proteases and proteomics: cutting to the core of human skin pathologies. Proteomics Clin Appl 8, 389- 402

259. Prassas, I., Chrystoja, C. C., Makawita, S., and Diamandis, E. P. (2012) Bioinformatic identification of proteins with tissue-specific expression for biomarker discovery. BMC Med 10, 39

260. Sanford, J. A., and Gallo, R. L. (2013) Functions of the skin microbiota in health and disease. Semin Immunol 25, 370-377

261. Rieg, S., Steffen, H., Seeber, S., Humeny, A., Kalbacher, H., Dietz, K., Garbe, C., and Schittek, B. (2005) Deficiency of dermcidin-derived antimicrobial peptides in sweat of patients with atopic dermatitis correlates with an impaired innate defense of human skin in vivo. J Immunol 174, 8003-8010

262. Rieg, S., Seeber, S., Steffen, H., Humeny, A., Kalbacher, H., Stevanovic, S., Kimura, A., Garbe, C., and Schittek, B. (2006) Generation of multiple stable dermcidin-derived antimicrobial peptides in sweat of different body sites. J Invest Dermatol 126, 354-365

150

263. Koomen, J. M., Li, D., Xiao, L. C., Liu, T. C., Coombes, K. R., Abbruzzese, J., and Kobayashi, R. (2005) Direct tandem mass spectrometry reveals limitations in protein profiling experiments for plasma biomarker discovery. J. Proteome Res. 4, 972-981

264. Mikhaylova, M., Schumacher, A., Borutzki, C., Neumann, J. R., Macharadze, T., El- Mousleh, T., Wahle, P., Zenclussen, A. C., and Kreutz, M. R. (2014) Analysis of Y- P30/Dermcidin expression and properties of the Y-P30 peptide. BMC Res Notes 7, 400

265. Gkegkes, I. D., Aroni, K., Agrogiannis, G., Patsouris, E. S., and Konstantinidou, A. E. (2013) Expression of caspase-14 and keratin-19 in the human epidermis and appendages during fetal skin development. Arch Dermatol Res 305, 379-387

266. Baechle, D., Flad, T., Cansier, A., Steffen, H., Schittek, B., Tolson, J., Herrmann, T., Dihazi, H., Beck, A., Mueller, G. A., Mueller, M., Stevanovic, S., Garbe, C., Mueller, C. A., and Kalbacher, H. (2006) Cathepsin D is present in human eccrine sweat and involved in the postsecretory processing of the antimicrobial peptide DCD-1L. J Biol Chem 281, 5406-5415

267. Emami, N., and Diamandis, E. P. (2007) Human tissue kallikreins: a road under construction. Clin Chim Acta 381, 78-84

268. Karakosta, T. D., Soosaipillai, A., Diamandis, E. P., Batruch, I., and Drabovich, A. P. (2016) Quantification of Human Kallikrein-Related Peptidases in Biological Fluids by Multiplatform Targeted Mass Spectrometry Assays. Mol Cell Proteomics 15, 2863-2876

269. Ishida-Yamamoto, A., Igawa, S., and Kishibe, M. (2011) Order and disorder in corneocyte adhesion. J Dermatol 38, 645-654

270. Peitsch, W. K., Doerflinger, Y., Fischer-Colbrie, R., Huck, V., Bauer, A. T., Utikal, J., Goerdt, S., and Schneider, S. W. (2014) Desmoglein 2 depletion leads to increased migration and upregulation of the chemoattractant secretoneurin in melanoma cells. PLoS One 9, e89491

271. Chavanas, S., Bodemer, C., Rochat, A., Hamel-Teillac, D., Ali, M., Irvine, A. D., Bonafe, J. L., Wilkinson, J., Taieb, A., Barrandon, Y., Harper, J. I., de Prost, Y., and Hovnanian, A. (2000) Mutations in SPINK5, encoding a serine protease inhibitor, cause Netherton syndrome. Nat. Genet. 25, 141-142

272. Komatsu, N., Saijoh, K., Jayakumar, A., Clayman, G. L., Tohyama, M., Suga, Y., Mizuno, Y., Tsukamoto, K., Taniuchi, K., Takehara, K., and Diamandis, E. P. (2008) Correlation between SPINK5 gene mutations and clinical manifestations in Netherton syndrome patients. J. Invest. Dermatol. 128, 1148-1159

273. Bennett, K., Heywood, W., Di, W. L., Harper, J., Clayman, G. L., Jayakumar, A., Callard, R., and Mills, K. (2012) The identification of a new role for LEKTI in the skin: The use of protein 'bait' arrays to detect defective trafficking of dermcidin in the skin of patients with Netherton syndrome. J. Proteomics 75, 3925-3937

151

274. Brannstrom, K., Ohman, A., von Pawel Rammingen, U., Olofsson, A., and Brattsand, M. (2012) Characterization of SPINK9, a KLK5-specific inhibitor expressed in palmo- plantar epidermis. Biol. Chem. 393, 369-377

275. Redelfs, L., Fischer, J., Weber, C., Wu, Z., and Meyer-Hoffert, U. (2016) The serine protease inhibitor of Kazal-type 9 (SPINK9) is expressed in lichen simplex chronicus, actinic keratosis and squamous cell carcinoma. Arch. Dermatol. Res. 308, 133-137

276. Magister, S., and Kos, J. (2013) Cystatins in immune system. J Cancer 4, 45-56

277. Mukherjee, S., Ukil, A., and Das, P. K. (2007) Immunomodulatory peptide from cystatin, a natural cysteine protease inhibitor, against leishmaniasis as a model macrophage disease. Antimicrob Agents Chemother 51, 1700-1707

278. Blankenvoorde, M. F., van't Hof, W., Walgreen-Weterings, E., van Steenbergen, T. J., Brand, H. S., Veerman, E. C., and Nieuw Amerongen, A. V. (1998) Cystatin and cystatin-derived peptides have antibacterial activity against the pathogen Porphyromonas gingivalis. Biol Chem 379, 1371-1375

279. Jasir, A., Kasprzykowski, F., Kasprzykowska, R., Lindstrom, V., Schalen, C., and Grubb, A. (2003) New antimicrobial cystatin C-based peptide active against gram-positive bacterial pathogens, including methicillin-resistant Staphylococcus aureus and multiresistant coagulase-negative staphylococci. APMIS 111, 1004-1010

280. Schauber, J., and Gallo, R. L. (2007) Expanding the roles of antimicrobial peptides in skin: alarming and arming keratinocytes. J Invest Dermatol 127, 510-512

281. Meyer-Hoffert, U., Wichmann, N., Schwichtenberg, L., White, P. C., and Wiedow, O. (2003) Supernatants of Pseudomonas aeruginosa induce the Pseudomonas-specific antibiotic elafin in human keratinocytes. Exp Dermatol 12, 418-425

282. Vergnolle, N. (2016) Protease inhibition as new therapeutic strategy for GI diseases. Gut 65, 1215-1224

283. Meneton, P., Bloch-Faure, M., Hagege, A. A., Ruetten, H., Huang, W., Bergaya, S., Ceiler, D., Gehring, D., Martins, I., Salmon, G., Boulanger, C. M., Nussberger, J., Crozatier, B., Gasc, J. M., Heudes, D., Bruneval, P., Doetschman, T., Menard, J., and Alhenc-Gelas, F. (2001) Cardiovascular abnormalities with normal blood pressure in tissue kallikrein-deficient mice. Proc Natl Acad Sci U S A 98, 2634-2639

284. Sledz, C. A., and Williams, B. R. (2005) RNA interference in biology and disease. Blood 106, 787-794

285. Ledford, H. (2015) CRISPR, the disruptor. Nature 522, 20-24

286. Chang, A., Yousef, G. M., Jung, K., Rajpert-De Meyts, E., and Diamandis, E. P. (2001) Identification and molecular characterization of five novel kallikrein gene 13 (KLK13;

152

KLK-L4) splice variants: differential expression in the human testis and testicular cancer. Anticancer Res 21, 3147-3152

287. Kapadia, C., Chang, A., Sotiropoulou, G., Yousef, G. M., Grass, L., Soosaipillai, A., Xing, X., Howarth, D. H., and Diamandis, E. P. (2003) Human kallikrein 13: production and purification of recombinant protein and monoclonal and polyclonal antibodies, and development of a sensitive and specific immunofluorometric assay. Clin Chem 49, 77-86

153

Appendices A.1 Tables and Figures Table A 1: Sweat proteases identified in this study

Gene name Protein name MEROPS ID

ADAMTS4 A disintegrin and metalloproteinase with M12.221 thrombospondin motifs 4

ANPEP Aminopeptidase N M01.001

C1R Complement C1R Subcomponent S01.192

C1RL Complement C1R Subcomponent-Like Protein S01.189

CAP1 Prostasin S01.159

CAPN1 Calpain-1 catalytic subunit C02.001

CAPN2 Calpain-2 catalytic subunit C02.002

CASP14 Caspase-14 C14.018

CFB Complement Factor B S01.196

CPA2 Carboxypeptidase A2 M14.002

CPA4 Carboxypeptidase A4 M14.017

CPD Carboxypeptidase D M14.001

CPE Carboxypeptidase E M14.006

CPM Carboxypeptidase M M14.005

CPVL Probable serine carboxypeptidase CPVL S10.003

CTSA Lysosomal protective protein S10.002

CTSB Cathepsin B C01.060

CTSC Dipeptidyl peptidase 1 C01.070

CTSD Cathepsin D A01.009

154

CTSF Cathepsin F C01.018

CTSH Pro-cathepsin H C01.040

CTSL Cathepsin L1 C01.032

CTSS Cathepsin S C01.034

CTSV Cathepsin L2 C01.009

CTSZ Cathepsin Z C01.013

DPP4 Dipeptidyl peptidase 4 S09.003

DPP7 Dipeptidyl peptidase 2 S28.002

FURIN Furin S08.071

GGH Gamma-Glutamyl Hydrolase C26.001

IDE Insulin-degrading enzyme M16.002

KLK1 kallikrein-related peptidase 1 S01.160

KLK10 kallikrein-related peptidase 10 S01.246

KLK11 kallikrein-related peptidase 11 S01.257

KLK13 kallikrein-related peptidase 13 S01.306

KLK5 kallikrein-related peptidase 5 S01.017

KLK6 kallikrein-related peptidase 6 S01.236

KLK7 kallikrein-related peptidase 7 S01.300

KLK8 kallikrein-related peptidase 8 S01.244

KLK9 kallikrein-related peptidase 9 S01.307

LGMN Legumain C13.004

LTF Lactotransferrin S60.001

METAP2 Methionine aminopeptidase 2 M24.002

MMP8 Matrix metalloproteinase-8 M10.002

155

MMP9 Matrix metalloproteinase-9 M10.004

PARK7 Protein Dj-1 C56.002

PEPD Xaa-Pro dipeptidase M24.007

PGC Gastricsin A01.003

PLG Plasminogen S01.233

PM20D1 Probable carboxypeptidase PM20D1 M20.011

PRCP Lysosomal Pro-X carboxypeptidase S28.001

PREP Prolyl endopeptidase S09.001

PRSS3 Trypsin-3 S01.174

PRTN3 Myeloblastin S01.134

SCPEP1 Retinoid-inducible serine carboxypeptidase S10.013

SPPL2A Signal Peptide Peptidase-Like 2A A22.007

THOP1 Thimet oligopeptidase M03.001

TPP1 Tripeptidyl-Peptidase 1 S53.003

156

Table A2: Sweat protease inhibitors identified in this study

Gene name Protein name MEROPS ID

A2M Alpha-2-macroglobulin I39.001

A2ML1 Alpha-2-macroglobulin-like protein 1 I39.007

AHSG Alpha-2-HS-glycoprotein I25.020

AMBP Protein AMBP I02.005

C3 Complement C3 I39.950

CAST Calpastatin I27.001

CD109 CD109 antigen I39.006

CST1 Cystatin-SN I25.010

CST2 Cystatin-SA I25.009

CST3 Cystatin-C I25.004

CST4 Cystatin-S I25.008

CST5 Cystatin-D I25.005

CST6 Cystatin-M I25.006

CSTA Cystatin-A I25.001

CSTB,CST6 Cystatin-B I25.003

HRG Histidine-rich glycoprotein I25.022

KNG1 Kininogen-1 I25.016

PI3 Elafin I17.002

SERPINA12 Serpin A12 I04.091

SERPINB12 Serpin B12 I04.016

SERPINB13 Serpin B13 I04.017

SERPINB2 Plasminogen activator inhibitor 2 I04.007

SERPINB3 Serpin B3 I04.008 157

SERPINB4 Serpin B4 I04.009

SERPINB5 Serpin B5 I04.980

SERPINB7 Serpin B7 I04.012

SERPINB8 Serpin B8 I04.013

SERPINC1 Antithrombin-III I04.018

SERPINF1 Pigment epithelium-derived factor I04.979

SLPI Antileukoproteinase I17.001

SPINK5 Serine protease inhibitor Kazal-type 5 I01.013

SPINK7 Serine protease inhibitor Kazal-type 7 I01.057

SPINK9 Serine protease inhibitor Kazal-type 9 I01.054

SPINT1 Kunitz-type protease inhibitor 1 I02.007

SPINT2 Kunitz-type protease inhibitor 2 I02.009

WFDC12 WAP four-disulfide core domain protein 12 I17.003

WFDC2 WAP four-disulfide core domain protein 2 I17.004

158

Table A3. Selected Sweat identified dermcidin peptides. Peptide sequence (AA) MW 63-109 SSLLEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLDSV 4703.547 SSLLEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDV 3947.146 SSLLEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHD 3848.077 SSLLEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVH 3733.05 SSLLEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGA 3496.923 SSLLEKGLDGAKKAVGGLGKLGKDAV 2511.451 SSLLEKGL 846.4932 SLLEKGLDGAKKAVGGLG 1712.991 LLEKGLDGAKKAVGG 1455.853 66-109 LEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLDSV 4416.399 LEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLD 4230.299 LEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVL 4115.272 LEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVKDV 4002.188 LEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVK 3788.092 LEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDV 3659.997 LEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHD 3560.929 LEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVH 3445.902 LEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAV 3308.843 LEKGLDGAKKAVGGLGKLGKDAVEDLESVGKGA 3209.775 67-101 EKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHD 3447.845 EKGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVH 3332.818 EKGLDGAKKAVGGL 1342.769 EKGLDGAK 817.4416 68-103 KGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVK 3545.966 KGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDV 3417.871

159

KGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHD 3318.802 KGLDGAKKAVGGLGKLGKDAVEDLESVGKGAVH 3203.775 KGLDGAKKAVGGLGKLGKDAVEDLESVGKGA 2967.648 KGLDGAKKAVGGL 1213.727 69-109 GLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLDSV 4046.178 GLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLDS 3947.109 GLDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDV 3289.776 GLDGAKKAVGGLGKLGKDAVEDLESVGKGAV 2938.621 GLDGAKKAVGGLGKLGKDAVEDLESVGKGA 2839.553 GLDGAKKAVGGL 1085.632 70-108 LDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLDS 3890.088 LDGAKKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLD 3803.056 LDGAKKAVGGLGKLGKDAVEDLESVGKGA 2782.532 LDGAKKAVGGLGKLGKDAVEDLESVG 2526.378 LDGAKKAVGGL 1028.61 71-98 DGAKKAVGGLGKLGKDAVEDLESVGKGA 2669.447 DGAKKAVGGLGKLGKDAVEDLESVGKG 2598.41 DGAKKAVGGLGKLGKDAVEDLESVGK 2541.389 DGAKKAVGGLGK 1100.643 DGAKKAVGGL 915.526 72-97 GAKKAVGGLGKLGKDAVEDLESVGKG 2483.383 73-96 AKKAVGGLGKLGKDAVEDLESVGK 2369.34 74-109 KKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLDSV 3632.986 KKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLD 3446.886 KKAVGGLGKLGKDAVEDLESVGKGAVHDVKDVL 3331.859

160

KKAVGGLGKLGKDAVEDLESVGKGAVHDVKDV 3218.775 KKAVGGLGKLGKDAVEDLESVGKGAVHDVKD 3119.707 KKAVGGLGKLGKDAVEDLESVGKGAVHDV 2876.585 KKAVGGLGKLGKDAVEDLESVGKGAVHD 2777.516 KKAVGGLGKLGKDAVEDLESVGKGAVH 2662.489 KKAVGGLGKLGKDAVEDLESVGKGAV 2525.43 KKAVGGLGKLGKDAVEDLESVGKGA 2426.362 KKAVGGLGKL 970.641 75-109 KAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLDSV 3504.891 KAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLDS 3405.823 KAVGGLGKLGKDAVEDLESVGKGAVHDVKDVLD 3318.791 KAVGGLGKLGKDAVEDLESVGKGAVHDVKDV 3090.68 KAVGGLGKLGKDAVEDLESVGKGAVHDVKD 2991.612 KAVGGLGKLGKDAVEDLESVGKGAVHDVK 2876.585 KAVGGLGKLGKDAVEDLESVGKGAVHDV 2748.49 KAVGGLGKLGKDAVEDLESVGKGAVHD 2649.421 KAVGGLGKLGKDAVEDLESVGKGAVH 2534.394 KAVGGLGKLGKDAVEDLESVGKGAV 2397.335 KAVGGLGKLGKDAVEDLESVGKGA 2298.267

161

Table A4. Corneodesmosomes proteins identified in this study

Unique Biological Unique spectrum Total spectrum sample name Protein name peptide count count count

Male Desmoglein-3 2 2 2

Male Desmocollin-2 18 22 23

Isoform 1B of Desmocollin- Male 1 276 340 487

Male Desmoglein-1 369 470 655

Male Corneodesmosin 1 1 1

Male Desmocollin-3 159 178 235

Male Desmoglein-2 2 4 4

Male Corneodesmosin 123 142 185

Male Isoform 2 of Desmoglein-4 1 2 6

Male Desmocollin-2 2 2 3

Isoform 1B of Desmocollin- Male 1 15 21 44

Male Corneodesmosin 4 4 7

Isoform 3B of Desmocollin- Male 3 5 6 8

Male Desmoglein-1 11 17 33

162

Female Desmoglein-3 3 3 3

Isoform 1B of Desmocollin- Female 1 287 371 617

Female Desmoglein-4 1 1 1

Female Desmocollin-3 152 174 252

Female Desmoglein-1 379 493 763

Female Corneodesmosin 118 136 185

Female Desmocollin-2 11 13 14

Female Desmocollin-2 1 1 1

Female Desmoglein-1 11 19 47

Isoform 1B of Desmocollin- Female 1 12 20 55

Isoform 3B of Desmocollin- Female 3 4 4 7

Female Corneodesmosin 3 3 6

163

Table A5. Cystatin proteins identified in this peptidomics study Total Percentage Biological Protein Unique Unique spectrum sequence sample name name peptide count spectrum count count coverage Cystatin- Male SN 10 11 11 0.376 Male Cystatin-A 67 76 107 1 Male Cystatin-S 3 3 3 0.284 Male Cystatin-C 3 3 3 0.171 Male Cystatin-B 19 22 24 0.449 Cystatin- Male SA 3 3 3 0.248 Male Cystatin-M 111 161 226 0.47 Cystatin- Female SN 5 5 6 0.255 Female Cystatin-A 48 53 68 0.816 Female Cystatin-C 9 9 9 0.377 Female Cystatin-B 13 15 16 0.378 Cystatin- Female SA 2 2 2 0.0993 Female Cystatin-M 94 127 192 0.49

164

Figure A1. The number of proteins identified after applying the selection criteria in human normal keratinocyte cell (HaCaT) secretome using the procedure described in Fig 3.1. Venn diagrams show the number of peptides identified in control and KLK7-treated cells.

165

166

167

Figure A2 : Expression pattern of substrate proteins in 8 types of cancer and normal tissues. A, tenascin (TNC); B, cysteine rich angiogenic inducer 61 (CYR61); C, connective tissue growth factor (CTGF); D, fibronectin (FN1); E, insulin-like growth factor binding protein- 3 (IGFBP3); F, cadherin 1 (CDH1).

168

A.2 Ongoing Studies

A.2.1 Characterization of mature form of KLK13

KLK13, also known as KLK-L4, is one of the first KLK genes discovered by our lab (286). Based on previous studies, KLK13 was widely detected in various human tissues and biofluids, with the highest mRNA levels in female reproductive system tissues (i.e. vagina and cervix) (33,103). To date, the putative substrates and biological function of KLK13 are not well characterized.

To gain a better understanding of KLK13, I have produced recombinant active-form KLK13 (36- 277 aa) from P. pastoris (using methods similar as KLK7 production in Chapter 2), and purified the protein with cation strong exchange chromatography. In humans, isoform 1 of KLK13 is the longest at 277 amino acids (aa). Of note, there is some debate regarding the pre- and pro- peptides sequence of KLK13. Previous studies indicated pre-peptide of KLK13 (1-20 aa) and pro-peptide of KLK13 (20-25aa)(287), while pre-peptide of KLK13 (1-16aa) was suggested by PubMed and Uniprot database. The majority of pro-peptides of KLKs end with R/K, indicating that they are cleaved by tryptic proteases. In KLK13, there is K25 located after the pro-peptide sequence. Previous studies in our lab suggested the KLK13 protein (26-277aa) may have little activity. Since proteases networks (i.e. KLKs cascade) exists in vivo, it is possible that chymotryptic-proteases may also contribute to the activation of pro-KLKs. Thus, due to the uncertainty of the sequence of the pro-peptide, we produced the recombinant active-form KLK13 (36-277aa) (F35 in KLK13).

The molecular mass of full-length mature KLK13 is predicted to be approximately 28 kDa, and KLK13 has one predicted glycosylation site at amino acid residue 225. As shown in Fig A3, HPLC Fraction 16 and 22 contained the highest concentrations of KLK13, and were further analyzed. The purity of KLK13 was confirmed by silver staining, western blotting and mass spectrometry. Two bands were observed on the reduced SDS-PAGE corresponding to molecular masses of ~30 kDa (I) and 28 kDa (II). Mass spectrometry analysis validated the presence of KLK13 in both bands, which was further confirmed by western blots (Fig. A3). PNGase F treatment suggested the ~30 kDa band was the glycosylated form of KLK13, while the ~28 kDa band was the non-glycosylated forms (Fig. A3). The N-terminal sequences of the 30 kDa and 28

169

kDa bands was found to be L30PGGY, which is in agreement with the N-terminal sequence of active full-length KLK13.

To further test the enzymatic activity and substrate specificity of purified KLK13, we employed AMC fluorogenic substrates assay. As shown in Table A6, recombinant KLK7 protein exhibited the highest activity against VPR-AMC with Km 0.1559 mM and Kcat/Km 1222.686 mM-1min-1. As expected, it also cleaved many other tryptic substrates. Current studies from our lab support that KLK13 was the most abundant KLKs in female reproductive tract (Carla M.J. Muytjens, Yijing Yu and Eleftherios P. Diamandis. Manuscript in submission). Interestingly, KLK13 detected in sweat and cervical vaginal fluids (CVF) by immune-based capture assay was the active form in both CVF and sweat and the KLK13 expression in CVF was found to be differentially regulated over the menstrual cycle (Fig. A4; Carla M.J. Muytjens, Yijing Yu and Eleftherios P. Diamandis. Manuscript in submission). Ongoing studies are focusing on the regulatory mechanism of KLK13 activity and identification of putative substrates of KLK13.

A.2.2 Identification of putative substrates of KLK9

KLK9 is another chymotryptic-like KLK, with little knowledge about its biological function and substrates. Our lab has produced the recombinant active-form KLK9 protein from mammalian expression system, and biochemical characterization of the purified KLK9 was performed. Moreover, a list of putative substrates of KLK9 was identified using the degradomics approach as described in Chapter 3. Of note, midikine was found to be cleaved by KLK9 in the in vitro optimized condition (Panagiota S. Filippou, Sofia Farkona, Davor Brinc, Yijing Yu, Ioannis Prassas and Eleftherios P. Diamandis. Manuscript in submission). Further biological function studies are under the way to investigate the role of KLK9 in biological settings.

170

Table A.6 Kinetic parameters for the hydrolysis of synthetic AMC substrates by KLK13. Vmax Kcat/Km Normalized Substrate (FU/min) Km (mM) Kcat(min-1) (mM-1min-1) Activity Tryptic-like Boc-VPR-AMC 32560 0.1559 190.616 1222.686 100 Boc-QGR-AMC 890.7 0.005498 2.56 467.14 38 Boc-QAR-AMC 13550 0.1265 40.327 318.79 26 Boc-LGR-AMC 17052 1.1 295.01 268.19 21.9 Boc-QRR-AMC 414.2 0.007399 1.194 161.37 13 Boc-LRR-AMC 2316.5 0.1499 6.679 44.56 3.6 Boc-LKR-AMC 1767.006 0.1562 5.09 32.61 2.7 Boc-GGR-AMC NR NR Tos-GPK-AMC NR NR Boc-EKK-AMC NR NR Chymotrypsin- like AAPF-AMC NR NR LLVY-AMC NR NR Negative Control AAPV-AMC NR NR

NR means no reaction was detected

171

Fig A.3. Characterization of purified recombinant active of protein KLK13. Panel A (left): Silver staining of reduced SDS-PAGE of purified mat-KLK13 (mat=mature). Two chromatographic fractions are shown: fraction 16 (lanes 1, 3) and 22 (lanes 2, 4), and two bands are observed: I (30 kDa) and II (28 kDa). All two bands were found to contain KLK13 sequences by mass spectrometry and were recognized by anti-KLK13 antibodies in western blots (Right panel). Panel B: Coomassie staining of reduced SDS-PAGE of purified mature KLK13 (FPLC fraction 16 and 22) with or without PNGase F treatment. Arrow refers to 28 kDa band of KLK13. Panel C: Amino acid sequence of active-form of KLK13. Catalytic amino acids were showed in brackets. The N-terminal sequence of bands I and II were identical (LPGGYT), which is in accordance with the sequence of active KLK13. Boxed amino acids were based on the data from N-terminal sequencing and underlined amino acid N225 is the potential glycosylation site of KLK13.

172

Catalytic activity of sweat

) u

. 10000 a

( rhK 13

c 8000 Concentrate sweat sample n

e Diluted sweat sample c

n 6000

o u

l 4000

M 2000

- R

P 0 V 0 5 10 Time (min)

Fig. A.4. Catalytically active KLK13 was detected in sweat with immune-capture activity assay. The experiment was followed the protocols previously described (67). The recombinant human KLK13 (rhK 13) represents the active form of recombinant KLK13 protein (6nM) was used as positive control. Concentrated sweat sample was prepared with a pool of 5 different individual sweat samples, and was diluted 1:10 times to make the diluted sweat sample in this study.

173