IDENTIFICATION AND CHARACTERIZATION OF A NOVEL KALLIKREIN GENE, KLK-L41KLK13

Albert Chang

A thesis submitted in conformity with the requirements for the Degree of Master of Science, Graduate Department of Laboratory Medicine and

Pathobiology, University of Toronto

O Copyright by Albert Chang 2001

- -

National Libmry 1+1 d-da BiMiotheque nationale du Canada

- .. . . - - AoguieitioriQa y" 8i iogrâphic Services services bibliiraphques

The author has granted a non- exclusive licence allowing the National Liôrary of Canada to reproduce, loan, distri'bute or sell copies of this thesis in microform, paper or elecüonic formats.

The author retains o w n d p of the copMght in this thesis. Neither the thesis nor substantid extracts fiom it may be printed or othenivise reproduced without the author's pemiission.

L'auteur a ac~ordé une licence non exclusive permettent à la BMioth&que nationale du Cana& de reproche, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/fïh, de reproduction sur papier ou sur fomat Clectronique .

L'auteur conserve la proprieté du droit d'auteur qui protège cette thèse. Ni la thése ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son

Canada

----a - IDENTIFICATION A & I I B C H A ~ C ~ ! t I ~ T I O N OF A NOVEL KALLIKREIN GENE, KLK-L4KLKl3

Albert Chang Master of Science, 2001

Depariment of Laboratory Medicine and Pathobiology University of Toronto

ABSTRACT

Kallikreins are a subgroup of senne proteases and these proteolytic enzymes have diverse

physiological functions in many tissues. Growing evidence suggests that many

kallikreins are implicated in carcinogenesis. In rodents. kallikreins constitute a large

multigene family, but in humans, only three genes were identified. By using the

positional candidate gene approach, we were able to identify a new kallikrein-li ke gene,

tentatively narned KLK-L4 [for kallikrein-like gene 41. This new gene maps to

chromosome l9q 13.3-q 13.4. is fonned of five coding exons and four introns, and shows

structural similarity to other kdlikreins and kallikrein-like genes. KLK-L4 is expressed in

a variety of tissues including prostate, salivary gland, breast and testis. Our preliminary

results show that K L K U is down-regulated, at the rnRNA level, in breast cancer tissues

and breast cancer ce11 lines. Its expression is ngulated by steroid hormones in the breast

cancer ceIl line BT-474. In addition, testis-specific mRNA splice variants have been

found. Their expression has been found in certain normal testicular tissues but not in their

cancerous adjacent counterparts. This gene and its splice variants may be involved in the

AL

pathogenesis andlor progression of various reproductive-organ cancers and may find

applicability as a novel cancer biomarkers.

I would like to express my gratitude to my supervisor, Dr. Eleftherios P. Diamandis, for

his support and encouragement. His enthusiasm and optimism provided the successful

foundation of this work. My gratitude ais0 extends to membea of the Diarnandis

laboratory including Dr. George Yousef, Dr. Liu-Ying Luo, Angeliki Magklara, Christina

Obiezu, Linda Grass and Antoninus Soosaipillai who provided valuable assistance,

suggestions and insights. Special thanks go to my cornmittee, Dr. Khosrow Adeli and Dr.

Bharati Bapat for their mentorship and advice. 1 wish to acknowledge Dr. Paul Walfish

who introduced me to the exciting field of medicai research. I am also indebted to my

family and friends for their tremendous support and optimism.

iii

Table of Contents

Abstract

Acknowledgements

Table of Contents

List of Abbreviations

List of Tables

List of Figures

Introduction

1. Serine Proteases

2. Kallikreins

2.1 Plasma Kallikrein

2.2 Tissue Kallikrein

3. The Human and Rodent Farnilies of Tissue Kallikreins

4. The Human Kallikrein Gene Locus

4.1 Locus Organization

4.2 Gene Organization

4.3 Nomenclature

5. Pmtein Homologies and Predicted Enzymatic Acti vity

6. Hormonal Regulation of Kalli krei ns

7. Tissue Expression of Kallikreins

7.1 mRNA Variants of Kallikrein Transcripts

8. Kallikrein Association with Human Disease

Prostate Specific Antigen (PSA, hK3)

Rationale of Study

Hy pot hesis

Objectives

Materiais and Methds

i

iii S . .

111

vii

viii

i x

1. Identification and Molecular Organization of KLK-LA - - -- - -

DNA sequence on chromosome 19 and prediction of new genes

Expressed sequence tag (EST) searching

Rapid amplification of cDNA ends (3' RACE)

Tissue expression

Breast cancer ce11 line and hormonal stimulation experiments

Reverse transcriptase polymerase chah reaction

Normal and malignant breast tissues

Structure analysis

2. Identification and molecular characterization of mRNA splice variants

of KLK-LA

Identification of ESTs related to KLK-LA

Reverse transcriptase polymerase chah reaction

Screening for KLK-L4 splice variants in testicular cancer tissue

1. Cloning of the KLK-Lcl gene

2. Structural characterization of the KLK-LA gene and i ts protein product

3. Mapping and chromosomal localization of the KLK-L4 p n e

4. Homology with the kallikrein multi-gene family

5. Tissue expression and hormonal regulation of KLK-IA gene

6. Expression of KIX-LA in breast cancer tissues and ce11 lines

7. Identification of KLK-LA splice variants

8. Screening for K L K U variants in testicular cancer tissues

Discussion

1. Genomic Characterization of KW(-L4

--A.- -- 2. Testis-Specific - . Splice Variants of --LA

Conclusions and Future Directions

References

LIST OF ABBREVIGTIONS . - -

ARE bp BCM O c

cDNA CIS DNA m p s EMSPI EST g hK HSCCE HUGO Kb KLX KKS LLNL LNCaP M mg min mL rnM mol mRNA NES 1 PCR PK PS A RACE RNA RT TGFB TLSP TK Pg

androgen response element base pair Baylor College of Medicine degree Celsius complimentary DNA carcinoma in situ deoxyri bonucleic acid deoxynucleoside triphosphates enarnel matrix senne proteinase 1 expressed sequence tag gtam hurnan kallikrein human straturn comeum chyrnotryptic enzyme Human Genome Organization ici lobase kallikrein kallikrein-kinin system Lawrence Livermore National Laboratory Lyrnph node-metstasized prostate cancer molar milligram minute milliliter millirnolar mole messenger ribonucleic acid normal epithelial cell-specific 1 gene. polyrnerase chah reaction plasma kalli krein prostate specific antigen rapid amplification of cDNA ends ribonucleic acid reverse transcriptase transforming growth factor trypsin-like serine protease tissue kalli krein microgram

vii

Table 1: nimers used for RT-PCR analysis 26

Table 2: EST clones with >95% homology to exons of KLK-LA 33

Table 3: Detection and relative abundance of KLK-LA mRNA 62 variant forrns in ten pairs of normal and cancerous testicular tissues

viii

LIST OF FIGURES . -a%.%-- -

Figure 1: Comparative genomic structure analysis of KLK-LA and its splice variants

Figure 2: Comparative genomic structure of the ESTs

Figure 3: Tissue expression of KLK-LA gene

Figure 4: Diagram showing the comparative genomic st~~cutre and PCR Produc t s

Figure 5: Genomic organization and partial genomic sequence

Figure 6: Plot of hydrophobicity/hydrophilicity analysis

Figure 7: Alignment of KLK-LA sequence with members of kallikrein family

Figure 8: A 3ûûkb map of aimost contiguous genomic sequence around chromosome 19q 13.3- 13.4

Figure 9: Dendogram of the predicted phylogenetic m e for KLK-LA and other kallikreins

Figure 10: Hormonal regulation of KLK-U in breast carcinoma ce11 lines

Figure 11: UE-S, UE-L, LAD-L, LAD-S, long KLK-LA protein translation

Figure 12: Plot of hydrophobicity and hydrophilicity of 2 putative proteins encoded by ME-S, LAE-L, LAD-L, LADS, long KLK-L4

Figure 13: Expression of KLK-LA classic, short and long fom in 10 matched testicular tumour pairs

Figure 14: Expression of KLK-LA mRNA variants UE-S, UE-L, LAD-L, L4D-S, in matched testicular tumour pairs

Figure 15: Expression of KLK-L4 classic, short and long fom in testicular NmOrS and embryonal carcinoma-like (EC-like) ce11 lines

INTRODUCTION - - - -

Serine Proteases

Serine proteases are proteolytic enzymes, with a characteristic serine-histidine-

aspartate triad that catalyzes the hydrolysis of peptide bonds (Stryer 1995). Many serine

proteases, including those involved in digestion, share similarities in the primary,

secondary and tertiary structure, as revealed by sequence analysis and X-ray

crystallography (Horton et aL, 1996). Substrate specificities of serine proteases can be

accounted for by relatively smail structural difference in the enzymes (Stryer 1995). The

hydrolysis of peptide bonds by senne proteases starts with a nonçovalent enzyme

substrate complex, orienting the substrate for reaction. An attask by the oxygen atom of

the hydroxyl group on the serine on the carbonyl carbon of the scissile peptide bond. A

proton is removed from the hydroxyl group of the serine, and a nucleophilic oxygen

attacks the carbonyl carbon of the peptide bond to form a tetrahedral intermediate, which

resembles the transition state. The substrate C-O bond changes from a double to a single

bond, allowing the negatively charged oxygen to fom hydrogen bonds. The His and Asp

residues share a proton through a hydrogen bond, and the imidazolium ring of His acts as

an acid catalyst, donating a proton to the nitrogen of the scissile peptide bond, facilitating

its cleavage. The carbonyl group from the peptide foms a covalent bond with the

enzyme, prducing an acyl-enzyme intermediate. After the peptide product with the new

amino terminus leaves the active site, water enters, and this starts the next stage, which is

acylation. The histidine draws r proton away from the water molecule, and the OH- ion

irnmediately attacks the carbonyl carbon atom of the acyl group that is attached to the

serine. A transient tetrahedral intermediate is formed again. The histidine then donates a --- L --- . 9- - - -- -

proton, leading to the collapse of this second tetrahedral intermediate. Hydrol ysis of the

acyl-enzyme intermediate starts when the water donates a proton to the His and provides

an OH- group to attack the carbonyl group of the ester. The enzyme then is ready for

another round of catalysis and the cycle begins again.

Kallikreins

Kallikreins are a group of serine proteases that are found in numerous tissues and

biological fluids. The term "kallikrein" onginated from in the 1930s by Werle and

colleagues, who found high levels of isolates in the pancreas (Bhwla ct al., 1992; Werle

1934). It was used to describe an enzyme that acts on precursor molecule (kininogen) for

nlease of a bioactive peptide (kinin) (Bhoola et al., 1992; Clements 1997). There are two

groups of kallikrein enzymes: plasma and tissue kallikrein, and they differ in molecular

weight, PI, substrate specificity, inununological characteristics, type of kinin released,

and functional importance (Bhwla et al., 1992). Regulatory processes, such as organ

perfusion, systemic blood pressure, sodium and water homeostasis, inflammation,

regulation and maturation of growth factors are regulated by humorally generated kinins

(Chen et aL, 1988; Yu et al., 199th). The regulatory control exhibited by kallikreins

forms an intricate cascade, thought to be involved in regulation of cellular events and

pst-translational processing of polypeptides (Clements 1994).

Piasma Kallikrein -- 2 A

Plasma kallikrein is encoded by a single gene, which is located on human

chromosome 4q35 (Yu et al., 1998a). The gene is composed of 15 exons and encodes for

an enzyme that releases the bioactive peptide bradykinin from a high molecular weight

precursor molecule (high molecular weight kininogen). Plasma kallikrein (PK) is

produced exclusively by the liver. It is involved in surface-dependent activation of blood

clotting, fibrinolysis, kinin generation and inflammation (Mahabeer and Bhoola 2000).

Sequence analysis of human and rat PK shows an overall homology of 75%. suggesting

high hornology between species (Mahabeer and Bhoola 2000). Current literature

suggests that PIC is produced mainly in the liver, but there are also other sources of PK

(Mahabeer and Bhoola 2000).

Tissue Kallikreln

Tissue kallikreins (also known as glandular kallikreins) are memben of a large

multigene family and have significant similarities on both the gene and protein level.

They have no homology to plasma kallikrein. Tissue kallikreins (TK) are acid

glycoproteins, which unlike trypsin, have high specificity for polypeptide substrate

cleavage. Their principal function is thought to be the highly selective cleavage of the

plasma protein kininogen at two peptide bonds to release kallidin (Lys-bradykinin)

(Drinkwater et al., 1988; MacDonald et al., 1988; Werle 1934). TK has been isolated and

identified in numerous tissues. Its immunoreactive and enzyrnatic activity has k e n found

in tissues such as submandibular gland, pmcreas. kidney, spleen, pituitary gland,

placenta, salivary gland, hypothalamus, neutrophils and brain ((Mahabeer and Bhoola

2000) and references therein). The terni kallikrein was initially used to describe an - - - - - -

enzyme that acts upon a precursor molecular (kininogen) for release of a bioactive

peptide (kinin) (Bhoola et al., 1992; Clements 1997; Clements 1994). Among the known

human and animal tissue kallihins, only one enzyme has the ability to release a

bioactive kinin from kininogen. In humans, this enzyme is known as pancreatic or rend

kallikrein, also known as hK1 (human kallikrein 1) protein. It is translated from the

KLKl gene, and it acts upon a liver-derived kininogen to release lysyl-bradykinin (also

known as kallidin), which is involved in the control of blwd pressure, electrolyte

balance, inflammation, and other diverse physiological processes (Clements 1997;

Clements 1994; Margolius 1998; Margolius et al., 1974). Tissue kallikrein (hkl) m2y

also digest other substrates, including growth factors, hormones, cytokines, to mediate

pleiotryopic effects (B hoola et al., 1992).

The Human and Rodent Familles of Tissue Kallikrein Genes - _. _i - .___: _ - - - - _ _ _ _ - 1 --_:- _ -- - _ .. - -

TK belongs to a kalliknin multigene family that has diverse patterns of tissue-

specific gene expression (Bhoola et al., 1992). AI1 genes in the TK family have five

coding exons and 4 introns (Mahabeer and Bhoola 2000; Yousef and Diamandis 2001).

The genomic structure of the tissue kallikrein gene family was examined in detail in both

humans and rodents in the 1980s (Evans et aL, 1987; Murray et al.. 1990; Wines et al.,

1989). It was then concluded that the mouse and rat gene families were composed of

many genes, clustered in the same chromosomal locus. The mouse tissue kallikrein gene

family is localized on chromosome 7 and consists of 24 genes (Evans et al., 1987; Yousef

and Diamandis 2001). The rat tissue kalli krein gene fami l y is composed of approximatel y

20 homologous genes (Wines et al., 1989; Yousef and Diarnandis 200 1). All tissue

kallikreins are closely related, regardless of species (Clements 1994; MacDonald et al.,

1988).

The human TK gene famil y (hK) is located on chromosome l9q 13.3- 13.4. Unti l

recently, the hK gene family was thought to have only three members: KLKI, KLIU,

KLK3, coding for hK1 (pancreatic/renal kallikrein), hK2 and hK3 (also known as

prostate specific antigen) (Evans et al., 1988; Riegman et ai., 1989a; Riegman et al.,

1992). Known as the classic kallikrein genes, they have been examined in much detail for

their role in physiological processes. A strong interest lies within hK2 and hK3 in the

prostate as cancer biomarken. Prostate specific antigen (hK3, or aiso known more

comrnonly as PSA) is widely used as in the diagnosis, monitoring and population

screening of prostate cancer (Diarnandis 2000; Diamandis et al., 2ûûûb). The introduction

of PSA testing has revolutionized the prostate cancer management, and the research field

is still expanding and evolving (Yousef and Diamandis 2001). Over the past few years, - - -

human glandular kallikrein 2 (hK2) has emerged as an additional prostatic cancer

biomarker, and it has been shown to be effective in improving identification and

differential diagnosis of prostate cancer. (Magklara et al., 1999; Rittenhouse et al., 1998;

Stenman 1999). As well, PSA diagnostic applications have extended beyond the prostate,

including breast and other cancers (Black and Diamandis 2000; Black et al., 2000a;

Diarnandis and Yu 1995).

In the last three years, new kallikrein genes that lie on chromosome 19q13.3-13.4

have been discovered. Research groups have cloned a number of novel senne protease

genes that show significant homologies with the classical kallikrein genes. The recent

detailed molecular description of the kallikrein gene locus has allowed researchers to

construct a physical map containing fifteen genes that share significant structural

similarities (Diamandis et al., 2000a; Yousef et al,, 2000b). It should be noted rhat the

genenc term "tissue kallikrein" is not restncted to the description of enzymes that release

bioactive peptides. The tem is now used to desci be a group of enzymes with highl y

conserved gene and protein structure, and colocalized in the sarne c hromosomai locus as

KLKl (Yousef and Diarnandis 2001). Thus, the terni "kallikrein" does not necessarily

imply that any of these family members (except KLKl) have kininogenase activity

(Yousef and Diamandis 2001). For those kallikrein family members that have been

tested, there were found to possess very low (hK2) or no kininogenase activity

(hK3/PSA) (Deperthes et al., 1997).

The Human Kallikrein Gene Locus -3. - - - - .-

Locus Organization

The availability of linear genomic sequences (Venter and et al 2001), including

those mund chromosome 19q 13.3- 13.4 (generated by the Lawrence Livermore National

Laboratory) allowed the precise localization of the fifteen members of the kallikrein gene

famil y to one nucleotide accuracy (Yousef et al.. 2 0 b ) . The three classical kalli kreins,

KLKl , KLK3, KLK2 cluster within a 60 kb region along wi th a newly discovered gene,

KLK15 (Yousef et al., 2ûûûc). The direction of transcription is from telomere to

centromen with the exception of KLK3 and KLK2. The genomic lengths of these genes

are relatively small, ranging from 4 to 10 kb (Yousef and Diamandis 2000). It is possible

that other kallikreins may exist centromenc to KLKl; however, it is unlikely that any

other members will tte found telomeric io KLK14, since the next neighbouring gene is a

Siglec (Foussias et al., 2000). Siglecs are genes that belong to the immunoglobulin

superfamil y and encode for transmembrane receptors that have the ability to bind sialic

acid (Crocker et OZ., 1999). These genes have no structural or functional similarity with

human kallikreins (Yousef and Diarnandis 200 1).

Gene Organization

Al1 members of the new human kallikrein gene family encode for senne

proteases. They consist of five c d n g exons, and the organization of al1 the genes is

similar, with the first coding exon having a short 5'-untranslated region, the second exon

containing histidine, the third exon harboring the aspartic acid and the fifth having the

senne. The histidine-aspartic-serine residues belong to the characteristic catalytic triad - -

conserved among al1 the kallikreins (Yousef and Diamandis 2(X)1). Be yond the stop

codon, there is a 3'- untranslated region of variable length. While the classical kallikreins

do not have 5'- untranslated regions, most members of this multigene family do (Yousef

and Diamandis 2001). In addition, the intron phases between coding exons of al1 these

genes (and those of the rodent kallikreins) are completely conserved, with phases 1-II+

O. (Yousef and Diamandis 2000).

Nomenclature

As many investigators cloned the fïfteen members of the kaIlihein gene family,

the human genome organization (HUGO) has recently proposed guidelines for human

gene nomenclature. Initially, some memben of the new kallikrein gene family were

clnssified under the designation "PRSS" (standing for protease serine) but this was

changed later, since of the lack of distinction between these genes and other senne

proteases that map in different locations of the genome (Yousef and Diamandis 2001).

The construction of the first detailed map allows for a more specific gene syrnbol

assignment. Since other kallikreins have been discovered in other species, official

nomenclature was agreed by HUGO to use the KLK (standing for kallikrein), consistent

with the three classical kallikreins (KLKl, KLK2, KLK3). Gene numbering starts from

centrornere to telomere on chromosome 19q13.4 with the exception of the three classical

kallikreins for which existing nomenclature was retained and one newly discovered gene,

which maps between KLKl and KLK2 (Yousef et al.. 2000~). It is possible, that in the

future, new members of this gene family will be identified, either centromeric to KLKl

or telomeric to KLK14. If new genes are identified, they would be sequentially

-y-- - L - - - - - -

numbered. starring with KLKl6 (Yousef and Diamandis 2001).

Protein Homolodes and Predicted Enmmatic Activity - - - - - --

Maximum homology between the 15 proteins encoded by the kalli krein gene

family is found around the aforernentioned amino acids (histidine, aspartic and serine)

that make up the catdytic triad. In general, the amino acid identity between the 15

proteins is between 40-8045 (Yousef and Diamandis 2001). The number and position of

cysteine residues are highly conserved among the 15 human kallikreins and also among

other senne proteases. Al1 members of this family contain about ten to twelve cysteine

residues, which are for disulfide bonds (Yousef and Diamandis 2001). A number of

invariant amino acids (which tend to be clustered around the active site of serine

proteases) have been descn bed (Day hoff 1 978).

All proteins encoded by the genes are initially synthesized as preproenzyrnes that

are proteolytically processed to yield proenzymes by removal of the N-terminal signal

peptide, followed by activation (also by proteolysis) to the mature, enzymatically active

foms. These cleavage sites have been predicted by computer programs, and have ken

experimentally verified by very few groups (Yousef and Diamandis 2001). Most pro-

forms of the enzymes are aciivated by cleavage at the carboxy terminal end of either

arginine or lysine (K) residues (the preferred trypsin cleavage site). Since most of the

human kallikreins have trypsin-like activity, they may potentially act as activating

enzymes for either themselves (auto-activation) or other pro-forms of kallilcreins.

Kallikreins may participate in cascade pathways similar to those for digestive enzymes,

coagulation, and apoptosis.

Serine proteases can be divided into two main evolutionary families, the trypsin-

like senne proteases and the subtilisin-like pro-protein convertases, which presumably

evolved through convergent evolution (Creighton 1993). The trypsin-like serine

proteases, like many other protein families, are believed to be derived from a single

ancestral gene that evolutionally duplicated, leading to other genes that gradually mutated

to related proteases and protease subfamily with new functions. This large family of

enzymes includes digestive enzymes, blood-clotting and complement enzymes,

granzyrnes and glandular kallikreins. Various serine proteases differ in their substrate

specificity due to small differences in the substrate binding site (Creighton 1993; Lovgren

et aL. 1999). Chyrnotrypsin, which requires an arornatic or bulky non-polar arnino acid

on the carboxy terminal side of the cleavage site, has a non-polar substrate binding

pocket (Creighton 1993). Trypsin-like serine proteases requires an arginine or lysine

residue carboxy tenninal to the cleavage site because there is an aspartic acid in their

binding pockets. The important amino acid of the binding pocket, responsible for

substrate specificity, is usually found six amino acids before the catalytic serine residue.

Out of the fifteen kallikreins, eleven have aspartic acid in this position are expected to

have trypsin-like activity (Yousef and Diamandis 2001). The four remaining enzymes,

hK3, hK7, hK9 and hK15 are expected to have chymotrypsin-like or other specifïc

enzymatic activity (Yousef and Diamandis 2001). The cleavage specificity of al1

kallikreins has yet to be determined, with the exception of hK3 (Liija 1993).

Hormonal Remdation of Kallikreins - -=- - - - . 2 A . - - - - L

The regulation of PSA (KLK3) by steroid hormones has been studied extensively.

Initially, two androgen response elements were identified in the PSA promoter at

positions -170 (AREI) and position -394 (ARE2) (Cleutjens et d, 1996; Riegman et al.,

1991b). These ARES have been functionaily tested and found to be active in LNCaP

prostate cancer cells. More recently, various regions of the 5'-sequences of the PSA gene

m n d -6 to 4 k b demonstrated presence of a putative androgen-response elements at

position 4, 136 (ARE3), which markedly affects PSA transcription androgen stimulation

(Schuur et al., 1996). The hormonal regulatim of the PSA gene is not tissue specific

since PSA has also been found to be regulated by steroid hormones in vitro and in vivo in

breast tissues and bnast carcinoma cell lines (Yu et al., 1994; Zargharni et al., 1997).

Various experiments have documented hormonal regulation of PSA by androgens in the

prostatic carcinoma cell line LNCaP and by androgens and progestins in the breast cancer

ce11 lines BT-474, T-47D and MFM223 (Magklara et al., 2000; Yu et al., 1994; Zarghami

et al., 1997).

The 5'-promoter sequences of KLK2 have been studied dong with identification

of androgen response elements in the prostatic carcinoma cell lines (Munha et al., 1993) . KLK2 was found to be upregulated by androgens and progestins in the breast cancer cell

line T47-D (Hsieh et al., 1997). More recently, a functionally active enhancer region was

discovered 3-5kb upstream from the transcription site of the KLK2 gene (Yu et al.,

1999). In addition, hk2 protein secretion and upregulation by androgens and progestins in

breast cancer cell lines BT-474, T-47D and MFM223 was found (Magklara et al., 2000).

KLKl expression has been studied in animals, and this enzyme has been found

not to be regulated by androgens in either the salivary gland or kidney (Clements et al.,

1988; van Leeuwen et uL, 1986; van Leeuwen et al., 1987). Similarly, no regulation of

KLKl gene by thyroid hormones bas ken demonstrated (van Leeuwen et al., 1987).

Some data has been described with KLKl transcriptionally estrogenically up-regulated.

The demonstration that KLKl expression in human endometrium is higher during the

middle of the menstnial cycle is also suggestive of KLKl up-regulation by estrogens in

this tissue (Clements et al., 1994).

The KLM gene was found to be upregulated by androgens in the prostatic

carcinoma cell line LNCaP (Nelson et al., 1999) and by androgens and progestins in the

breast carcinoma cell line BT-474 (Yousef et al.. 1999). The mode of regulation of KLK2

and KLK4 appears to be similar to that of PSA. Putative androgen response elements in

the promoter region of KLK4 have been identified but not tested (Stephenson et al.,

1999). The remaining 11 human kallikrein genes have been recently identified. Most

preliminary studies regarding hormonal regulation of these new genes have been

perfomied, and most genes appear to be up-regulated by estrogens, androgens and

progestins. but with different potencies (Yousef and Diamandis 2001). It is possible that

some of these genes are honnonally regulated through indirect mechanisms, involving

trans-acting elements (Luo et al., 2000). Investigations characteri ring the promoter and

enhancer regions of these genes have yet to be perfotmed.

- % A L . - & . - Tissue Ex~ression of Kaiiikreins

KLKl gene expression is highest in pancreas, kidney and salivary glands

(Clements 1997). The other two classical kallikrein genes KLK3 and -2, were

thought to be expressed exclusively in the prostate (Papsidero et al., 1980; Wang et al.,

1981); however, recently, using highly sensitive immunological techniques , RT-PCR

and immunohistochemistry, it has been demonstrated that both genes are expressed in

diverse tissues but at lower concentrations than prostatic tissues. In particular, hK3 and

hK2 protein and rnRNA have been found in significant amounts in female breast and at

lower levels in many other tissues. KLK4 also appears to have prostatic-restricted

expression but by RT-PCR, it was demonstrated that it is also expressed in breast and

other tissues. Such frequent coexpression of many kallikreins may suggest a functional

relationship. For example, hK3 and hK2 are frequently coexpressed in the same tissues

and body fluids. In vitro data clearly indicates that hK2, which has trypsin-like activity,

cm activate the proform of PSA (Kumar et al., 1997; Lovgren et al., 1997; Takayama et

al., 1997). Other functional relationships between members of the kallikrein gene family

have not been discovered yet.

mRNA Variants of Kallikrein Transcripts

A number of variant transcnpts have ken identified for the classic and new

kallikreins. The physiological and diagnostic significance of these banscripts has yet to

be defined. Other foms of kallikreins have also been described, such as kallikreins bound

to proteinase inhibitors, intemally clipped kallikreins, and circulating profoms (Catalona

1996; Catalona et al., 1998; Lilja 1997; Stenman et al., 1991). The putative proteins

----- - encoded by these variant transcnpts have not ken isolated. By open reading kame

analysis, it has been predicted that most transcnpts will produce tmncated proteins due to

frarneshifts originating from deleted exons(Heuze et al., 1999; Liu et al., 1999; Rae et al.,

1999; Riegman et al., 1988; Tanaka et al., 2000).

Kallikrein Association with Malinnancv and Disease

Among the human kallikrein family members, the only enzyme with efficient

kininogenase activity is hK1. The biological effects are mediated mainly through kinin

release. Kinins are biologically active peptides thought to be involved in inflammation,

pain generation and regulation of blood pressure, and exert their actions via kinin

receptors. The kailikrein-kinin system is also involved in disease processes such as

hypertension (Margolius et al., 1974), rend disease (Horwitz et al., 1978) and cancer

(Matsumura et al., 1988; Roberts and Gullick 1989). Two kinin receptors have been

identified, kinin B 1 and B2 receptors, which are linked to G-protein coupled second

messengers (Regoli and Barabe 1980). The receptors modulate the cellular actions of

kinins, and are present in low copy number in most natural cells (Blaukat et al., 1999).

The kinin B 1 and B2 receptors have the characteristic membrane-spanning regions linked

by three extracellular and four intracellular lwp regions (Mahabeer and Bhoola 2000).

The B f receptor shares a lirnited sequence homology with the B2 receptor (RodeIl et al.,

1995). B1 receptors have a greater affinity to the kinin metabolites of bradykinin and

kallidin, and appear to induced in some pathologicai processes, such as chronic

inflammation, cancer and trauma (Mahabeer and Bhoola 2000). Kinin action mediated by

B2 receptors has a greater affhity for intact bradykinin and kallidin (Wang et al., 1996).

Molecular studies regarding cDNA encoding of kinin receptors have established the

existence of B 1 and 8 2 recepton as the products of distinct genes (Hess 1997).

The kallikrein-kinin system (KKS) has been implicated the process of

tumourigenesis thmugh the kinin action (Maeda et al., 1999). TIC is known to be present

in several tumours, and may affect tumour growth and proliferation by a number of - %

mechanisms. Preliminary reports indicate that TK is present in breast (Rehbock et al.,

1995). lung (Mahabeer and Bhwla 2000), stomach (Koshikawa et al., 1992), pituitary

(Jones et al., 1990) and utenne tumours (Clements and Mukhtar 1994). Recent studies

demonstrated the pnsence of TK, B2 and B1 receptors in tumour astrocytes and

esophageai carcinoma @lamini et al., 1999; Raidoo et al.. 1999). These results imply

that the KKS may have diagnostic and prognostic relevance. Inhibitors of KKS

components may be therapeutically valuable in neoplastic diseases (Dlamini et al.. 1999).

The role that TK plays in tumour growth needs to be determined. Studies have

demonstrated varying levels of hKLKl and B2 receptor gene expression in endometrial,

pituitary, and prostate cancer tissues (Clements and Mukhtar 1994; Clements 1994)

(Clements et al., 1996). This suggests that there is a functional KKS present in these

tumour systems, and may be important in tumourigenesis (Clements 1997; Clements et

al., 1996) and angiogenesis (Plendl et al., 2000). T K is known to be present in several

tumoun in which increased h W 1 gene expression has been demonstrated. One

example is the human astrocytic tumour cells, in which increased levels of TK have been

demonstrated, while in normal astrocytic cells, there is no immunoreactive TK (Raidoo et

al., 1997). The sequence of KLKl needs to be determined to confirm that it is the hKLKl

gene that is king expressed.

Prostate Specific Antigen (PSA, hK3)

Among ail other kallikreins, the best studied is PSA (hK3), especially in its

application to prostate cancer diagnostics. Although PSA concentration is generaily

elcvated - - - 2 in the - - serum of prostate cancer patients, mRNA down-regulation of PSA in

prostate cancer tissue was observed, in cornparison wi th normal and h yperplastic

prostatic tissues (Hakalahti et al., 1993; Pretlow et al., 199 1). Furthemore, lower tissue

PSA concentration is associated with more aggressive forms of prostate cancer

(Abraharnsson et al.. 1988; Stege et al., 2000). These data agree with those published for

breast cancer, where PSA down-regulation was observed in comparison to normal or

hyperplastic breast tissues, and also in more aggressive forms of breast cancer (Yu et al.,

1996; Yu et al., 1995; Yu et al., 1998b). In addition, lower PSA levels in nipple aspirate

fluids were associated with higher risk for developing breast cancer (Sauter et al., 1996).

Other published data suggest that PSA rnay be a tumour suppressor (Balbay et al., 1999).

an apoptotic inducer (Balbay et al., 1999), negative regulator of cell growth (Lai et al.,

1996), and angiogenic inhi bitor (Fortier et al., 1999).

Another set of investigations suggests PSA may be associated with unfavourable

prognosis/outcome in cancer. For example, breast tumours with higher PSA content do

not nspond to tamoxifen therapy (Foekens et al., 1999). In addition, patients with breast

tumours w hic h produce PS A after medrox yprogesterone acetate (a s yn thetic

progestinlandrogen), have a worse prognosis than patients with tumours that do not

produce PS A @iamandis et al., 1999). Numerous reports demonstrate that PSA may

cleave insulin-like growth factor binding protein-3, thus liberating insulin-like growth

factor, which is a rnitogen for prostatic stroma1 and epithelial cells (Cohen et al., 1991).

PSA may activate latent transfonning growth factor-6 (TGFB), stimulate ce11 detachment,

and facilitate Nmour spreading (Killian et ai., 1993). Like other senne proteases, PSA

-----;- - - may mediate proteolysis of basement membrane, leading to invasion and metastasis

- - - -

(We bber et al., 1995).

Hence, current clinical data produce conflicting results in defining PSA's role in

cancer. Differences in methodology, purity, patient selection, and source of PSA

preparations may account for differences in PS A response (Yousef and Diamandis 200 1).

Furthemore, the lack of knowledge of biological pathways in which PSA is participating

poses significant difficulties in interpreting these clinical observations (Diamandis 2 0 ) .

Nonetheless, data suggests PSA involvement in not just prostate cancer (as its acronym

suggests), but in breast cancer as well. Further studies have yet to determine roles of

PSA and other kallikreins in various fonns of cancer.

-a- - - A

RATIONALE OF STUDY

In Our effort to identify new kallikrein-like genes that might be useful as

diagnostic andor prognostic marken for cancer, we studied a genornic area of -300 kb

around chromosome l9q 13.3-q 13.4, where the known human kalli krein genes are

localized (the classical ka1 likreins). Our laboratory studies concentrated on characterizing

the human kallikrein gene farnily at the nucleotide level, and we identified seven novel

putative genes on chromosome l9q 13.3- 13.4. One of these genes was tentative1 y called

KLK-LA, also known as KLK13. Nothing is known about this gene except that its

predicted exons harbour a catalytic enzymatic tnad that is conserved among PSA and the

other kallikreins. In addition, testis-specific mRNA splice variants of KLK-LA (-13)

were discovered. Given that this gene shares homology to PSA and other kallikreins, our

present investigation was to genomically charncterize KLK-LA and its mRNA splice

variants and to examine its role in several reproductive cancers.

Based on the following facts that:

1) kallikreins are members of a large multigene fmily and have significant similarities

on both the gene and protein level

2) kalli kreins are found in various cancers, including reproducti ve-organ ones, in vanous

molecular forms

3) many kallikreins are homonally regulated by steroids. such as androgens, progestins

and estrogens

4) certain kallikreins (e.g. PSA and hK2) have valuable clinical utility and significance

it is hypothesized that KLK-LA is a homonally regulated gene that exists in multiple

molecular forms in various tissues. This gene and its splice variants may have important

clinical utility and significance.

OBJECTIVES . . . - . .

9 To identify and genomically characterize KLK-LA

variants

and its tissue-specific splice

m To characterize and examine expression of KLK-LA and its splice variants in various

reproductive-organ tumors

9 To perfonn hormonal regulation studies exmining KLK-IA expression in breast

cancer ceIl lines T-47D, MCF-7, BT-474 and BT-20

MATERIALS AND METHODS

1. Identification and Molecular Cbaracterization of KLK-L4

DNA sequence on chromosome 19 and prediction of new genes

We have obtained sequencing data of approximately 300Kb of nucleotides,

around chromosome 19q13.3-q 13.4, from the web site of the Lawrence Livermore

National Laboratory (LLNL) (htto://www-bio.llnl.~ov/genomel~enome.html) - and

constructed an almost contiguous stretch of genornic sequences. A number of computer

prograrns were used to predict the presence of putative new genes in this genornic area

(Yousef 1999).

Exprossed sequence tag (EST) searching

The predicted exons of the putative new gene were subjected to homology search

using the BLASTN algorithm (Altschul et al., 1997) on the National Center for

Biotechnology Information web server (http://www ncbi.nlm.nih.gov/BLAST/) against

the human EST database (dbEST). Clones with > 95% homology were obtained from the

I.M.A.G.E. consortium (Lennon et al., 1996) through Research Genetics Inc, Huntsville,

AL. The clones were propagated, purified and sequenced from both directions with an

automated sequencer, using insert-flanking vector primers.

Rapid ampMIication of cDNA ends (3' RACE)

According to the EST sequence data and the predicted structure of the gene, two

gene-specific prirners were designed and two rounds of RACE reactions (nested PCR) -L-J-- - --A - -

were performed with 5p1 Marathon Readym cDNA of human testis (Clontech, Pa10 Alto,

CA, USA) as a template. The reaction mix and PCR conditions used were according to

the manufacturer's recommendations.

Tissue expression

Total RNA isolated from 26 different human tissues was purchased from

Clontech. We prepared cDNA as described below, and used it for PCR reactions with

different sets of prirners (Table 1). Tissue cDNAs were amplified at various dilutions.

Breast cancer ce11 Une and hormonal stimulation experiments

The breast cancer ce11 lines BT-474, T47-D, MCF-7 and BT-20 were purchased

from the American Type Culture Collection (ATCC), Rockville, MD. Cells were cultured

in RPMI medi a (Gi bco BRL, Gaithersburg, MD) supplemented wi th glutamine (200

mmoüL), bovine insulin (10 mg/L), fetal bovine serum (10%). antibiotics and

antirnycotics, in plastic flasks, to near connflency. The cells were then aliquoted into 24-

well tissue culture plates and cultured to 50% confluency. 24 hours before the

experiments, the culture media were changed into phenol red-free media containing 10%

charcoal-stripped fetal bovine serum. For stimulation experiments, various steroid

hormones dissolved in 100% ethanol were added into the culture media, at a final

concentration of lu8 M. Cells stimulated with 100% ethanol were included as controls.

Ethanol concentration was ~ o - ~ M . The cells were cultured for 24 hours, then harvested

for mRNA extraction.

a- - - -

Reverse transcriptase polymerase chah reaction

Total RNA wos extracted from the breast cancer tissues and ce11 lines using

~ r i z o l ~ nagent (Gibco BRL) following the manufacturer's instructions. RNA

concentration was detemined spectrophotometrically. 2 pg of total RNA was reverse-

transcribed into fiat strand cDNA using the ~ u ~ e r s c r i ~ t ~ preamplification system

(Gibco BRL). The final volume was 20 pl. Based on the combined information obtained

from the predicted genomic structure of the new gene and the EST sequences, two gene-

specific primers were designed (LA-FI and U-RI, see Table 1) and PCR was carried out

in a reaction mixture containing 1 pl of cDNA, 10 rnM Tris-HCI (pH 8.3), 50 rnM KCI,

1.5 m M MgCl*, 200 pM cNTPs (deoxynucleoside triphosphates), 150 ng of primers and

2.5 units of AmpliTaq Gold DNA polperase (Roche Molecular Systerns, Branchburg.

NJ, USA) on a Perkin-Elmer 9600 thermal cycler. The cycling conditions were 94°C for

9 minutes to activate the Taq Gold DNA polymerase, followed by 43 cycles of 94OC for

30 s, 63OC for 1 minute and a final extension st 63°C for 10 min. Equd arnounts of PCR

products were electrophoresed on 2% agarose gels and visualized by ethidium bromide

staining. Al1 prîrners for RT-PCR spanned at least 2 exons to avoid contaminition by

genornic DNA.

To verify the identity of the PCR products, they were cloned into the pCR 2.1-

TOPO vector (Invitmgen, Carlsbad, CA, USA) according to the manufacturer's

instructions. The inserts were sequenced from both directions using vector-specific

primers, with an automated DNA sequencer.

Table 1. Rimers used for reverse transcription polymerase chah reaction (RT-PCR)

Gene Primer name Sequence' KLK-LA LA-Fl AACTCTACAATGTGCCAACA

LA-A AGGCTGCCCTACTAGTGCAA LA-B ATATTGCCTAGGTGGATGTG LA-D AAGACTTCAAGGAGCCAAGC LA-E GACCCTTCACCTCCCAAAAT

LA-X 1 CTAGTGATCGCCTCCCTGAC

PSA

Actin

PSAS TGCGCAAG'iTCACCCTCA PSAAS CCCTCTCCITACTTCATCC

ACTINS ACAATGAGCTGCGTGTGGCT ACTINAS TCTCCTTAATGTCACGCACGA

1. Al1 nucleotide sequence are given in the 5' +3' orientation.

Norxnai breast tissues were obtained from women undergoing reduction

mammoplasties. Breast tumor tissues were obtained from female patients at participating

hospitals of the Ontario Provincial Steroid Hormone Receptor Rogram. The normal and

tumor tissues were immediately frozen in liquid nitrogen after surgical resection and

stored in this manner until extracted. The tissues were pulverized with a h m e r at dry

ice temperature and RNA was extracted as described above, using Trizol reagent.

Structure analysis

Multiple alignment was performed using the Clustal X software package available

at: ft~://ft~.ebi.ac.u~ublsoftware/doslclustalw/clustalx/ (clustalx 1.64b.msw.exe) and the

multiple alignment prograrn available from the Baylor College of Medicine (BCM),

Houston, TX, USA (kiwi.imeen.bcm.tmc. edu:8808/search-Iaunchernauncher/html).

Phylogenetic studies were performed using the Phylip software package available ai:

htt~://evolution.eenetics. wnshin~ton.edu/phvli tme.html. Distance matnx analysis

was perfonned using the b'Neighbor-JoiningNPGMA'T prograrn and parsimony anal ysis

was done using the "Protpars" prograrn. Hydrophobicity study was performed using the

BCM search launcher progrms (htti>://dot.i meenhcm. tmc.edu:933 llsea-searchlstruc-

predict.htm(). Signal peptide was predicted using the S igna lP server (httv://www.cbs.

dtu.dklservices1 simal). Rotein structure anal ysis was perfonned b y the "S APS"

(structural anal ysis of protein sequence) prograrn (h tt~:lldot.im~en.bcm.tmc.edu:

933 Ilsea-search/stmc-~redict,html).

---- *.- - 2. Identification and molecular chamcterization of mRNA ~ ~ l i c e variants of KLK-

L4 -

IdentifiePtion of ESTs related to KLK-L4 gene

The cDNA sequence of the KLK-LA gene was subjected to homology search

using the BLASTN algorithm at the National Center for Biotechnology Information Web

Server (htt~:l/www.ncbi.nlm.nih.~ov/BLAST) against the human EST database (dbEST).

Clones with >95% homology were obtained from the I.M.A.G.E. consortium through

Research Genetics Inc., Huntsville, AL, USA. The clones were propagated, pun fied and

sequenced from both directions using insert-flanking vector pnmers. Based on the

combined information obtained from the K U - L 4 genornic sequence (Genbank accession

#AF13SO24) and the ESTs, two gene-specific fonvard primers were designed, LA-D and

L4-E (Table 1 and Fig. 1). The reverse primer LA-RI binds to exon 5 (Table 1 and

Figure 1). In addition, a second reverse primer, named LA-R2, was designed to bind to

exon 2 of the KLK-IA classic fom (Table 1 and Flg. 1). Screening of testicular cDNA

with LA-XI (a forward primer binding to exon 1; see Figure 1 and Table 1) and LA-R2

was then performed. Al1 selected primers were designed to amplify specific altematively

spliced forms, as further exempli fied below .

Reverse transcriptase polyrnerase chah reacüon

Total RNA isolated from 26 different human tissues was purchased from

Clontech, Pa10 Alto, CA. Two pg of total RNA was reversed-transcribed into fiat strand

cDNA using the Superscript preamplification system (Gibco BRL, Gaithersburg, MD).

Figure 1: Comparative genomic structure of the KLK-IA variants UD-L, LAD-S, L4E- L, IAE-S, KLK-LA long and short variants and of the classic KLK-LA gene. Exons are represented by solid bars and introns by the connecting lines. Numbers in the boxes and above the lines represent the sizes of exons and introns, respectively, in base pairs. The mow heads represent stop codons and the asterisks start codons. Numbers in parenthesis indicate length of ptedicted polypeptide in amino acids. Rimers and their directions are illustrated on the bottom. For details see text,

The final volume was 20 pl. - Using - the three gene-specific forward pnmers with

reverse primer U R I , the tissue panel was screened by PCR. PCR was carried out in a

reaction mixture containing 1 pl cDNA, 2.5 mM MgC12, 25mM PCR buffer, 200 p M

dNTPs (deoxynucleoside triphosphates), 150 ng primers and 1 unit of HotStar Taq

polymerase (Qiagen, Valencia, CA) on a Eppendorf Mastercycler gradient system

(Eppendorf, Westbury, NY). The cycling conditions were 95OC for 15 minutes to activate

the HotStar Taq polyrnerase, followed by 35 cycles of 94OC for 30 s, 63OC for 1 minute

and a final extension at 72°C for 10 minutes. The amplified splice variants were separated

using 1.5% agarose gels. The isolated bands were then purified using the Qiagen gel

purification kit. The PCR products were cloned into the pCR 2.1 -TOPO vector

(Invitmgen, Carlsbad, CA. USA) in order to venfy their identities. Insens were

sequenced from both directions wing vector-speci fic pnmers with an automated DNA

sequencer,

Screening for the KLK-LA splice variants In testicular cancer tissue

Included in this study were tissue samples from patients who had undergone

radical orchiectomy for testicular cancer at the Chdte University Hospital, Germany,

and the National University Hospital, Denmark. Patient age ranged from 23-60 years with

a median of 36. Matched testicular tissue samples were obtained from the tumon and

adjacent morphologicdly normal parts of the same testis. Al1 patients had a

histologicallyîonfinned diagnosis of primary testicular cancer and received no treatment

before surgery. Of the ten matched samples, five tumors were seminomas, three were

categorized as embryonal carcinomas and two were teratocarcinomas. Other tissue

samples included one Leydig ceIl tumor and two samples of testicular tissue containing .-..- . - a...-- .-= L - - - -

tubules with pre-invasive carcinoma-in situ (CIS) within morphologically normal

pannchyma. In addition, the expression of KLK-LA and its splice variants was examined

in two embryonal cell lines derived from a human teratocarcinorna, one pluripotent

(NTERA-2) and one nullipotent (2102Ep).

Tissue specimens were rninced with a scalpel, on ice. and immediately transferred

into a 2ml polypropylene tube. Tissue samples were then homogenized, and RNA

extnicted using the RNeasy Mini Spin column method fmm Qiagen, according to

manufacturer's recommendations. The RNA concentration was deterrnined

spectrophotometncally, and 2 pg of total RNA was reverse-transcribed into fint strand

cDNA as described above.

RESULTS

Cloning of the KLK-L4 gene

Cornputer analysis of the genomic sequence around chromosome 19 q13.3q13.4

predicted a putative new gene formed of at least 3 exons. To experimentally verify the

existence of this gene, the putative exons were subjected to sequence homology search

against the human expressed sequence tag (EST) database (dbEST), and four EST clones

with > 97% homology were identified (Table 2). Al1 ESTs were cloned from testicular

tissue. These clones were obtained and inserts were sequenced from both directions. The

"Blast 2 sequences" program was used to compare the EST sequences with the predicted

exons, and final selection of the exon-intron splice sites was done according to the EST

sequences. The presence of many areas of overlap between the various EST sequences

allowed us to funher verify the structure of the new gene. The coding and genomic

sequence of the gene has been deposited in GenBank (accession # AF135028).

As shown in Figure 2, three ESTs match almost perfectly with the predicted 3

exons (exons 3,4,5) of the gene and one EST matches perfectly with predicted exons 3

and 5. However, each of the ESTs extends funher upstream with different exonic

patterns, suggesting the presence of different splice variants. Attempts to translate these

clone sequences demonstrated the presence, in some ESTs, of intempting stop codons in

al1 three possible reading h e s . Homology search of the three common exons against

the GenBank database revealed a cDNA sequence from the German Human Genome

Roject. This clone has an identical exon 2 as the long form of KLK-L4 gene (this form

will be described below) but has an extended exon 3 that ends with a stop codon (Figure

----* -- - - - - -. -. - Table-2: EST clones with >95% homology to exons of KLKU GenBank # Tissue of I.M.A.GE. ID

origin AA399955 Testis 7431 13

AM4677 1 Testis 1392889

AI002101 Testis 1619045

AI032327 Testis 1644236

Exon number EST ID

AA399955

AA8477 1

A1002 101

AI032327

U=Xl U-E LA-D U - B L4-A IA-Fl IL, -b .II, -b + +

Figure 2. Comparative ~ n o m k stnicnue of the ESTs (Table 2), the clone fiom lk Oerman Gemme Project, and the ciassic fonn KLK-LA. Exons aie represented by solid bars and m t m s by the connecting lines. Exon numbers on top of sdid bars refer to our GenBank submission #AF135024. The EST IDs represent GenBank accession numbers. Asterisks represnt the positions of stop codons. Horizontal amnvs Bdkate the direction of the PCR primrs (descriid in Table 1) and anowheads their pition abng the exons. Vertical dotted iines show alignment of identical fragments. For inore detaik, see text.

1). This clone was isolated from uterine - tissue - and is translated into a tmncated protein

product of 196 amino acids which is followed by a 3' untranslated region (GenBank

accession # AUlS0220).

Screening of cDNAs from 26 different tissues by RT-PCR, using gene-specific

pnmers for exons 3 and 5 (LA-FI and LA-RI) (Table 1 & Figure 2) revealed that this

gene is expressed in many tissues. Four tissues that show the highest level of expression

(salivary gland, mammary gland, prostate, and testis) (Figure 3) and uterus (the EST

clone ALû50220 was isolated from this tissue) were selected for identification of the full

stnicture of the gene. Different PCR reactions were performed using one reverse primer

(L4-RI) together with each of the forward pnmers located in upstream exons that were

found in the different EST clones (pnmers L4-A, U - B . L4-D, LA-E) (Table 1 & Figure

2). The PCR reactions were performed under different experimental conditions, using the

EST clones as positive controls, and the PCR products were sequenced. None of these

forms were found in any of the tissues, except in testis [data not shown].

By RT-PCR of the KLK-L4 gene using primers LA-RI and LA-FI, it was found

that the gene is expressed in a wide variety of tissues (Figure 3). In order to obtain the

structural forms that exist in these tissues. a hornology study was performed. Aligning

the predicted polypeptide of the KLK-LA gene with al1 other kallikreins and kallikrein-

like genes, suggested, by homology, that at least two more exons should be present

upstream of the predicted three exons. The genomic fragment upstream of the third exon

was subjected to further computer analysis for gene prediction, and exon 2 was identified . based on: a) a consensus exodintron splice site b) preservation intron phase II after this

exon, in agreement to intmn phases of al1 other known kallikreins (see discussion for

Figure 3. Tissue expression of KLK-LA gene as determined by RT-PCR. Actin and PSA are control genes. KLK-LA is highly expressed in breast, prostate, salivary gland and testis.

details] c) presence of the histidine residue of the cataiytic triad (H'~) surrounded by a - . . --

welltonserved peptide motif [see below] just before the end of this exon d) comparable

exon length to other kallikrein genes. A potential first exon was also predicted from the

upstream genomic sequence, based on the preserved intron phase (phase 1), and the

existence of an in-frarne start codon that is located at a comparable distance [in relation to

other kallikreins] from the end of this exon. To venfy this predicted structure, a PCR

reaction was perfomed using one reverse primer (LA-RI) together with another fonvard

primer that is located in the predicted first exon (primer L4-X 1) (Table 1 & Figure 2).

Two main PCR bands were obtained from the tissues examined; the expected 819 bp

band (predominant) and an additional minor band of about 650 bp (Figure 4). Cloning

and sequencing of these two bands revealed that the gene exists in two main fomis i?i

these tissues; the classic fom (our GenBank submission # AF135024) and another form

(referred to as the shon KLK-L4 variant) that utilizes an upstream alternative splice

donor site, located inside exon 3, thus creating an mRNA product that that is 214 bp

shorter. This alternative splice site causes frame-shifting of the coding region that will

generate a predicted stop codon at the beginning of exon 4, giving rise to a truncated

protein product that does not contain the senne residue of the catalytic triad (Figures 4

and 5).

Aligmng the classic KLK-LA form with the ESTs (Figure 2) demonstrated that al1

ESTs utilize a different splice donor site located 80 bp downstrearn from the end of exon

3. These additional 80 bp contain an in-frame stop codon at nucleotide position 5505

which will lead to the formation of a shorter polypeptide product. They also utilize an

alternative pol yaden ylation signal located at p s i tion 8706 (numbea refer to our

KLK-i.4 long form

Short KLK-LA variant

Fipre 4. Upper Panel: Diagram showing the comparative genornic structure of the long KLK-LA form and the short KLK-LA variant. Exons are represented by boxes and introns by the connecting lines. Exon numben refer to Our GenBank submission # AF135024. The black region indicates the extra fragment (214 bp) that isfound in the long, but not in the short form of the gene (see text for details). The positions of the stop codons of the two foms are marked with astensks. Frame shifting occurs as a result of utilization of an alternative splice site, and a stop codon is generated at the beginning of exon 4 in the short fonn. Lower Panel: PCR ptoducts of the amplification of the KLK-LA gene using U - R I and LA-XI primers (Figure 1 and Table 1). Note the predominant longer fonn (also hown as classic) and a minor bond representing the shoa fom of KLK-LA rnRNA. (M); Markers with sizes in bp shown on the left. Tissues used: (l), salivary gland; (2), mammary gland; (3), prostate; (4), testis; (5). uterus; (6). breast cancer tissue; (7), negative control.

--- ., - - TCAGGCCCCG~CCG~C~TGCC~TCCC~TCCCGATCCCGG ~ T G G ccc CTG GCC - - -- - -- M W P L A CTA GTG ATC GCC TCC CTG ACC TTG GCC TTG TCA GGA G.. .&taaga.. .. intron 1 . . ... ttaccag L V I A S L T L A L S G

GT GTC TCC CAC GAG TCT TCC AAG GTT CTC AAC ACC AAT GGG ACC ACT GGG TTT G V S Q E S S K V L N T N G T S G F

CTC CCA GGT CGC TAC ACC TGC TTC CCC CAC TCT CAG CCC TCG CAC GCT GCC L P G G Y T C F P H S Q P W Q A A

CTA CTA GTG CAA GGG CGC; CTA CTC TGT GGG GGA GTC CTG GTC CAC CCC AAA L L V Q G R L L C G G V L V H P K

TGG GTC CTC ACT GCC GCA CAC TGT CTA AAG GA @tgt . .... intron 2.. .. . .. ciic-g G CGC W V L T A A F I C L K E G

CTC AAA GTT TAC CTA CGC AAG CAC GCC CTA GGG CGT GTC GAA GCT GGT GAG L K V Y L G K H A L G R V E A G E

CAG GTG AGG GAA GTT GTC CAC TCT ATC CCC CAC CCT GAA TAC CCG AGA AGC Q V R E V V H S I P H P E Y R R S

CCC ACC CAC CTG M C CAC CAC CAT CAC ATC ATG CTT CTG GAG CTG CAC TCC P T H L N H D H E I I M L L E L Q S

CCG GTC CAC CTC ACA GGC TAC ATC CAA ACC CTG CCC CTT TC( CAC AAC AAC CGC P V Q L T G Y I Q T L P L S H N N R

CTA ACC CCT GGC ACC ACC TGT CGG CTG TCT GCC TGG GGC ACC ACC ACC ACC L T P G T T C R V S G W G T T T S

CCC CAG Gsatgciic.. . intron 3.. ... tcccc %TG AAT TAC CCC AAA ACT CTA CAA TCT GCC p Q V N Y P K T L Q C A

AAC ATC CAA CTT CGC TCA GAT GAG GAG TGT CGT CAA GTC TAC CCA GGA AAG N I Q L R S D E E C R Q V Y P G K

ATC ACT GAC M C ATG TTG TGT CCC CGC ACA AAA GAG GGT CGC AAA GAC TCC I T D N M L C A G T K E G G K D S

TGT GAG gtatgca.. . intron 4. .... aactcag CGT GAC TCT GGC CCC CTG GTC TGT AAC c E G D R E G P t v

C N

AGA ACA CTG TAT GGC ATC GTC TCC TGG GGA GAC TTC CCA TGT GGG CAA CCT R T L Y G I V S W G D F P C G Q P

CAC CGC CCT GGT GTC TAC ACC CGT GTC TCA AGA TAC GTC CTC TGG ATC CGT D R P G V Y T R V S R Y V L W I R

GAA ACA ATC CGA AAA TAT GAA ACC CAC CAC CAA A M TCG TTG AAG CGC CCA E T C R K Y E T Q Q Q K W L K G P

C M @ AAGTTGAGAAATGTACCGGCTTCCATCCTGTCACCATOAC Q

A T ~ G T C T G ~ A G C C ~ C T C T G C T C ~ A T T C C C A G T G ~ ~ A G ~ ~ A G T G A T C C A T G T C ~ G A A A A A T ( j ( , T C A A T C T C A G C T A A C A T T C C A T G m c A G ~ ~ A ~ c A G ~ A ~ ~ c A G m TGC AGTCTCCC A G A T G T T G C A T C C C T G A A A C A T C T C C A

Figure 5. Genomic orgmization and partial genomic sequence of the KLK-LA gene. Intmnic sequences are not shown except for the splice junction areas. Introns are shown with lower case lettea and exons with capital letters. For full sequence, see GenBank accession # AH35024 The start and stop codons are encircled and the exon -intron junctions are underlined. The translated amino acids of the coding region are shown undemeath by a single letter abbreviation. The catalytic residues are boxed. The putative polyadenylation signal is underlined.

Genbank submission # AF13SON). The clone from the Geman Genome Project utilizes

another splice donor site that is located further downstream, inside intron 3, and ends up

with a poly A tail without having a fourth or fifth exon. The same stop codon (position

5505) will be in-frarne, and therefore, a truncated protein product is predicted to be

formed (Figure 2).

In order to obtain the î'end of the gene, a 3'RACE reaction was perfomed, and

an additional 375 bp fragment of 3' untranslated region, downstrem from PCR primer

IA-RI, was obtained. This fragment was further confimed to be present in al1 tissues

tested, by perfonning a PCR naction using primers L4-FI and LA-R3 (Table 1 & Figure

2). This fragment ends with a putative polyadenylation signal variant (TATAAA).

Structural characterization of the KLK-L4 gene and its protein product

The classic fom of the KLK-LA gene is presented in Figure 4. KLK-LA is

fomed of five coding exons and four intervening introns, spanning an area of 8,905 bp of

genomic sequence on chromosome 19q13.3-q13.4. The Iengths of the coding regions are

52, 187, 269,137 and 189 bp, respectively. The predicted protein coding region of the

gene is fonned of 83 L bp, encoding a deduced 277-aminoacid protein with a predicted

molecular mass of 30.6 kDa (Figure 5). The introdexon splice sites (mGT.. ..AGm,

w here m is any base) and their fianking sequences are in agreement with the consensus

splice site sequence. A potential translation initiation codon is present at position 45 of

the predicted first exon (numbers refer to Our GenBank submission # AF135024). The

cDNA extends at least 382 bp further downstream from the stop codon and a putative

pol yaden ylation signal (TATAAA) is present at the end of this region (Figure 5).

Hydrophobicity analysis revealed that the amino-terminal region is quite -

hydrophobie (Figure 6), consistent with the possibility that this region may harbor a

signal sequence, analogous to other serine proteases. Figure 6 also shows the presence of

several evenly distributed hydrophobic regions throughout the KLK-L4 polypeptide,

which are consistent with a globular protein, similar to other serine proteases (Liu et al.,

1996). Computer analysis of the amino acid sequence of KLK-L.4 predicted a cleavage

site between aminoacids 20 and 21 (GVS-QE). Sequence homology with other senne

proteases (Figure 7) predicted another potential cleavage site (L~S") in close proximity.

Most other kallikreins are activated by cleavage after arginine or lysine [data not shown].

Thus, although the protein product has not as yet k e n directly characterized, it is very

likely to be a secreted protein. The dotted region in Figure 7 indicates an 1 1-arninoacid

lwp characteristic of the classical kallikreins (PSA, KLKI, and KLK2) which is not

found in KLK-IA or other members of the kallikrein multi-gene family (Little et al.,

1997; Liu et al., 1996; Yoshida et al., 1998; Yousef 1999; Yousef et al., 1999).

Sequence analysis of eukaryotic serine proteases indicates the presence of twenty

nine invariant amino acids (Dayhoff 1976) . Twenty eight of them are conserved in the

KLK-L4 protein and the remaining arnino acid (QI'* instead of P) is not conserved among

al1 other kallikreins (Figure 7). Ten cysteine residues are present in the putative mature

KLK-L4 protein. These are conserved in ail the senne pmteases that are aligned in

Figure 7, and would be expected to form disulphide bridges.

The presence of aspartate @) in position 239 suggests that KLK-IA will possess a

trypsin-like cleavage pattern, similar to most of the other kallikreins (e.g., KLKl, K m ,

TLSP, neuropsin, zyme, prostase, and EMSP) but different from PSA which has a

Figure 6. Plot of hydrophobicity and hydrophilicity of the KLK-U protein, as compared with the glandular kallikrein gene 2 (KLK2). Note the hydrophobie region at the amino terminus, suggesting presence of a signal peptide.

7 - WAX-LSELIQPLPL ERDeSANP-TÇCHIL -A#;--DFm 1 QCAYIKL- HA-1- D- 8 KU-L4 S-LTGYXQTLPL SlMURLTWIWRvS CWOrrPSeQVNYPKT L4CANIQtRSDEDCR QWPCKITDPiSMLCAG 9 TLSP SRIS-IIwAVRPLTL S!SRCVTAG-TSCLIS GWCSrçSWLRLm tRCANITIIMQrmE NAYK;NITDmKAS VQ- 10 nauropsin DPAÇ-LGSKVKPISL mm-QKCryS C;WCRPrçFRR4FPDi' CXAEMUFPgKKCE DA-1- SSKG-m 11 NEÇ1 W - G - L PYRCAQK-WCOVA Gt'GXT- LXSSITILSPWCE -CAG LDRG-QDPCQSO#P

8 a . 8 - a 8 8 8 - O . 271- 285 286 300 301 315 316 330 331

1 p e t i m e KAEiCCQVCN#NYTN ------------ 254 2 KWt-L2 w r & L V S W G bYeCARRIR#rVYTN LCKFTKWIQFMQAN S----------- 293

Figure 7. Alignment of the deduced amino acid sequence of KLK-L4 with membea of the kallikrein rnulti-gene family. Genes are (from top to bottom, and in brackets are the GenBank accession #): KLK- LI/prostase ( A m 2 158 1 ), enamel rnatrix senne proteinase 1 (EMSP) (NP-004908). KLK-L2 (AF135028), PSA (P07288), KLIU (P2OlS l), KLKI (NP(NPûû2248), trypsinogen (W7477), zyme (Q92876), KLK-LQ (AF135024), trypsin-like senne protease (TLSP) (BAA334W), KLK-U (AFl35026),neuropsin (BAA28673), and the normal epithelial cell-specific 1 gene (NESl) (043240). Dashes represent gaps to bring the sequences to better alignment. The residues of the catalytic triad are typed in bold, and conserved motifs aound them are highlighted in grey.The 29 invariant senne protease residues are denoted by ( *), and the cysteine residues by (+). The predicted cleavage sites are indicated by (4). The dotted ana represents the kallikrein loop sequence. The trypsin-like cleavage pattern of KLK-LA with the D residue, is indicated by (O).

serine (S) residue in the corresponding position, and is known to have ch ymotrypsin li ke - . - * - - - & * .- : - - - L.

activity (Figure 7) (Clements 1997).

Mapping and chromosomal localization of the KLK-IA gene

Alignment of the KLK-LA gene and the sequences of other known kallikrein

genes within the 300 Kb area of interest [the human kallikrein gene family locus],

enabled us to precisely localize al1 known genes and to determine the direction of

transcription, as shown by the arrows in Figure 8. The PSA gene lies between KLKl and

KLKZ genes and is separated by 13,3 19 base pairs (bp) from KLIU, and both genes are

transcribed in the same direction (centromere to telomere). Al1 other kallikrein-like

genes are transcribed in the opposite direction. KLK-LA is 13 kb centromeric from KLK-

L6 [GenBank accession # AF1612211, and 2 1 kb more telomeric to KLK-L5 (GenBank

accession #AF 135025).

Homology with the kallikrein multi-gene family

Alignment of the arnino acid sequence of the KLK-L4 protein (long form) against

the GenBank database and the known kallikreins, using the BLAST algorithm (Altschul

et al.. 1997), indicated that IUK-LA has 5 1 % amino acid sequence identity with the

TLSP and zyme genes, 49% identity with KLK-L2 and 47% and 45% identity with PSA

and KLK2 genes, respectively. Multiple dignment study shows that the typical catalytic

triad of senne proteases is conserved in the KLK-Lcl gene (EI'08, D"~, and s") and, as is

the case with al1 other kallikreins, a well conserved peptide motif is found around the

mino acid residues of the catalytic triad (Le. histidine (WLLTAAHC), serine

The New Human Kallikrein Gene Locus (1 Sq13.4,300kb)

Figure 8. An approximate 300 Kb region of almost contiguous genomic sequence amund chromosome l9q 13.3- q 13.4. Genes are represented by horizontal mows denoting the direction of the coding seqwnce. For gene names. see under abbreviations. Names give below each mow represent official nomenclature as defined by the Human Genome Organization (HUGO).

(G-EGGP), and aspartate (DLMLI)) (Figure - 6) (Evans et ai., 1988; Little et al., 1997;

Liu et al., 1996; Schedlich et al., 1987; Yoshida et al., 1998; Yousef 1999). In addition,

several other residues were found to be fully or partially conserved among the human

kalliknin gene family, as further shown in Figure 6. To predict the phylogenetic

nlstedness of the KLK-LA gene with other senne proteases, the amino acid sequences of

the kallikrein genes were aligned together using the "Clustal X" multiple alignment

program and a distance matnx tree was predicted using the Neighbor-joiningNPGMA

rnethod (Figure 9). Phylogenetic analysis separated the classical kallikreins (KLKI,

KLK2, and PS A) and grouped KLK-L4 with zyme, TLSP, KLK-L3, neuropsin, and

NES 1 genes, consistent with pnviously published studies (Mason et al., 1983; Nelson et

al., 1999) and indicating that this group of genes probably arose from a comrnon

ancestral gene by duplication.

Tissue expression and hormonal replation of the KLK-LA gene

As shown in Figure 2, the KLK-LA gene is pnmarily expressed in mammary

gland, prostate, salivary gland and testis, but, as is the case with other kallikreins, lower

levels of expression are found in many other tissues. In order to venfy the RT-PCR

specificity, the PCR products were cloned and sequenced.

Three steroid hormone receptor-positive breast cancer cell lines were used as a

mode1 (BT-474, T47-D, BT-20) to verify whether the KLK-LA gene is under steroid

hormone ngulation. The BT-20 ce11 line is known to be devoid of steroid-hormone

receptors, and was used as a negative control for the BT-474 and T47-D. PSA was used

as a control gene, known to be up-regulated by androgens. Our prelirninary results

Figure 9. Dendrograrn of the predicted phylogenetic tree for some kallikrein and serine pmtease genes. The neighbor-joininSnrpGMA method wasused to align KLK-LA with other serine proteases and members of the kallikrein gene family. The tree grouped the classical kallihins (KLKI, KLK2, and PSA) together and aligned the KLK-LA gene in one group with zyrne, NESI, neuropsin, KLK-L3, and TLSP. Other senne proteases were aligned in different groups, as shown.

indicate that KLK-LA is up-regulated by progestins and androgens and strongly by P L U A-- %- - -= -..-

estrogens in the breast cancer cell line T47-D (Figure 10A).

Expression of KLK-L4 in breast cancer tissues and cell lines

To characterize the extent and frequency of expression of the KLK-IA gene in

breast tumon. we used cDNA derived from 3 normal and 19 malignant breast tissues and

3 breast cancer cell lines. The data were interpreted by cornparison of band intensities.

Out of the 19 tumors, KLK-LA gene expression was undetectable in 7, lower than normal

tissues in 9, comparable to the normal tissues in 1, and higher than nomal tissues in 2

tumors (data not shown). Without hormonal stimulation, the BT-474 (Fig 10B) lines had

no detectable KLK-L4 mRNA; however T47D (Fig 10C) had mRNA KLK-LA

expression. BT-20 was used a control cell line. In addition, prelirninary results suggest

that this gene is down-regulated in the majority (16119) of breast tumors (data not

shown).

Identification of testis-specific KLK-L4 s~ l i ce va riants

Two EST clones with ~ 9 7 % homology with this new gene were identified

(Genbank accessions #fi846771 and AI02101); both of them were cloned from the

testis. These EST clones were obtained and inserts were sequenced from both directions.

Sequences were then compared with the known genomic sequence of the region (see our

Genbank accession #AF135024) and selection of the intronlexon splice sites were made

according to EST sequences. Introdexon splice sites conformed to the consensus

sequences (AG.. .GT).

T47-D PSA

BT-474 PSA

BT20 PSA

Figure 10. Hormonal regulation of the KLK-Lc) gene in the breast carcinoma ce11 lines. Steroids wen added at lu8 M final concentrations. Actin (not regulated by steroid hormones}, and PSA (upregulated by androgens and progestins) are control genes. KLK- LA is up-regulated by androgens and progestins and to a lesser extent by estrogens. H20 was used to check for PCR specificity in al1 PCR reactions. For more details, see text. M.= marker, 1. alcohol, 2. estradiol, 3. norgestrel, 4. dihydrotestosterone, 5. aldosterone, 6. dexamethasone, -ve. negative control.

- - In our previous study, we obtained - - - indications that there may be vhous splice

variants of the KLK-LA gene, since 1) multiple PCR bands were observed with primer~

LA-Xl and LA-R1, which arnplify the whole coding sequence of KLK-IA and 2) we have

identified two ESTs with some exons which were different from those of our previously

reported gene. Preliminary information revealed that at least part of this alternative

splicing occurred between exons 1 and 2 of the classic KLK-LA form. We have therefore

designed primers LA-D and IA-E, binding to this region (please refer to Figure 1 for areas

of primer binding). Screening of cDNAs from 26 different human tissues by RT-PCR,

using gene-specific primea LA-D or LA-E (forward) and LA-Rl (reverse), was then

perfonned. Only testis cDNA produced severai PCR bands (data not shown). Sequencing

of these bands indicated the presence of four mRNA variants: two of these were

characterized from LA-DU-RI screening and two from L4-EU-RI screening (Figure

1). In cornparison to the classic KLK-L4 form, the L4-D and LA-E variants contain

additional exons that we have tentatively named exon D and E, respectively, that are not

present in any of the previously reported splice variants. We also noted (Figure 1) that

exon 3 of the classic form was spliced differently in these LA-D and L4-E splice variants.

Exon 3 for each of the splice variants was either shorter or longer than exon 3 of the

KLK-LA classic form. However, exons 2,4 and 5 of the classic KLK-IA gene, are

identical in al1 splice variants, as shown in Figure 1. We decided to name the four new

variants UD-S, LAD-L, LAE-S, IAE-L, for IAD short, LAD long, LAE short and L4E

long, respectively. To verify that these splice variants, identified by RT-PCR, can also be

defined precisely on the genornic sequence of KLK-LA, the BLASTN algorithm (Altschul

et al., 1-7) w ~ u s e d . Exon identif&-on was-further confimed based on the consensus

exon-intron splice sites.

In order to precisely chmcterize the long KM-LA variant that was identified

previously in our initial study, we designed a reverse primer binding on exon 2 of the

classic form (LA-R2) (Figure 1). Scnening of the testis cDNA with LA-XI and LA-R2

primers revealed the presence of exon D, but not exon E, in this variant. In addition, exon

3 of this variant was shorter than that of the KLK-LA classic fom (Figure 1). Thus, the

structure of the testicular long fom, which was observed previously, has now ken

precisely defined. We have previously observed three KM-LA variants: the long variant

(described above), the classic form, and the short variant, which is spliced earlier in exon

3 and has a putative stop codon at exon 4 (see Figure 1). The short variant does not

contain exons D or E.

Translation of the mRNA variants in al1 possible reading frames indicated that

there is one common reading frame that is predicted to produce a polypeptide for LAD-S,

LAD-L and the KLK-Lcl long fom (Figure 11 A), and there is a separate common reading

frarne for the L4E-S and L4E-L variants (Figure 11B). The other variant (short KLK-LA

form) has a different reading frame sirnilar to the classic form of KLK-LA. and is

predicted to encode for a truncated protein. Possible start positions for the IAE-S and

L4E-L variants were at nucleotide 564 or 644 of the genomic sequence. However, the

likely statt codon would be at position 644 since there is a stop codon after 13 amino

acids for the putative protein encoded by the reading frame that starts at position 564. The

cDNAs extend further downstream from the stop codon and a putative polyadenylation

signal (TATAAA) is present in exon 5. We did not observe a catal ytic triad or exon

Figure 11A: Testis, Long fom and LAD-S and LAD-L protein translation: ---. - - - - - - -- - - - . .

TCAG GCCCCGCCCG C C C K I C C ~ C C ~ C G A T CCCGGGCC ATG TGG CCC CTG GCC CTA GTG ATC M W P L A L V I

GCC TCC CTG ACC TTG GCC TTG TCA GGA G gaag.. . . . . . . . . . . . . . . . intron 1.. . . . . . . . . . . . . . .ttacc% A S L T L A L S G

GT GAA GAC TTC AAG GAG CCA AGC CAG GGG CCT AGA GAT GTA ATG GCG GCT G E D F K E P S Q G P R D V M A A

TCC CGA CCA GAG GCC TTC CCA AAG GGC TTG ACC CTT GAG CCA AGA CCT GAA S R P E A F P K G L T L E P R P E

AAA GGA GGG ATC TGT GGG TGC CTG GCA CCT GGC ACC ATC CTT GGC K G G 1 C G C L A P G T I L G

CTG AAG GTG GGG TGG CTT TTC TCC TCT GGC GAC ACT CCC TGG ATT CAT GCC L K V G W L F S S G D T P W I H A

CGT GCC ACT CCT GAG TGC CAC ACC CTA GGC TAG GAGACCCACACGCTFCGC R A T P E C H T L G *

CTTGTGGAGTCCTCAACAACCTGGCGAG

C A G C T C A G A ~ ~ ~ C f f G G G ~ A C ~ G G T l T C C C G G C l Y i A G G I T G G C C c r C C G A C C C CAGACCCITCAC~CCAAAATACCCIY:GCAGCAGCCCC1Y3CCGCGTIY3MGGCrrCCTG XC~C~CK;GAAAGC~GAMGAC ATGGGTT CGCGTCCTGA CGCTGCCGCT TTGAGCCAGT AGCCTAGCAG CTGCTITGTG CCTAAATTGT TTTCATCTGG AAA ATG GGC TTA ATC TAT

M G L I Y

AAG TGC TTA CCA GAG AAG GTC ACT GTG AAT ATT GAA ACG AG~aatg ... intron 1.. .accs K C L P E K V T V N I E T

G TGT CTC CCA GGA GTC TTC CAA GGT TCT CAA CAC CAA TGG GAC CAG TGG GTT TCT R C L P G V F Q G S Q H Q W D Q W V S

CCC AGG TGG CTA CAC CTG CTT CCC CCA CTC TCA GCC CTG GCA GGC TGC CCT ACT AGT P R W L H L L P P L S A L A G C P T S

GCA AGG GCG GCT ACT CTG TGG GGG AGT CCT GGT CCA CCC CAA ATG GGT CCT CAC A R A A T L W G S P G P P Q M G P H

TGC CGC AC A CTG TCT AAA GGA gatg tgg . . . . . . . . . intron 2.. . . . . .. tgcaccca CS

C R T L S K G

GGG GCT CAA AGT TTA CCT AGG CAA GCA CGC CCT AGG GCG TGT GGA AGC TGG TGA G A Q S L P R Q A R P R A C G S W *

GCAGGTGAGGGAAGmGTCCACTCTATCCCCCACCCTGAATACCGGAGAAGCCC CACCCACCTGAACCACGACCATGACATCATGCTTCTGGAGCTGCAGTCCCCGGT CCA~CACAGGCTACATCCAAACCCTGCCCCTTTCCCACAACAACCGCCTAACCCC TGGCACCACCTGTCGGGTGTCTGGCTGGGGCACCACCACCAGCCCCCAGGGTAT GCACCCACACAGGTGGCCTGAGGCCCCATAGGAGTGGCTGGGGAAACAGGGGCA GAGATGGGAGGGAAGGTCTGAG

Figure Il: Gmomic organizatim and partid genomic sequenns of the KLK-LA variants MD-S, MD-L and long KLK-LA fom. Intronic sequences are not shown except for the splice junction areas. Introns are shown with lowercase letters. and exons with capital letters. The stop codon is indicated by an asterisk, the start codon (ATG) is in bold and the exon-intron junctions are underlined. The translated amino acids of the putative coding region are shown underneath by a single letter abbreviation. The sequence shown starts at exon 1 and ends at exon D; the sequences of exons 2,3,4 and 5 are not shown.

--.-- - -- L A % L A-

structiiFë ~haractëristic of conserveir liàniEëiKgeneX (Yousef and Diamandis ZOO I) in the

predicted translated proteins. Protein h ydrophobicity was studied using the Baylor

College of Medicine search launcher prograrns at

htt~://dot.imgen.bcm.tmc.edu:933 llsea-search/stnic-predic~html. We found the N-

terminus of the predicted proteins to be hydrophobic (Figure 12), suggesting that if the

M A S are translated, the proteins could be secreted, similarly to al1 kallikrein proteins.

Screeninn for KLK-L4 variants in testicular cancer tissues

Total RNA from matched tumor and adjacent morphologically normal tissue

sarnples from patients with various testicular cancers was screened by RT-PCR for the

newly characterized variant mRNAs. The samples were first screened wi th LA-XI and

LA-RI, then with L4-D and LA-Rl, and finally with L4-E and LA-RI primers. Primers

LA-X1 and LA-RI amplified the short, classic and long foms of KLK-L4 (Figure 13).

while the LA9 and L4-E pnmers detected the L4D & IAE splice variants (Figure 14),

respectively. Of the 10 pairs of cDNAs tested, the classic KLK-LA fom was low or

undetectable in the tumor tissues in four pairs, increased in one tumor tissue, and

remained unchangecl in five pairs (Figure 13). Thne normal specimens expressed the

long KLK-LA splice variant (NI, N4, NS). However, their adjacent cancerous tissues did

not, suggesting that this splice variant may be differentially expressed in certain types of

testicular cancer (Figure 13). Screening of the tissue samples with the LA-DI L4-RI and

L4-EIL4-RI primers indicated the presence of al1 four splice variants (LADS, IAD-L,

L4E-S, UE-L) in four out of the ten normal tissues (Ni, N5, N9, Nio ) (Figure 14).

Figure 12a: LAD-L, L4D-S. long KLK-L4

Figure 12: Plot of hydmphobicity and hydrophilicity of the 2 putative proteins, encoded by the rnRNA variants MD-S, LAD-L and KLK-IA long form (left; 93 arnino acids) and IAE-S and ME-L (right; 98 amino acids). Negative numbers indicate hydrophobicity and positive ones hydrophilicity. N denotes the N-temiinus and C denotes the C-terminus of the protein. Note the hydrophobic region at the arnino terminus, suggesting the presence of a signal peptide.

Figure 13: Expression of the KLK-LA variants KLK-LA long (upper band; clearly shown in lanes NI, N4 and Ns), KLK-[A classic form (Piast intense band; clearly shown in lane N3) and WK-L4 short form (lower band, clearly shown in lane Ns), in matched, paired cancerous and non- cancerous testicular tissues, using primers LA-XI and L4-RI (primer location is shown in Figure 1). N= normal tissue, C=cancerous tissu?, M= marker (1 kb molecular weight ladder), -ve= negative control. Numben shown on the left side correspond to the marker size. Expected bwd sizes are: 877bp for the KLK-LA long variant, 819bp for the KLK-LA classic fom, and 608bp for the KLK-LA short variant.

Figure 14: Expression of the KLK-LA mRNA variants. LAD-S. LAD-L, and long KLK- - - e h ---A

LA fom (panel A); UE-S-, L4E-L (pGëlB); in matctied, paired cancerous and non- cancerous testicular tissues using (A) primers LA-D and U - R I and (B) primers IA-E and L4-R1.The PCR products of MD-S and long KLK-LA foms are identical. N= normal tissue. C= cancerous tissue. M= marker (1 kb molecular weight ladder), -ve= negative contml. Numbers shown on the left side correspond to the marker size. Expected band size for LAD-S is 837bp (shown clearly in lanes Ni, Ns, N9, Nlo; lower band) LAD-L is 1136bp (shown clearly in lanes Ni, N4, Ns. N9, Ni*; middle band), UE-L is 1080bp (shown clearly in lanes Ni, Ns, N9, Nio) and ME-S is 781 bp (shown clearly in lanes Ni. Ns, Nia). In the upper (A) panel we occasionall y saw a PCR band that is larger than UD- L which could not be identified (this band is visible in lanes NI and NI; top band)

Much like Our pnvious observation with the long and classic splice variant~, the four -

variants were present only in the normal tissues; the adjacent tumors did not express any

of these splice variants. These data are summarized in table 3. In general, the variant

mRNAs are detectable in a fraction of testicular normal tissues but not in their adjacent

cancerous counterparts.

We then examined expression of IUK-LA and its variants in three samples of

morphologically nomal testicular parenchyma (adjacent to overt tumors), two with

carcinoma in situ tubules. one without carcinoma in situ (CIS), one Leydig-cell tumor as

well as in two human embryonal carcinoma cell lines. We found that only the normal

parenchyma expressed the long KLK-LA variant (Figure 7) and the L4D-S, LAD-L, LAE-

S, UE-L vanants (data not shown). However, these variants were not present in any of

the tumors.

--E - - - Tabk 3: Detection and relative abundance of KLK-LA mRNA variant forms in ten pairs of n6id ancicancerous testicuiar tissues

Relative mRNA Abundance

Splice mRNA variant ' Classic form

Long form

L4D-S

L4D-L

L4E-S

L4E-L

Short form

Detected in Detected in Approx. Higher in Higher in normal tissue2 cancer Eqwl normal cancer

9 9 5 4 1

3 O - 3 -

4 O - 4

4 O - 4

4 O - 4

4 O - 4 -

2 O - 2

1. For splice variant definition and structure, please see text. 2. We used 10 pairs of nonnal/cancemus tissues from patients with testicular cancer.

Please refer to Material and Methods for more details.

Actin

Figure 15: Expression of the KLK-LA variants KLK-LA long. classic and short form, in testicular tumors and embryonal carcinoma-like (EC-like) ce11 lines. Pnmen used were LA-XI and LA-RI (primer location is shown in Figure 1). For further details see Figure 1. 1. Testicular carcinoma in situ; 2. Leydigcell tumor; 3. Normal parenchyrna; 4. Seminoma; 5. Non-seminoma; 6. Pluripotent EC-like ceIl line NTERA-2; 7. Nullipotent EC-like ce11 line 2102Ep; 8. Negative control. Numben shown on the left side correspond to the marker size. Expected band sizes are mentioned in Figure 1. For discussion see text.

DISCUSSION

1. Genomic Characterization of KLK-L4

Kallikreins are a subgroup of serine proteases traditionally defined by their ability

to release vasoactive peptides (kinins) from kininogens (Clements 1997). However, it is

now recognized that in both humans and rodents, kallikreins exhibit a variety of functions

in diffennt tissues. KLK-L4 is defined as a kallikrein-like gene based on the criteria of

structural homology and chromosomal locaiization (Mason et al., 1983). W i n et al.

proposed that the serine protease genes could be classified into five different groups

according to intmn position ( h i n et al., 1988). The established kallikreins (KLICI,

KLK2, and PSA), trypsinogen and chymotrypsinogen belong to a group that has: (1) an

intron just downstream from the codon for the active site histidine residue, (2) a second

intron downstream from the exon containing the codon for the active site aspartic acid

residue. and (3) a third intron just upstrearn from the exon containing the codon for the

active site senne residue. Figure 10 shows that KLK-LA meets the above mentioned

critena; moreover, is located in close proximity to other kallikrein genes on the

chrornosomal locus L9q 13.3-q 13.4 (Figure 8).

Our preliminary finding that the KLK-LA gene may be down-regulated in a subset

of breast cancers is not surprising. There is now growing evidence that many of the

kallikreins and kallikrein-like genes that are clustered in the sarne chromosomal region

(Figure 7) are related to malignancy. PSA is the best marker for prostate cancer so fa.

(Diamandis 2000). A recent report provided evidence that PSA has antiangiogenic

activity,l,-d that this activity may be related to its function as a senne protease (Fonier et - - - -

al., 1999). This study suggested that other serine proteases. including new members of

the kallikrein multigene farnily of enzymes, should also be evaluated for potential

antiangiogenic activity (Fortier et al., 1999). Recent reports suggest that hK2 (encoded

by the KLK2 gene) could be another useful diagnostic marker for prostate cancer (Black

et al.. 2ûûûb; Stenman 1999). NES 1 appears to be a tumor suppressor gene (Goya1 et al.,

1998). The protease M gene was shown to be differentially expressed in primary breast

and ovarian tumors (Anisowicz et al.. 1996), and the human stratum comeum

chymotryptic enzyme has been shown to be expressed at abnormally high levels in

ovarian cancer (Undenvood et al.. 1999). Another recentl y identified kalli krein-li ke

gene, located close to KLK-L4 and ten tati vel y named tumor-associateci, di fferen tiall y

expressed gene- 14 (TADG14) [an alternatively spliced form of neuropsin, see Figure 71

was found to be overexpressed in about 60% of ovarian cancer tissues (Underwood et al..

1999) . Also, prostase/KLK-L1, another new l y discovered kallikrein-li ke gene, is

speculated to be linked to prostate cancer (Yousef et al., 1999). Thus, extensive new

literature suggests multiple connections of many kallikrein genes to various fotms of

human cancer. This subject has recently been reviewed (Diamandis et al.. 2000b).

The removal of intervening RNA sequences (introns) from the pre-messenger

RNA in eukaryotic nuciei is a major step in the regulation of gene expression (Adams et

al.. 1996). RNA splicing provides a mechanism whereby protein isoform diversity can

be generated and the expression of particular proteins with specialized functions can be

resvicted to certain ce11 or tissue types during development (Adams et al., 1996). RNA is

synthesized by RNA polyrnerase II, called a primary transcript. It undergoes several

processing steps befon becorning a functional product (Lodish et al., 1995). It is now --- 2 &- -

know that after transcription begins, the 5' end of the nascent RNA is capped with 7-

methyl-guanylate. Transcription by RNA polyrnerase II tenninates at any one of the

multiple sites about 0.5-2kb down Stream of the last exon in the transcript. Nascent RNA

and mRNA intermediates first emerge from RNA polyrnerase until they are transported

into the cytoplasm. usually associated with chaperone proteins as a histone complex. 4

These proteins were first characterized as k ing major components of heterogeneous

ri bonucleoprotein partic les (hnRNPs) whic h contain heterogeneous nuclear RNA

(hnRNA). HnRNP proteins interact with different portions of pre-mRNA and contribute

to the structure recognized by RNA-processing factors. RNA splicing occurs via a

ribonucleoprotein complex called a spliceosome. At least one hundred proteins ue

estimated to be involved in RNA splicing. After pre-mRNAs are processed in the

nuclease, each of the resulting mature mRNAs is transported to the cytoplasm. The

translation of mRNAs takes place in the cytoplasm on polyribosomal complexes. The

concentration of rnRNA is a function of both its rate of synthesis. In additional, the

presence of RNA may not correlate with protein synthesis. Hence, our splice variants

reported in this study may not be necessarily related to functional proteins and further

investigation is warranted to examine this further.

The sequence elements in the pre-mRNA at the 5' and 3' splice sites in

metazoans have very loose consensus sequence; only the first and the last two bases

(GT..AG) of the introns are highly conserved (Sambrook et al., 1989). These sequences

cannot be the sole detenninants of splice site selection, since identical, but not ordinarily

active, consensus sequences can be found within both exons and introns of many

- - eukaryotic genes. Other protein factors and sequences downstream of the splice sites are ----- - - ?.

also involved.

The existence of multiple splice forms is frequent among kallikreins. Distinct

RNA species are transcribed from the PSA gene, in addition to the major L .6-kb

transcnpt (Adams et al., 1996). Several distinct PSA transcripts have ken described by

others (Riegman et al., 1989b). Interestingly, one of these clones lacks the 3'

untranslated region and the first 373 nucleotides of the open reading frame, and has an

extended exon that contains a stop codon, a pattern that is comparable with some

alternative forms of the KLK-LA cDNA, as described here (Figure 1). Heuze et al.,

reported the cloning of a full-length cDNA corresponding to a 2.1 kb PSA mRNA. This

fom results from the alternative splicing of intron 4 and lacks the serine residue that is

essential for catalytic activity (Heuze et al., 1999). Also, Reigman et al reported the

identification of two altematively spliced forms of the human glandular kallikrein 2

(KLK2) gene (Riegman et al., 199 la). A novel transcript of the tissue kallikrein gene

(KLKl) was also isolated from the colon (Chen et al., 1994). Interestingly, this transcript

lacks the first two exons of the tissue kallikrein gene, but the last three exons were fully

conserved, a pattern that is similar to our findings with some ESTs containing parts of the

KLK-LA gene (Figure 1). Neuropsin, a recentl y identi fied kallikrein-li ke gene, was

found to have two altematively spliced forms, in addition to the major form (Mitsui et al.,

1999; Underwood et al., 1999). Here, we describe the cloning of the KLK-LA gene and

the identification of a number of alternative mRNA foms. These fonns may result from

alternative splicing (Sambrook et al., 1989), retained intronic segment (Riegman et al.,

1989a), or from the utilization of an alternative transcription initiation site (Chen et al.,

1994). Because the classic form of KLK-L4 and the major alternative splice variant

[short KLK-L4 variant] (Figure 3) have an identical 5' sequence required for translation.

secntion and activation, it is possible to assume that both mRNAs encode for a secreted

protein (Heuze et al., 1999).

In order to investigate the relative predominance of the long KLK-LA and related

forms, cDNA from various tissues was amplified by PCR. Although, in general, it is

dificult to use PCR for quantitative cornparisons between mRNA species, in this

experiment, [mRNAs of comparable sizes, using one set of primers under identical

conditions], such a cornparison is reasonable (Riegman et al.. 1991a). In al1 five normal

tissues examined [breast, prostate, testis, salivary gland and utexus] the classic form of

KLK-LA was the predominant, with minimal level of expression of the short form (Figure

3). The clone from the German Human Genorne project is pndicted to encode for a

inincated protein at the carboxyl terminus. As a result, the serine residue ( s2I8 ),

essential for the catalytic activity, will be lacking.

The presence of alternative1 y spliced forms may be related to malignanc y. Recent

literature suggests that distinct molecular forms of PSA could be expressed differently by

malignant versus benign prostate epithelium (Baffa et al.. 1996). Aberrant PSA mRNA

splicing in benign prostatic hyperplasia, as opposed to prostate cancer, has been described

by others (Henttu et al., 1990). In addition. it has been recentl y postulated that different

prostatic tissues potentially harboring occult cancer could account for the presence of

various forms of PSA (Baffa et al., 1996). Clearl y, the alternative1 y spliced forms of

KLK-LA should be examined and compared in detail, in various normal and malignant

tissues. In the next section we will discuss characterization of five novel, testis-specific -=.-A---- A--- :A -

Although testicular cancer accounts for only 1% of al1 tumon in males, it is the

most comrnon malignancy between 15 to 34 years of age . The incidence of testicular

cancer in the United States has almost doubled since the 1930s and continues to climb,

while more effective treatments have ied to a decline in mortality (Kinkade 1999). Earl y

diagnosis of testicular cancer is important because the doubling time of testicular tumors

is relatively short (10 to 30 days) (Kinkade 1999). Cumnt testicular tumor markers

include human chorionic gonadotropin (hCG), alpha-fetoprotein (AFP), lactate

dehydrogenase (LDH), and placental alkaline phosphatase (PAP) (Gatti and Stephenson

1998); however, sensitivity and specificity of these markers are not hi&. For example,

AFP is not specific for testicular neoplasms, and its high senun concentration has been

documented in lung, gastric, pancreatic and hepatocellular cancers (Gatti and Stephenson

1998). AFP and hCG are currently the most widely applied markers for testicular tumor

staging, despite their relatively low specificity and sensitivity (Dean and Mou1 1998);

however, they are only useful in about half of the genn ce11 tumor cases, because

seminomas usually do not secrete AFP or hCG. Hence, new tumor markers could be

valuable tools for both diagnosing and staging of various types of testicular cancer. As

KU-LA and its newly-identified splice variants appear to be preferentially expressed in

the testis, we investigated their expression in a series of testicular tumors.

Our previously identified kallikrein-like gene, the KLK-LA classic form (official

nomenclature, KLK13), was found to be differentidly expressed in five out of ten

matched (norxnaVtumor) testicular tissues (Table 1). Using expressed sequence tag (EST)

anal ysis, sequencing and PCR, we have c harac terized several novel testis-speci fic KLK- - - A % - - - - - -- - -

LA mRNA variants, tentatively named LAD-S, LAD-L, ME-S, L4E-L and KLK-L4 long

form, and noted the frequent presence of these splice variants in normal tissues but not in

the adjacent tumors (Table 1 and Fig 14). The lack of expression of the --LA splice

variants in testicular genn ce11 tumors provides information conceming the biology of

these tumors. It is now commonly accepted that testicular germ ceIl tumors originate

from a cornmon pre-invasive precursor, the carcinoma-in-situ (CIS) ce11 (Skakkebaek

1972). and that CIS in tum, is derived from gonocytes transformed early in life

(Skakkebaek et al., 1987). It is not yet known whether or not CIS cells express the KLK-

IA splice variants, but it is unlikely, because seminoma has nearly identical phenotypic

features, including expression of the number of markers. Further studies will be necessary

to confimi whether the mRNA variant absence is related to tumor progression or

aggressiveness.

By direct sequencing and alignment strategies, we were able to define precisely

the exon composition for each of the putative splice variants, further confirmed by the

pnsence of AG.. .GT splice sites that are well-conserved arnong vertebrates (Iida 1990).

It is clear that the variants arise from alternative splicing of exon 3 at three different

locations, as well as from two additional exons, D and E (Fig. 1). We did no< investigate

experimentally if the variant mRNAs are mslated but open reading frame anal ysis

indicated that if the mRNAs are translated, they will lead to production of tnincated

proteins of various lengths, as shown in Figure 1. None of the variant mRNAs (with the

exception of the classic KLK-L4 form) encodes for r senne protease with a conserved

catalytic triad (Diamandis et al., 2ûûûb); however, they are predicted to be secreted

proteins due to conservation of their hydrophobie signal peptide sequences (Yousef et al.,

2ûûûa).

Other kallikreins have been also noted to have splice variants of unknown

physiologie devance (Henttu et al., 1990; Heuze et aL, 1999; Liu et al., 1999; Tanaka et

al., 2000). For example, Henttu et al. noted that the amount of aberrant PSA mRNAs

which would produce variant proteins (if translated) was about 18% to 38% (Henttu et

al., 1990). These rnRNA variants were noted to have extended 3' untranslated regions

(UTR), which are speculated to be essential for mRNA stability or targeting of rnRNA

into cells (Stenman 1999). Differences in the transcriptional regulation of the PSA gene

between benign prostatic hyperplasia and prostate carcinomas, resulting in a 3-fold

variation in PSA arnount in carcinoma tissues (vs. normal or benign hyperplasia) have

been reported (Mitsui et al., 1999). A mRNA variant of PSA, which lacked the critical

serine residue necessary for catalytic activity, was noted to produce a protein that would

cross-react with anti-PSA antibodies used in clinical assays (Heuze et al., 1999). Another

PSA mRNA variant was reported to be expressed predominantly in the prostate tissue,

but not in blood (Tanaka et al., 2000). Such a mRNA splice variant, if translated, would

also produce a variant protein that may offer enhanced clinical utility (Tanaka et al.,

2000). Recently, a highly sensitive splice variant-specific RT-PCR assay for human

glandular kallikrein 2 (hK2) was reported, which was able to detect metastatic disease in

men with clinically localized prostate cancer (Slawin et al., 2000). Thus, further

examination of our own reported splice variants is warranted to determine their potential

diagnostic and thetapeutic applicability.

--- - 3 - In summary. we have cloned and characterized five new KLK-LA mRNA

variants, narned LAD-S, LAD-L, L4E-S, LAE-L and KLK-L4 long fom. Unlike their two

other counterparts (the KLK-LA classic and short fom), Our new variants do not seem to

be present in any other tissue except the testis. Our preliminary results suggest that these

variants are not expressed in testicular tumors. More studies will be necessary to

elucidate the biological role of these mRNA variants in the testis.

- - CONCLUSIONS AND FUTURE DIRECTIONS

In surnmary, we have cloned and chmcterized a new gene, kallikrein-like geoe 4,

also known as kallikrein gene 13 (KLK13) and its splice variants. We describe their

genomic structure, chromosornai localization, and tissue expression pattern. Our

preliminary results suggest that this gene is down-regulated in breast cancer, and also,

that the gene is hormonally regulated in the breast cancer ce11 line BT-474. Its splice

variants are differentially expressed in testicular cancer, suggesting that their functions

Vary within normal and cancerous testicular tissue. This gene (and its splice variants)

should be examined as a potential marker for cancer diagnosis, prognosis. and

monitoring. My development of recombinant protein and monoclonal antibodies has been

made and the expectation for project is to develop a highly specific sandwich

immunoassay. Such assay would be able to detect the presence of KLK13 in serum and

various tissue extracts. Furthemore, indications that this recombinant protein is

enzymatically-active will lead up to substrate analysis and thus elucidate the possible

physiological functions of KLK13. Understanding the function of the gene and its splice

variants will indicate if this gene is useful as a therapeutic target.

Abrahamsson, P. A., Lilja, H., Falkmer, S. and Wadstrom, L. B. (1988). Immunohistochemical distribution of the t h e predorninant secretory proteins in the parenc h p a of h yperplastic and neoplastic prostate glands. Prostate 12: 39-46.

Adams, M. D., Rudner, D. 2. and Rio, D. C. (1996). Biochernistry and regulation of pre-mRNA splicing. Curr Opin Ce11 Bi01 8: 331-9.

Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search prograrns. Nucleic Acids Res 25: 3389-402.

Anisowicz, A., Sotiropoulou, O., Stenman, G., Mok, S. C. and Sager, R. (1996). A novel protease homolog differentially expressed in breast and ovarian cancer. Mol Med 2: 624-36.

Baffa, R., Moreno, J. G., Monne, M., Veronese, M. L. and Gomella, L. G. (1996). A comparative analysis of prostate-specific antigen gene sequence in benign and malignant prostate tissue. Urology 47: 795-800.

Balbay, M. D., Pettaway, C. A., Kuniyasu, H., Inoue, K., Ramirez, E., Li, E., Fidler, 1. J. and Dinney, C. P. (1999). Highly metastatic hurnan prostate cancer growing within the prostate of athymic mice overexpresses vascular endothelial growth factor. Clin Cancer Res 5: 783-9.

Bhoola, K. D., Figueroa, C. D. and Worthy, K. (1992). Bioregulation of kinins: kallikreins, kininogens, and kininases. Phannacol Rev 44: 1-80.

Black, M. H. and Diamandis, E. P. (2000). The diagnostic and prognostic utility of prostate-specific antigen for diseases of the breast. Breast Cancer Res Treat 59: 1-14.

Black, M. H., Giai, M., Ponzone, R., Sismondi, P., Yu, H. and Diamandis, E. P. ( 2 0 a ) . Serum total and free prostate-specific antigen for brrast cancer diagnosis in womeri [In Process Citation]. Clin Cancer Res 6: 467-73.

Black, M. H., Magklara, A., Obiezu, C., Levesque, M. A., Sutherland, D. J., Tindall, D. J., Young, C. Y., Sauter, E. R. and Diamandis, E. P. (2ûûûb). Expression of a prostate-associated protein, human glandular kallikrein (hm), in breast tumours and in normal breast secretions. Br J Cancer 82: 36 1-7.

Blaukat, A., Herzer, K., Schroeder, C., Bachmann, M., Nash, N. and Muller- Esterl, W. (1999). Overexpression and functional characterization of kinin receptors reveal subtype-specific phosphorylation. Biochemistry 38: 1300-9.

- A

Catdona, W. 1. (S99t9. Clhicd utfiity ofmeisüïëments of free and total prostate- specific antigen (PSA): a review. Prostate Suppl7: 64-9.

Catalona, W. J., Partin, A. W., Slawin, K. M., Brawer, M. K., Fianigan, R. C., Patel, A., Richie, J. P., deKemion, J. B., Walsh, P. C., Scardino, P. T. et al., (1998). Use of the percentage of free prostate-speci fic antigen to enhance di fferentiation of prostate cancer from benign prostatic disease: a prospective multicenter clinical trial. Jama 279: 1 542-7.

Chen, L. M., Murray, S. R., Chai, K. X., Chao, L. and Chao, J. (1994). Molecular cloning and characterization of a novel kallikrein transcript in colon and its distribution in human tissues. Braz J Med Bi01 Res 27: 1829-38.

Chen, Y. P., Chao, J. and Chao, L. (1988). Molecular cloning and characterization of two rat nnal kalli krein genes. Biochemistry 27: 7 1 89-96.

Clements, J. 1997. The molecular biology of the kallikreins and their roles in inflammation. In The Kinin System (ed. S . Farmer), pp. 71-97. Academic Press, New York.

Clements, J. and Mukhtar, A. (1994). Glandular kallikreins and prostate-specific antigen are expressed in the human endometrium. J Clin Endocrinul Metab 78: 1536-9.

Clements, J., Mukhtar, A., Ehrlich, A. and Yap, B. (1994). Glandular kallikrein gene expression in the human uterus. Broz J Med Bi01 Res 27: 185563.

Clements, J. A. (1994). The human kallikrein gene family: a diversity of expression and function. Mol Ce11 Endocrinol99: C 1-6.

Clements, J. A., Matheson, B. A., Wines, D. R., Brady, J. M., MacDonald, R. J. and Funder, J. W. (1988). Androgen dependence of specific kallikrein gene farnily members expressed in rat prostate. J Bi01 Chem 263: 16132-7.

Clements, J. A., Mukhtar, A., Verity, K., Pullar, M., McNeill, P., Cummins, J. and Fuller, P. 1. (1996). Kallikrein gene expression in human pituitary tissues. Clin Endocrinul (O@ 44: 223-3 1.

Cleutjens, K. B., van Eekelen, C. C., van der Korput, H. A., Brinkmann, A. O. and Trapman, J. (1996). Two androgen response regions cooperate in steroid hormone regulated activity of the prostate-specific antigen promoter. J Bi01 Chem 271: 6379-88.

Cohen, P., Peehl, D. M., Lamson, G. and Rosenfeld, R. G. (1991). Insulin-like growth factors (IGFs), IGF receptors, and IGF-binding proteins in primary cultures of prostate epithelial cells. J Clin Endocnnol Metab 73: 40 1-7.

Cnighton, T. E. 1993. Proteins: Structures and Molecular Properties, 2nd. W.H. -- - A - - -- -

Freeman and Compaiy, New York.

Crocker, P. R., Vinson, M., Kelm, S. and Drickamer, K. (1999). Molecular analysis of sialoside binding to sialoadhesin by NMR and site-directed mutagenesis. Biochem J 341: 355-61.

Dayhoff, M. 0. (1976). The origin and evolution of protein superfamilies. Fed Proc 35: 2 132-8.

Dayhoff, M. 0. (1978). Atlas of protein sequence and structure. Nat1 Biomed Res Found 5: 79-8 1.

Dean, R. C. and Moul, J. W. (1998). New tumor markers of testis cancer. Urol Clin North Am 25: 365-73.

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: 102-6.

Diamandis, E. P. (2000). Prostate-speci fic antigen: a cancer fighter and a valuable messenger? Clin Chem 46: 896-900.

Diamandis, E. P., Helle, S. L, Yu, H., Melegos, D. N.9 Lundgren, S. and Lonning, P. E. (1999). Rognostic value of plasma prostate specific antigen after megestml acetate matment in patients with metastatic breast carcinoma. Cancer 85: 891-8.

Diamandis, E. P., Yousef, G. M., Clements, J., Ashworth, L. K., Yoshida, S., Egelnid, T., Nelson, P. S., Shiosaka, S., Little, S., Lilja, H e T Stenman, U. H., Rittenhouse, H. G. and Wain, H. (2ûûûa). New nomenclature for the human tissue kallihin gene farnily [In Process Citation]. Clin Chem 46: 1855-8.

Diamandis, E. P., Yousef, G. M., Luo, L. Y., Magklara, A. and Obiezu, C. V. (2ûWb). Themw human kallikrein gene family: implications in carcinogenesis. Trends Endocrinol Metab 1 1: 54-60.

Diamandis, E. P. and Yu, H. (1995). New biological functions of prostate-specific antigen? [editorial]. J Clin Endocrinol Metub 80: 15 15-7.

Dlamini, Z., Raidou, D. and Bhoola, K. (1999). Visualisation of tissue kallikrein and kinin receptors in oesophageal carcinoma. Immunophannacology 43: 303- 10.

Drinkwater, C. C., Evans, B. A. and Richards, R. 1. (1988). Kallikreins, kinins and growth factor biosynthesis. Trends Biochem Sci 13: 169-72.

-- Evans, B. A., Drinkwater, C. C. and Richards, R. 1. (1987). Mouse glandular EdLXiri ~nës.~StniCture' andpa.isequenceanaTySis o f t k kallikein gene locus. J Bi01 Chem 262: 8027-34.

Evans, B. A., Yun, Z. X., Close, J. A., Tregear, G. W., Kitamura, N., Nakanishi, S., Callen, D. F., Baker, E., Hyland, V. J., Sutherland, O. R. and et al. (1988). Structure and chromosomal localization of the human rend kallikrein gene. Biochemistry 27: 3 124- 9.

Foekens, J. A., Diamandis, E. P., Yu, H., Look, M. P., Meijer-van Gelder, M. E., van Putten, W. L. and Klijn, J. G. (1999). Expression of prostate-specific antigen (PSA) correlates with poor response to tamoxifen therapy in recurrent breast cancer. Br J Cancer 79: 888-94.

Fortier, A. H., Nelson, B. J., Grella, D. K. and Holaday, J. W. (1999). Antiangiogenic activity of prostate-specific antigen. J Nat1 Cancer Inst 91: 1635-40.

Foussias, GeT Yousef, G. M. and Diamandis, E. P. (2000). Identification and molecular characterization of a novel member of the siglec family (SIGLECI)). Genomics 67: 171-8.

Gatti, J. M. and Stephenson, R. A. (1998). Staging of testis cancer. Combining serum markers, histologic parameters, and radiographie imaging. Ur01 Clin North Am 25: 397-403.

Goyal, J., Smith, K. M., Cowan, J. M., Wazer, D. E., Lee, S. W. and Band, V. (1998). The role for NES I senne protease as a novel tumor suppressor. Cancer Res 58: 4782-6.

Hakalahti, L., Vihko, P., Henttu, P., Autio-Hannainen, H., Soini, Y. and Vihko, R. (1993). Evaluation of PAP and PSA gene expression in prostatic hyperplasia and prostatic carcinoma using northem-blot analyses, in situ h ybridization and immunohistochemical stainings with monoclonal and bispecific antibodies. Int J Cancer 55: 590-7.

Henttu, P., Lukkarinen, 0. and Vihko, P. (1990). Expression of the gene coding for hurnan prostate-specific antigen and related hGK-1 in benign and malignant tumors of the human prostate. Int J Cancer 45: 654-60.

Hess, J. F. 1997. Molecular pharmacology of kinin receptors. In The Kinin System (ed. S. O. Farmer), pp. 45-55. Academic Ress, London.

Heuze, N., Olayat, S., Gutman, N., Zani, M. L. and Couriy, Y. (1999). Molecular cloning and expression of an alternative hKLK3 transcript coding for a variant protein of prostate-specific antigen. Cancer Res 59: 2820-4.

- - Horton, H. R., Moran, L. A., Ochs, R. S., Rawn, J. D. and Scrimgeour, K. G. 1996. ~&ci'les of ~iochemistry, 2ndTPientice-Hdl, Inc., Upper Saddle River, NJ.

Horwitz, D., Margolius, H. S. and Keiser, H. R. (1978). Effects of dietary potassium and race on urinary excretion of kallikrein and aldosterone in man. J Clin Endocrinol Metab 47: 296-9.

Hsieh, M. L., Charlesworth, M. C., Goodmanson, M., Zhang, S., Seay, T., Klee, G. G., Tindall, D. J. and Young, C. Y. (1997). Expression of human prostate-specific glandular kallikrein protein (hK2) in the breast cancer cell line T47-D. Cancer Res 57: 265 1-6.

Iida Y. (1990). Quantification analysis of 5'-splice signal sequences in mRNA precursors. Mutations in 5'-splice signal sequence of human beta-globin gene and beta- thalassemia. J Theor Bi01 145: 523-33.

Invin, D. M., Robertson, K. A. and MacGiliivray, R. T. (1988). Structure and evolution of the bovine prothrombin gene. J Mol Bi01 200: 31-45.

Jones, T. H., Figueroa, C. D., Smith. C., Cullen, D. R. and Bhoola, K. D. (1990). Charactenzation of a tissue kallikrein in human prolactin-secreting adenornas. J Endocth01 124: 327-3 1.

Killian, C. S., C o d , D. A., Kawinski, E. and Constantine, R. 1. (1993). Mi togenic response of osteoblast ce1 1s to prostate-speci fic an tigen suggests an activation of latent TGF-betr and a proteolytic modulation of ce11 adhesion receptors. Biochem Biophys Res Commun 192: 940-7.

Kinkade, S. (1999). Testicular cancer. Am Fam Physician 59: 253944,2549-50.

Koshikawa, N., Yasumitsu, H., Umeda, M. and Miyazaki, K. (1992). Multiple secntion of matrix serine proteinases by hurnan gastric carcinoma ce11 lines. Cancer Res 52: 5046-53.

Kumar, A., Mikolajczyk, S. D., Goel, A. S., Millar, L. S. and Saedi, M. S. (1997). Expression of pro form of prostate-specific antigen by mammalian cells and its conversion to mature, active fonn by human ka1 likrein 2. Cancer Res 57: 3 1 1 1-4.

Lai, L. C., Erbas, H., Lennard, T. W. and Peaston, R. T. (1996). Prostate-specific antigen in breast cyst fluid: possible role of prostate-specific antigen in hormone- dependent breast cancer. Int J Cancer 66: 743-6.

Lennon, G., Auffray, C., Polymeropoulos, M. and Soares, M. B. (1996). The I.M.A.G.E. Consortium: an integrated molecular anaiysis of genomes and their expression. Genomics 33: 1 5 1-2.

- - Lilja, H. (1993). Structure, function and regulation of the enzyme activity of

pr6stateapiicific antigen. WortdJ m o l IZ: 18ü-Tn.

Lilja, H. (1997). Prostate-specific antigen: molecular foms and the human kallikrein gene famil y. Br J Ur0179 Suppl 1: 44-8.

Little, S. P., Dixon, E. P., Noms, F., Buckley, W., Becker, G. W., Johnson, M., Dobbins, J. R., Wyrick. T., Miller, J. R., MacKellar, W. et al., (1997). Zyme, a novel and potentially amyloidogenic enzyme cDNA isolated frorn Alzheimer's disease brain. J Biol Chem 272: 25 135-42.

Liu, X. F., Essand, M., Vasmatzis, G., Lee, B. and Pastan, 1. (1999). Identification of three new altemate human kallikrein 2 transcripts: evidence of long transcript and alternative splicing. Bioclrem Biophys Res Commun 264: 833-9.

Liu, X. L., Wazer, D. E., Watanabe, K. and Band, V. (1996). Identification of a novel senne protease-like gene, the expression of which is down-regulated dunng breast cancer progression. Cancer Res 56: 337 1-9.

Lovgren, J., Airas, K. and Lilja, H. (1999). Enzymatic action of human glandular kallikrein (hK2) substrate specificity and regulation by Zn2+ and extracellular protease inhibitors. Eur J Biochem 262: 78 1-789.

Lovgren, J., Rajakiski, K., Karp, M., Lundwall, A. and Lilja, H. (1997). Activation of the zymogen form of prostate-specific antigen by human kallikrein 2. Biochem Biophys Res Commun 238: 549-555.

Luo, L. Y., Grass, L. and Diamandis, E. P. ( 2 0 ) . The normal epithelial cell- specific 1 (NESI) gene is up-regulated by steroid hormones in the breast carcinoma ce11 line BT-474. Anticancer Res 20: 98 1-6.

MacDonald, R. J., Margolius, H. S. and Erdos, E. G. (1988). Molecular biology of tissue kallikrein. Biochem J 253: 3 13-2 1.

Maeda, H., Wu, J., Okamoto, T., Maruo, K. and Akaike, T. (1999). Kallikrein- kinin in infection and cancer. Immunophannacology 43: 115-28.

Magklara, A., Grass, L. and Diamandis, E. P. (2000). Differential steroid hormone regulation of human glandular kallikrein (hK2) and prostate specific antigen (PSA) in breast cancer ce11 lines. Breast Cancer Res Treat In press:

Magklara, A., Scorilas, A., Catalona, W. J. and Diamandis, E. P. (1999). The combination of human glandular kallikrein and free prostate- specific antigen (PSA) enhances discrimination between prostate cancer and benign prostatic hyperplasia in patients with moderatel y increased total PSA. Clin Chem 45: 19606.

-*

Mahabeer, R. and Bhoola, K. D. (2000). Kallikrein and kinin receptor genes. Pfi~rnirniaco1Tlier 88: 77-89,

Margolius, H. S. (1998). Tissue kallikreins: Structure, regulation, and participation in mammalian physiology and àisease. Clin Rev Allergy Immunol16: 337- 49.

Margolius, H. S., Horwitz, D., Pisano. J. J. and Keiser, H. R. (1974). Urinary kallikrein excretion in hypertensive man. Relationships to sodium intake and sodium- retaining steroids. Circ Res 35: 820-5.

Mason, A. J., Evans, B. A., Cox, D. ReT Shine, J. and Richards, R. 1. (1983). Structure of mouse kallikrein gene family suggests a role in specific processing of biologically active peptides. Nature 303: 300-7.

Matsumura, Y., Kimura, M., Yamamoto, T. and Maeda, H. (1988). Involvement of the kinin-generating cascade in enhanced vascular permeability in tumor tissue. Jpn J Cancer Res 79: 1327-34.

Mitsui, S., Tsumoka, N e T Yarnashiro, K., Nakazato, H. and Yamaguchi, N. (1999). A novel form of human neuropsin, a brain-related serine protease, is generated by alternative splicing and is expressed preferentially in human adult brain. Eur J Biochem 260: 627-34.

Murray, S. Re, Chao, J., Lin, F. K. and Chao, L. (1990). Kallikrein multigene families and the regulation of their expression. J Cardiovasc Phannacol 15: S7- 16.

Murtha, P., Tindall, D. J. and Young, C. Y. (1993). Androgen induction of a human prostate-specific kallikrein, hKLK2: characterization of an androgen response element in the 5' promoter region of the gene. Biochemistry 32: 645964.

Nelson, P. S., Gan, L., Ferguson, C., Moss, P., Gelinas, R., Hd, L. and Wang, K. (1999). Molecular cloning and characterization of prostase, an androgen- regulated senne protease wi th prostate-restricted expression. Proc Nad Acad Sci (I S A 96: 3 1 14-9.

Papsidero, L. D., Wang, M. C., Valenzuela, L. A., M u ~ h y , G. P. and Chu, T. M. (1980). A prostate antigen in sera of prostatic cancer patients. Cancer Res 40: 2428-32.

Plendl, J., Snyrnan, C., Naidoo, S., Sawant, S., Mahabeer, R. and Bhoola, K. D. (2000). Expression of tissue kallikrein and kinin receptors in angiogenic microvascular endothelial cells. Bi01 Chem 381: 1 103- 15.

Pretlow, T. O., Pretlow, T. P., Yang, B., Kaetzel, C. S., Delmoro, C. M., Kamis, S. M., Bodner, D. R., Kursh, E., Resnick, M. 1. and Bradley, E. L., Jr. (1991). Tissue concentrations of prostate-specific antigen in prostatic carcinoma and benign prostatic hyperplasia. Int J Cancer 49: 645-9.

Rae, F Bulmer, B., Nicol, D. and Clements, J. (1999). The human tissue -E ---- - *'-

kallilrreins (KLKs 1-3)and a novel faXl-&tmiihsci$i Ge expressed in a rend cell carcinoma cDNA Library (In h e s s Citation]. Immunophamacology 45: 83-8.

Raidw, D. M., Ramsaroop, R., Naidw, S. and Bhoola, K. D. (1997). Visualisation of tissue kallikrein and kinin receptors in human astrocytomas. Br J Phannacol 122: 107,

Raidw, D. M., Sawant, S., Mahabeer, R. and Bhoola, K. D. (1999). Kinin receptors are expressed in human astmcytic tumour cells. Immunophamucology 43: 255- 63.

Regoli, D. and Barabe, J. (1980). Pharmacology of bradykinin and related kinins. Phannacol Rev 32: 1-46.

Rehbock, J., Buchinger, P., Hermann, A. and Figueroa, C. (1995). Identification of immunoreactive tissue kallikrein in human ductal breast carcinomas. J Cancer Res Clin Oncol 121: 64-8.

Riegman, P. H., Klaassen, P., van der Korput, J. A., Romijn, J. C. and Trapman, l. (1988). Molecular cloning and characterization of novel prostate antigen cDNA's. Biochem Biophys Res Commun 155: 18 1-8.

Riegman, P. H., Vlietstra, R. J., Klaassen, P., van der Korput, J. A., Geurts van Kessel, A., Romijn, J. C. and Trapman, J. (1989a). The prostate-specific antigen gene and the human glandular kallikrein-1 gene are tandemly located on chromosome 19. FEBS k t t 247: 123-6.

Riegman, P. H., Vlietstra, R. J., Suurmeijer, L., Cleutjens, C. B. and Trapman, J. (1992). Characterization of the human kallikrein locus. Genomics 14: 6-1 1.

Riegman, P. H., Vlietstra, R. J., van der Korput, H. A., Romijn, J. C. and Trapman, J. (1991a). Identification and androgen-regulated expression of two major human gtandutar kdtikrein- f (hGK- 1) mRNA spccies. Mat Celf EndocrinoI 76: 18 1-90.

Riegrnan, P. H., Vlietstra, R. J., van der Korput, J. A., Bnnkrnann, A. 0. and Trapman, J. (1991b). The promoter of the prostate-specific antigen gene contains a functional androgen responsive element. Mol Endocrinol5: 1921-30.

Riegman, P. H., Vlietstra, R. J., van der Korput, J. A., Romijn, J. C. and Trapman, J. (1989b). Characterization of the prostate-specific antigen gene: a novel human kallikrein-like gene. Biochem Biuphys Res Commun 159: 95-102.

Rittenhouse, H. G., Finlay, J. A., Mikolajczyk, S. D. and Partin, A. W. (1998). Human Kallikrein 2 (ha) and prostate-specific antigen (PSA): two closely related, but distinct, kallikreins in the prostate. Crit Reu Clin Lab Sei 35: 275-368.

- - Roberts, R. A. and Gullick, W. - - -- J. (1989). - - - Bradykinin - - - receptor number and sensivity to ligand stimulation of mitogenesis is increased by expression of a mutant ras oncogene. J Ce11 Sci 94: 527-35.

Rodell, T. C., Naidoo, Y. and Bhoola, K. D. (1995). Role of kinins in inflammatory responses. Prospects for dmg therapy. Clin Immunother 3: 18-32.

Sambrook, J., Fritsch, E. F. and Maniatis, T. 1989. Molecular Cloning: A Lobortatory Manual, Second Edition. Cold Sping Harbor Laboratory, NY.

Sauter, E. ReT Daly, M., Linahan, K., Ehya, H., Engstrom, P. F., Bonney, G., Ross, E. A., Yu, H. and Diamandis, E. (1996). Prostate-specific antigen levels in nipple aspirate fluid correlate with breast cancer risk. Cancer Epidemiol Biutnurkers Prev 5: 967-70.

Schedlich, L. J., Bennetts, B. H and Moms, B. J. (1987). Primary structure of a human glandular kalli krein gene. DNA 6: 429-37.

Schuur, E. R., Henderson, G. A., Kmetec, L. A., Miller, J. DeT Lamparski, H. G. and Henderson, D. R. (1 996). Prostate-speci fic antigen expression is regulated b y an upstream enhancer. J Bi01 Chem 271: 7043-5 1 .

Skakkebaek, N. E. (1972). Possible carcinoma-in-situ of the testis. Lrrncet 2: 5 16- 7.

Skakkebaek, N. E., Berthelsen, J. G., Giwercman, A. and Muller, J. (1987). Carcinoma-in-situ of the testis: possible origin from gonocytes and precursor of al1 types of genn ce11 tumours except spermatocytoma. Int JAndrol10: 19-28.

Slawin, K. M., Shariût, S. F., Nguyen, C., kventis, A. K., Song, W., Kattan, M. W., Young, C. Y., Tindall, D. J. and Wheeler, T. M. (2000). Detection of metastatic prostate cancer using a splice variant-specific reverse transcriptase-polymerase chain reaction assay for human glandular kallikrein. Cancer Res 60: 7142-8.

Stege, R., Grande, M., Carlstrom, K., Tribukait, B. and Pousette, A. (2000). Prognostic significance of tissue prostate-specific antigen in endocrine-treated prostate carcinomas. Clin Cancer Res 6: 160-5.

Stenman, U. H. (1999). New ultrasensitive assays facilitate studies on the role of human glandular kallikrein (hK.2) as a marker for prostatic disease [editorial; comment]. Clin Cheni 45: 753-4.

Stenman, U. H., Leinonen, J., Alfthan, H., Rannikko, S., Tuhkanen, K. and Alfthan, O. (1991). A complex between prostate-specific antigen and alpha 1- antichymotrypsin is the major form of prostate-specific antigen in serum of patients with prostatic cancer: assay of the complex improves clinical sensitivity for cancer. Cancer Res 51: 222-6.

A - - St-ephenson, S. A., Verity, K., Ashworih-, L. K. and Clements, J. A. (1999).

Localization of a new prostate-specific antigen-related serine protease gene. KLK4, is evidence for an expanded human kallikrein gene family cluster on chromosome l9q 13.3- 13.4. J Bi01 Chem 274: 232 10-4.

Stryer, L. 1995. Biochemistry, 4th. W.H. Freeman and Company, New York. 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 Bi01 Chem 272: 2 1582-8.

Tanaka, T., Isono, T., Yoshiki, T., Yuasa, T. and Okada, Y. (2000). A novel form of prostate-specific antigen transcript produced by alternative splicing. Cancer Res 60: 56-9.

Undenvood, L. J., Tanimoto, H., Wang, Y., Shigemasa, K., Parmley, T. H. and O'Brien, T. J. (1999). Cloning of tumor-associated differentially expressed gene-14, a novel senne protease overexpressed by ovarian carcinoma. Cancer Res 59: 4435-9.

van Leeuwen, B. H., Evans, B. A., Tregear, G. W. and Richards, R. 1. (1986). Mouse glandular kallikrein genes. Identification, structure, and expression of the renal kalli krein gene. J Bi01 Chem 261: 5529-35.

van Leeuwen, B. H., Penschow, J. D., Coghlan, J. P. and Richards, R. 1. (1987). Cellular basis for the di fferential response of mouse ka1 li krein genes to homonai induction. Embo J 6: 1705-13.

Venter, J. C. and et al (200 1). The sequence of the human genome. Science 291: 1304-1350,

Wang, D. Z., Song, Q., Chen, L. M., Chao, L. and Chao, 1. (1996). Expression and cellular localization of tissue kallikrein-kinin system in human adrenal gland. Am J Physiol27l: F709- 1 6.

Wang, M. C., Papsidero, L. D., Kuriyama, M., Valenzuela, L. A., Murphy, G. P. and Chu, T. M. (198 1). Prostate antigen: a new potential marker for prostatic cancer. Prostate 2: 89-96.

Webber, M. M., Waghray, A. and Bello, D. (1995). Prostate-specific antigen, a senne protease, facilitates human prostate cancer ce11 invasion. Clin Cancer Res 1: 1089- 94,

Werle, E. (1934). Zur kenntnis des haushalts des kallikreins. Biochem Z 269: 415- 434.

- . - - Wines, D. R.. Brady, J. M., Pritchett, D. B., Roberts, J. L. and MacDonald, R. J.

(1989). ÔrgGization and expression ofthe-ratkallikrein gene f h l y . J Bi01 Chem 264: 7653-62.

Yoshida, S., Taniguchi, M., Hirata, A. and Shiosaka, S. (1998). Sequence analysis and expression of human neuropsin cDNA and gene. Gene 213: 9-16.

Yousef, G. M., Luo, L. Y., and Diamandis, E.P. (1999). Identification of novel human kalli krein-like genes on chromosome l9q 13.3-q 13.4. Anticancer Res 19: 2843-52.

Yousef, G. M., Chang, A. and Diamandis, E. P. (2000a). Identification and characterization of KLK-L4, a new kallikrein-like gene that appears to be down-regulated in breast cancer tissues, J Bi01 Chem 275: 1 189 1-8.

Yousef, G. M., Chang, A., Scorilas, A. and Diamandis, E. P. (2ûûûb). Genomic organization of the human kallikrein gene famil y on chromosome 19q 13.3-q 13.4 [in Process Citation]. Biuchem Biophys Res Commun 276: 125-33.

Yousef, G. M. and Diamandis, E. P. (2000). The expanded human kallikrein gene farnily: locus characterization and molecular cloning of a new member, KLK-L3 (KLK9). Genomics 65: 1 84- 194.

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

Yousef, G. M., Obiezu, C. V., Luo, L. Y., Black, M. H. and Diamandis, E. P. (1999). Prostase/KLK-L1 is a new member of the human kallikrein gene farnil y, is expressed in prostate and breast tissues. and is honnonally regulated. Cancer Res 59: 4252-6.

Yousef, G. M., Sconlas, A., Jung, K., Ashworth, L. K. and Diamandis, E. P. (2000c). Molecular cloning of the human kallikrein 15 gene (KLK 15): Upregulation in prostate cancer. J Bi01 Chem

Yu, D. C., Sakamoto, G. T. and Henderson, D. R. (1999). Identification of the transcriptional regulatory sequences of human kalli krein 2 and their use in the construction of calydon virus 764, an attenuated replication comptent adenovirus for prostate cancer therapy . Cancer Res 59: 1498-504.

Yu, H., Bowden, D. W., Spray, B. J., Rich, S. S. and Freedman, B. 1. (1998a). Identification of human plasma kallikrein gene polymorphisms and evduotion of their role in end-stage rend disease. Hypertension 31: 906- 1 1.

Yu, H., Diamandis, E. P., Levesque, M., Giai, M., Roagna, R., Ponzone, R., Sismondi, P., Monne, M. and Croce, C. M. (1996). Prostate specific antigen in breast

cancer, benign breast disease and normal breast tissue. Breast Cancer Res Treat 40: 171- --; - . - - .

8 *

Yu, HeT Diarnandis, E. P., Zarghami, N. and Grass, L. (1994). Induction of prostate specific antigen production by steroids and tamoxifen in breast cancer ce11 lines. Breast Cancer Res Treat 32: 29 1 -300.

Yu, H., Giai, M.. Diarnandis, E. P., Katsaros, D., Sutherland, D. J., Levesque, M. A., Roagna, R., Ponzone, R. and Sismondi, P. (1995). Prostate-specific antigen is a new favorable prognostic indicator for women with breast cancer. Cancer Res 55: 2lO4- 10.

Yu, H., Levesque, M. A., Clark, G. M. and Diamandis, E. P. (1998b). Prognostic value of prostate-specific antigen for women with breast cancer: a large United States cohort study. Clin Cancer Res 4: 1489-97.

Zargharni, N., Grass, L. and Diamandis, E. P. (1997). Steroid hormone regulation of prostate-specific antigen gene expression in breast cancer. Br J Cancer 75: 579-88.