Citation
In Search of a Parotid Secretory Protein Protease: A Focus on Glandular Kallikrein 22

Material Information

Title:
In Search of a Parotid Secretory Protein Protease: A Focus on Glandular Kallikrein 22
Creator:
ALVARADO, JAVIER BRIAN ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Amino acids ( jstor )
CDNA libraries ( jstor )
Complementary DNA ( jstor )
DNA ( jstor )
Gels ( jstor )
Plasmids ( jstor )
Polymerase chain reaction ( jstor )
Proteins ( jstor )
Saliva ( jstor )
Salivary glands ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Javier Brian Alvarado. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/31/2006

Downloads

This item has the following downloads:


Full Text











IN SEARCH OF A PAROTID SECRETARY PROTEIN PRO TEASE: A FOCUS ON
GLANDULAR KALLIKREIN 22













By

JAVIER BRIAN ALVARADO


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2006

































Copyright 2006

by

Javier Brian Alvarado
















ACKNOWLEDGMENTS

I wish to thank Dr. Ammon Peck for allowing me to conduct research on this

interesting project in his lab. He has taught me the value of good leadership and afforded

me the opportunity to explore my own ideas and creative thinking. Also, I wish to thank

Dr. Smurti Killedar for her support and scientific advisement. She is an intelligent, kind

and honest person who has provided me with the background introduction for this

exciting proj ect. Dr. Seunghee Cha has also been a very important resource for Sjoigren' s

Syndrome throughout this research. She has provided me with clinical insight and

information as to the general nature of the Sjoigren' s Syndrome.

I would like to thank Ms. Joyce Conners for her vigilant effort of keeping my

administrative affairs on the right track. I would also like to thank Janet Cornelius for her

management of the laboratory and general laboratory practical advice. I wish to thank Dr.

Sally Litherland and her laboratory staff for their helpful advice and kindness.

I would like to thank the entire Peck Lab, both past and present. Specifically, I wish

to thank Eric Singson, Jin Wang, Daniel Saban, Cuong Nguyen, and Jeff Olpako for their

support, friendship, and technical expertise.

Finally, I wish to thank my family for their support and faith in my abilities. My

cousin, D. Chris Mclendon, deserves special thanks for being my best friend and a

remarkable scientific mind. And most importantly, I wish to thank my mom and dad,

Kathy and Javier Alvarado, for their undying support and unconditional love they have

provided for me throughout my life



















TABLE OF CONTENTS


page

ACKNOWLEDGMENT S ........._.._.. ...._... ............... iii...


LIST OF FIGURES .............. ....................vi


AB S TRAC T ......_ ................. ..........._..._ viii..


CHAPTER


1 INTRODUCTION ................. ...............1.......... ......


Sj oigren' s Syndrome ................. ...............1...............
NOD M house M odel .............. .. .. .................... ...............4

Temporal Changes in Salivary Composition of NOD Mouse Model...........................6
Glandular Kallikrein ................ ...............10.................


2 MATERIALS AND METHODS .............. ...............15....


M house M odels ................ ...............15...
cDNA Library Construction .............. ...............15....
cDNA Library Screening for mGK22 .............. ........ ....... .......1
Analysis of Plasmid DNA by Gel Electrophoresis and Enzyme Digestion ...............18
Expression of Recombinant mGK22 ......__....._.__._ ......._._. ............1
Protein Analysis............... ...............20
SD S-PAGE .............. ....._ ...............20....
Commassie Blue Staining............... ...............20
Membrane Blotting............... ...............21
W western Blots .............. ...............21....
HPLC-PSP Assay .............. ...............22....
PSP Peptide Synthesis .............. .... ......_ ..... ._ ............2
Detection of PSP Peptide Cleavage by HPLC .............. ...............23....

3 RE SULT S .............. ...............25....


cDNA Library Screening for mGK22 ................ ...... ....... .. ...........2
Analysis of cDNA Cloned mGK22 and pET200 D-topo Constructs............._.._.. ......27
Analysis of mGK22 Expression ......_.._.............. .........._ ...._.. .......29
Commassie Blue Staining............... ...............30
W western Blot Analysis....................... ... ... .. .. ................3
HPLC-PSP Analysis to Determine PSP Proteolytic Activity in Crude Lysates.........32













4 DI SCUS SSION ............. ...... .__ ............... 5...


LIST OF REFERENCES ............. ...... .__ ...............61..


BIOGRAPHICAL SKETCH .............. ...............64....
















LIST OF FIGURES


Figure pg

2-1 Amino acid sequence of2~us musculus parotid secretary protein. ..........................24

3-1 Photograph of replica-plated nitrocellulose membrane hybridized with
oligonucleotide probe from cDNA library screening ................. ......................35

3-2 Alignment of nucleotide sequences from cDNA library screening. ................... .....37

3-3 Alignment of translated gene sequences from cDNA library Screening. ................39

3-4 Alignment of nucleotide sequences of mGK6 from screening of NOD.B 10.H2b
submandibular tissue cDNA library and M~us musculus mGK6 (BC010754).. .......40

3-5 Alignment of amino acid translation of the mGK6 nucleotide sequences from
screening of NOD.B10.H2b submandibular tissue cDNA library and amino acid
sequence of2~us musculus mGK6 (BC010754) .............. ...............41....

3-6 Alignment of nucleotide sequences of mGK26 from screening of NOD.B 10.H2b
submandibular tissue cDNA library and M~us musculus mGK26 (NM_0 10644).. ..42

3-7 Alignment of amino acid translation of the mGK26 nucleotide sequences from
screening of NOD.B10.H2b submandibular tissue cDNA library and amino acid
sequence of2~us musculus mGK26 (NP_034774.1) .............. .....................4

3-8 Translated amino acid sequence mGK22/pET200 D-topo plasmid construct........44

3-9 Results of gel electrophoresis analysis of PCR amplification using mGK22
specific prim ers.. ............ ...............45.....

3 -10 Results of gel electrophoresi s analysis of plasmid constructs ........._..... ...............46

3-11 Results of gel electrophoresis analysis of hindIII digested plasmid constructs.......47

3-12 Results of SDS-PAGE analysis and commassie blue staining of crude whole cell
lysates induced with IPTG with uninduced crude whole cell lysates .........._..........48

3-13 Analysis of crude whole cell lysates induced with IPTG using western blot
analysis with anti-6Xhistidine antibody .................... ...............4










3-14 Analysis of pellet and supernatant of crude lysates induced with IPTG using
western Blot analysis with anti-6Xhistidine antibody................ ...............5

3-15 HPLC-PSP assay chromatogram of 40C1L of PSP peptide incubated with 40C1L
of lysis buffer. ........._..._. ........ ........ ...._ .. ...._.__ ........._.....51

3-16 HPLC-PSP assay chromatogram of 40C1L of PSP peptide incubated with 5 CIL of
NOD.B 10.H2b and 3 5 CL of lysis buffer. .....__.___ ... .... .___ .. ...._.......5

3-17 HPLC-PSP assay chromatogram of 40C1L of PSP peptide incubated with 40 CIL
of crude whole cell lysates from the mGK22 expression..........._.._.. ........._.._.. ..53

3-18 HPLC-PSP assay chromatogram of 40C1L of PSP peptide incubated with 40 CIL
of supernatant of the crude cell lysates from the mGK22 expression. ........._.._........54
















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

IN SEARCH OF A PAROTID SECRETARY PROTEINT PRO TEASE: A FOCUS ON
GLANDULAR KALLIKREIN 22

By

Javier Brian Alvarado

August, 2006

Chair: Ammon B. Peck
Major Department: Oral Biology

Sjoigren's Syndrome is an autoimmune disorder that leads to a decrease in saliva

and tear production by affecting the salivary and lacrimal gland functions. The NOD

mouse model is an animal model that exhibits an autoimmune exocrinopathy similar to

that of Sj ogren' s Syndrome, making it an excellent model for the study of Sjoigren' s

Syndrome in humans.

Previous research has observed that the salivas of NOD mice exhibiting Sjoigren's

Syndrome contain an unknown protease that aberrantly cleaves parotid secretary protein

at a specific N-terminal point containing the NL-NL amino acid sequence. In an effort to

identify the unknown PSP cleaving protease in the NOD saliva, earlier work determined,

using purification and inhibition assays, the activity might be caused by an enzyme with

strong homology to glandular kallikrein 22. The focus of the current research has been to

test the idea that mGK22 is capable of cleaving PSP and to screen the cDNA library of









NOD mouse submandibular gland tissue for other kallikrein-like proteins that may be

capable of cleaving PSP.

Screening of cDNA library provided evidence that many kallikrein and mutant

genes known to have similar function and homology with mGK22 are expressed in the

NOD mouse submandibular gland tissue. The homologies and functional similarity of

these other kallikreins and mutant enzymes make them strong candidates as PSP cleaving

enzymes. In addition it was possible to successfully clone mGK22 from NOD

submandibular tissue and express it in a prokaryotic cell line. However, the crude extract

of the prokaryotic cells expressing mGK22 showed no PSP protease activity, suggesting

but not proving that GK22 may not be the PSP proteolytic entity. The lack of PSP

proteolytic activity of mGK22 could be the result of expressing mGK22 in a bacterial,

prokaryotic rather than an eukaryotic system. Further experimentation utilizing

eukaryotic expression systems and enzyme purification directly from NOD saliva is

probably necessary to further characterize mGK22 as well as identify if the PSP

proteolytic activity is kallikrein mediated.















CHAPTER 1
INTTRODUCTION

Sjoigren's Syndrome (Sj S) is an autoimmune disorder that affects the salivary and

lacrimal glands resulting in decreased saliva and tear production. Sj S is also characterized

by mononuclear lymphocytic infiltration of the lacrimal and salivary glands. Most Sj S

patients report the sensation of chronic dry mouth (xerostomia) and dry eyes

(keratoconjunctivitis sicca). Moreover, Sj S patients report dryness of other mucosal

surfaces such as lungs, gastrointestinal tract, vagina, and skin. However, these symptoms

are subj ective and can be the result of causalities other than Sj S such as prescription

medication side effects. Therefore, several obj ective tests have been used to confirm Sj S

in patients (Delaleu et al., 2005). Although there are several tests for specific aspects of

Sj S, there are no widely accepted criteria for Sj S classification. However, most

classification models for Sj S include the following manifestations: presence of ocular and

oral sicca symptoms, measured decreased salivary and tear flow, lymphocytic infiltration

of salivary glands, and presence of specific autoantibodies. The etiology of the disease

remains to be elucidated. As a result, Sj S in patients often goes under diagnosed and

untreated.

Sjsgren's Syndrome

Sj S is classified in two ways: primary and secondary Sj S. The classifications are

based upon clinical symptoms and the presence of other autoimmune disorders. Sj S's

association with other autoimmune disorders and presence of autoantibodies implies that

there are both local and systemic aspects of the disease. Primary Sj S exists when the









lacrimal and salivary glands are affected and no other connective tissue autoimmune

diseases are involved. Secondary Sj S occurs when the typical Sj S symptoms occur in

association with other autoimmune disorders such as rheumatoid arthritis, sclerodema, or

systemic lupus erythmatosus.

Sj S displays a sexual dimorphism and age preference. Women are diagnosed with

Sj S more often than men at a ratio greater than 9:1i. This is one of the highest female to

male ratios when compared to other autoimmune rheumatic diseases. Moreover, in

women between the age of 40 and 60years old the disease is even more prevalent.

Patients with Sj S also have a high occurrence of lymphocytic malignancies especially in

B-cell lymphocytes (Voulgarelis et al., 2003).

Although the etiology of Sj S remains unknown, many proposals of Sj S

pathogenesis have been put forth. One explanation is that the characteristic decrease of

saliva and tear production in Sj S is the result of glandular destruction caused by an

inflammatory autoimmune attack on the acinar cells of lacrimal and salivary glands.

Disruption of acinar cell apoptotic pathways is thought to play a critical role in the T-cell

mediated glandular destruction. Histopathological analyses confirms that the

lymphocytes infiltrating the minor salivary gland of Sj S patients consist mostly of CD4+

T-cells and less of B-cells. Macrophages and dendritic cells have also been found in the

salivary gland infiltrates of Sj S patients (Aziz et al., 1997; Zeher et al., 1991). A focal

score of 1 or greater, where a focus is defined as a cluster of >50 lymphocytes in a 4mm2

area, from lower lip biopsies are generally considered to be abnormal. The infiltrates are

positioned peri-ductal and peri-vascular to the lacrimal and salivary glands (Cha et al.,

2002). Acinar epithelial atrophy and fibrosis have also been observed in histological









evaluations of salivary and lacrimal glands in Sj S. It has been proposed that lymphocytic

infiltrations cause Sj S symptoms via disruption of apoptotic pathways. However, there

have been many discrepancies between glandular destruction and hyposalivation in Sj S

patients observed by researchers (Delaleu et al., 2005). Histopathogical evidence from

labial salivary gland biopsies of lymphocytic infiltration is one of the key components of

Sj S diagnosis.

The manifestations of Sj S extend far beyond the sicca symptoms of salivary and

lacrimal glands. The extra glandular symptoms can affect musculoskeletal, pulmonary,

vascular, gastrointestinal, hepatobiliary, hematological, dermatological, renal, and

nervous systems. This suggests that Sj S pathology has a systemic component affecting

other tissues and organ systems.

Indeed one of the hallmarks and diagnostic criteria of Sj S is the presence of organ

specific and non-organ specific autoantibodies in SjS patient's sera. Antibodies against

nuclear proteins such as SS-A (Ro) and SS-B (La) have been commonly found in the sera

of Sj S patient' s. Patients with Sj S commonly show high amounts oflIgG and

hypergammaglobulemia in serum analysis. Interestingly, rhemeutoid factors (RF) such as

IgM-RF and IgA-RF are found in the sera and saliva of Sj S patients (Atkinson et al.,

1989). In addition, antibodies reactive against the type3 muscarinic acetylcholine receptor

(M3R) have been detected in the sera of Sj S patients. Although the presence of

autoantibodies is prevalent in Sj S patients, the role of autoantibodies in the pathogenesis

of Sj S remains obscure.

Antibodies against M3R are thought to play a pivotal role in the pathogenesis of

Sj S by some researchers. Since secretion of water and electrolytes by acinar cells is









directly induced by acetylcholine and substance P, disruption of acetylcholine ligation

with M3R would seem to affect saliva production (Baum et al., 1993). The importance of

M3R in saliva production was confirmed via knockout mice experiments (Bymaster et

al., 2003). Moreover, the infusion of monoclonal anti-M3R antibody from NOD mice or

Sj S patient' s sera into NOD and other mouse strains resulted in a loss of secretary

function (Robinson et al., 1998). This suggests that the sicca symptoms associated with

Sj S could be the result of glandular dysfunction rather than acinar cell destruction. Thus,

B-cell activation may be a critical point in Sj S pathology indicating the importance of a

humoral response in the pathology of Sj S.

Although the presence of lymphocytes and acinar tissue destruction seem to point

to glandular destruction as the main cause of the sicca symptoms in Sj S, there are many

documented discrepancies between lymphocytic infiltration of salivary tissue and

hyposalivation. Another contributing explanation could be that the sicca symptoms of Sj S

are caused by an autoantibody mediated glandular dysfunction. The etiology of Sj S

remains to be elucidated in spite of the considerable data investigating the pathology of

Sj S. However, many hallmarks and characteristics of Sj S have been identified. Moreover,

research has identified many biomarkers that are used to help diagnose Sj S in patients.

Due to the fact that Sj S symptoms occur in the late stages of the disease, the diagnosis of

Sj S in patients remains a maj or problem amongst clinicians and researchers.

NOD Mouse Model

In an effort to further investigate Sj S pathology, many animal models have been

used in Sj S research. Limitations of using human tissue in Sj S research range from

ethical issues, environmental and dietary variance amongst subjects, and genetic

diversity. The non-obese diabetic (NOD) mouse strain displays many symptoms with









considerable similarities to human Sj S pathology. The NOD mouse develops type I

diabetes via a unique H2g7 maj or histocompaltab~ility (MHC) haplotype (Leiter and

Atkinson, 1998). The expression of Ag7 with no concomitant surface expression of an I-E

molecule in the NOD mouse leads to a difference in binding affinity and affects antigen

presentation. Also, the NOD mouse contains multiple mutations resulting in less potent

IL-2, low expression of Fcy receptor and high expression of prostaglandin synthase 2 in

macrophages, lack of complement-dependent lysis, and functional NK1+ T-cells. The

NOD mouse displays different incidence of diabetes rates between male and female in

different colonies indicating many interactions of endocrine factors, environmental

factors, and multiple genes that may confer protection from or susceptibility to the

disease.

The NOD.B 10.H2b is the congenic partner strain to the NOD mouse whose MHC

locus has been replaced with the non-diabetogenic MHC locus of C57BL/10. The

NOD.B10.H2b mouse displays an autoimmune exocrinopathy characterized by

autoantibody generation, lymphocytic infiltration, and Sj S-like sicca symptoms of the

lacrimal and salivary glands. However, NOD.B 10.H2b displays Sj S autoimmune

exocrinopathy without developing type I diabetes, making it an excellent model for the

study of primary Sj S in humans (Carnaud et al., 1992; Robinson et al., 1998c).

Additional studies on the C57BL/6 recombinant strain, which contained the insulin

dependent diabetes (Idd) susceptibility loci Idd5 derived from chromosome 1 of the NOD

mouse, displayed biochemical changes of autoimmune exocrinopathy without the loss of

secretary function (Brayer et al., 2001). Thus, it appears that chromosome 1 of the NOD









mouse controls the biochemical events and other loci critical to immune infiltration and

loss of secretary function.

Another Sj S animal model is the NOD-scid mouse. The NOD-scid mouse was

developed by breeding a homologous scid (severe combined immunodeficiency) locus

into the NOD mouse background. NOD-scid mice are unable to produce B and T cells.

As a result, the NOD-scid mouse has no functional immune system. The NOD-scid

mouse has enabled researchers to explore the non-autoimmune component of Sj S

pathology. Indeed, the NOD-scid mouse shows no lymphocytic infiltration of salivary

tissue and no decrease in saliva production. However, the NOD-scid mouse shows

changes in protein composition and absent proteolytic activity in saliva when compared

to the control mouse strain. This indicates that alteration of the NOD salivary gland

occurs independently of lymphocytic infiltration. Moreover, salivary gland dysfunction

may cause the immune response that leads to glandular destruction (Robinson et al.,

1996).

Temporal Changes in Salivary Composition of NOD Mouse Model

Research on the NOD-scid mouse has led to the suggestion that Sj S pathogenesis

occurs in a two stages: an asymptomatic stage and a symptomatic stage (Cha et al.,

2002). The asymptomatic stage precedes the symptomatic stage and is hallmarked by

several biochemical changes in the glandular function of the NOD mouse. One of the

biochemical changes of the NOD mouse is the presence of new and or aberrantly altered

proteins in the NOD saliva. Specifically, the internal cleavage of parotid secretary protein

(PSP) has been observed in the saliva of the NOD mouse and is considered a biomarker

for the disease (Robinson et al., 1998).









PSP is a secretary glycoprotein found in abundance in the parotid gland of mouse

and rat. It is also found in the lacrimal, sublingual, and submandibular glands at different

ages. Although the function of PSP is not known, one study showed that it does bind to

bacterial surfaces in a zinc-dependent manner suggesting that it may be capable of

controlling bacterial growth (Robinson et al., 1998). Another study identified human PSP

as having anti-bacterial function similar to bactericidal/permeability-increasing protein

(Geetha et al., 2003).

Researchers also noted an age-dependent variance of the composition of salivary

proteins in the NOD-scid mouse. Observations of the saliva composition of the NOD-scid

mouse showed that its salivary protein profile and flow were similar to the BALB/C and

prediabetic mice (Robinson et al., 1998). However, the observations also showed that

the saliva composition of the NOD-scid mouse changed as the mouse aged. Changes in

the salivary composition of the NOD-scid mouse include: increase in amylase expression,

increase in proteolytic activity, and aberrant expression of PSP.

The temporal changes in the saliva profile of the NOD mouse were resolved by

SDS-polyacrylamide gel electrophoresis. The gel analysis showed a disappearance of a

32kD band and the appearance of a 20kD band in the saliva of NOD-scid mouse over 15

weeks of age. Coincidentally, a 27kD band appeared in the saliva of the NOD-scid over

15 weeks of age as well. When the 32kD, 27kD, and 20kD bands were N-terminally

sequences they were found to be homologous to the published murine-PSP sequence with

interesting distinctions. The 32kD and 20kD bands were shown to have N-terminal

sequences that were identical to the secreted form of PSP that is cleaved at the start

sequence. The 27kD band was found to be identical to the PSP sequence that has been









internally cleaved at a position 27 amino acids downstream from the protein start

sequence indicating the presence of an internally cleaved PSP isoform. The internal

cleavage site was determined to be at a specific NL-NL amino acid sequence of PSP

located 27 amino acids downstream from the protein start sequence. This specific NL-NL

site appears to be an unusual cleavage site and does not serve as a substrate for any

known proteases. Database searches of known proteases have been unable to Eind any

proteases that cleave PSP at an NL-NL site (Day, 2002). Thus, the cleavage of PSP

appears to be caused be an unknown unique protease in the saliva of the NOD mouse.

Western blot analysis using a polyclonal anti-murine PSP antibody confirmed the

presence of aberrantly expressed PSP in mouse saliva of the NOD mouse. The western

blots were used to compare saliva from 10-week-old NOD-scid, 20-week-old NOD-scid,

20-week-old BALB/c, and diabetic NOD mice. The saliva 32Kd and 20Kd band was

found in both thel0 week old NOD-scid and 20 week old BALB/c. On the other hand, the

20-week-old NOD-scid and the diabetic NOD mouse contained the 27Kd and 20Kd

isoforms of PSP. Moreover, the presence of the internally cleaved 27 kD PSP isoform is

indicative of PSP cleavage that is present in the saliva of older NOD-scid and NOD mice

and absent in the younger 10 week old NOD-scid and BALB/c mice.

PSP cleavage in the NOD mouse precedes the appearance oflymphocytic infiltrates

of the salivary glands. Also, PSP cleavage in the NOD-scid mouse occurs at the time

when lymphocytic infiltrates would begin to appear in the salivary glands of the NOD

mouse. This sequential synchronization of proteolytic activity and lymphocytic

infiltration of the NOD mouse and NOD-scid suggest that the changes in the exocrine

glands are independent oflymphocytic infiltration and that the autoimmunity may be









caused by changes in the morphology of the exocrine glands. Thus, PSP cleavage could

be one of the hallmark biochemical changes of the asymptomatic phase of Sj S in the

NOD mouse model.

PSP cleavage occurs in the NOD and many of its congenic strains, but not in the

normal mice. Thus, the cleavage is present in the saliva of the diabetic NOD, NOD-scid,

and NOD.B 10.H2b mice. However, the PSP cleavage does not occur in the saliva of the

10-week old NOD-scid, 8-week old pre-diabetic NOD, C57BL/6 and BALB/c mice.

Therefore, this PSP proteolytic activity could function as a biomarker for SjS.

In 2002, researchers developed a High Performance Liquid Chromatography

(HPLC) assay to detect the cleavage fragments of a PSP-like peptide (Day, 2002). The

PSP-like peptide employed by the HPLC assay is a 15 amino acid peptide that

corresponds to amino acids 20 to 34 of the published sequence for mouse PSP. The PSP-

like peptide also includes the unusual NL-NL cleavage site that serves as a substrate for

the unknown PSP cleaving enzyme. In an effort to identify the protein responsible for

PSP cleavage, purification methods were used to extract the PSP cleaving protease from

NOD.B10.H2b SaliVa. The HPLC assay was used to detect the PSP cleaving proteases

susceptibility to various specific protease inhibitors.

The protease inhibitor experiments conducted on the NOD.B10.H2b SaliVa TOVealed

some interesting findings and candidate proteins that could be responsible for the PSP

cleavage observed in the NOD.B10.H2b saliva. One interesting discovery from the

experiments was total inhibition of the PSP cleavage by serine protease inhibitors.

Conversely, the PSP cleavage was not affected by cysteine protease inhibitors and

chelating agents designed to inhibit zinc-dependent metalloproteases as well as other









proteases stabilized by calcium (Day, 2002). The protease inhibitor assays suggest that

the unknown PSP cleaving protease in the NOD mouse could be a serine protease.

The partial purification and protein analysis of NOD.B 10.H2b SaliVa alSO

suggested candidate genes responsible for PSP cleavage. The NOD.B10.H2b SaliVa WaS

purified by Sephadex G-100 gel filtration column, which is used for batch separations of

large peptides. The fractions that retained PSP cleaving activity were analyzed by SDS-

PAGE and commassie blue staining then their protein profiles were compared to normal

control mice. The protein profiles of the NOD mouse showed a unique protein band that

was not present in the normal control mouse. When this protein band was N-terminally

sequenced the following amino acid sequence was revealed: ILGXFKXEKDSQPXQ.

This amino acid sequence was then searched against a database of known protein

sequences and was shown to have strong sequence homology to mouse glandular

kallikrein 22 (mGK22) (Day, 2002). Moreover, mGK22 is a known to belong to a serine

proteinase gene family whose inhibitor was shown to prevent PSP peptide cleavage.

Although the N-terminal fragment showed near 100% homology with mGK22, it

contained a single amino acid discrepancy with the published mgk22 amino acid

sequence. Specifically, mGK22 contains asparagine residue at position 34 whereas the N-

terminal fragment sequenced from previous research displayed an aspartic acid residue at

that position. The research performed by Day on the NOD.B10.H2b SaliVa Strongly

supports the hypothesis that the PSP cleaving enzyme in the NOD saliva is mGK22 or a

protein with high homology to mGK22.

Glandular Kallikrein

The mouse glandular kallikreins (GK) are a group of biologically active peptides

that function as highly specific esterases and are encoded by closely linked genes located









on chromosome 7. To date there are 28 GK genes in the mouse, 14 of which code for

functional proteins. The kallikrein serine proteinase family was originally defined by

their ability to release bioactive kinnin from high molecular mass precursors (Olsson and

Lundwall, 2002). For example, a maj or GK found in kidney, pancreas, and salivary

glands has been shown to cleave the precursor kininogen to release bradykinin which is a

vasoactive peptide thought to play an important role in regulating blood flow (Schschter,

1980). GK' s are thought to be involved in a wide variety of peptide processing pathways

and may represent potential regulatory steps in the conversion of inactive precursors into

biologically active peptides (Evans et al., 1986).

GKs, also known as tissue kallikreins, are a family of glycoproteins of varying

molecular mass ranging from 25-40 kD. They are related to trypsin, chymotrypsin, and

other serine proteases. The GKs all possess a histidine residue at amino acid position 41,

an aspartate residue at amino acid position 96, and a serine residue at amino acid position

189. These three amino acids form what is known as the catalytic triad, a structure

thought to be critical to the formation of the serine protease catalytic site (Young et al.,

1978). Moreover, mouse GKs -22, -9, and -13 contain the aspartate residue at amino acid

position 183 that is thought to be required for cleavage at basic amino acids (Kreiger et

al., 1974). However, unlike trypsin, GK's show a high degree of substrate specificity

(Evans et al., 1986). In fact in amino acid sequence comparisons, the GK's display a high

degree of homology with each other except in regions that are thought to be important in

determining substrate specificity. Thus the actions of GKs are highly specific suggesting

that their role (if any) in bioprocessing is exclusive to certain pathways. The high degree

of specificity and large multigene family that encodes GK's supports the hypothesis that









GK' s have an integral role in the processing of a wide variety of hormone and growth

factor precursors (Mason et al., 1983).

mGK22 is a 29 kD protein that is expressed in the salivary glands of mice in a pre-

pro zymogen form. mGK22 is activated by cleavage of the zymogen peptide at an

arginine residue located 24 amino acids downstream from the peptide start site. Although

mGK22 has characteristics similar to trypsin, it lacks the trypsin calcium-binding loop

and fails to form trypsin's six disulfide bridges. In contrast, mGK22 has the characteristic

kallikrein loop beginning at amino acid position 77 and forms five disulfide bridges

(Blaber et al., 1987). mGK22, also known as epidermal growth factor binding protein

(EGF-BP) type A, is one three GKs known to bind and cleave the mouse 9 kD epidermal

growth factor (EGF) precursor at the carboxy terminus to produce the mature growth

factor. Interestingly, EGF is a maj or protein produced by the salivary glands and secreted

in saliva. Other kallikrein EGF-BPs are EGF-BP type B and type C coded by mGKl3 and

mGK9 genes respectively. Research has shown that mGKl3 and mGK26 have a 99%

homology leading researchers to conclude that they represent allelic variations of the

same gene (Olsson and Lundwall, 2002). Although mGKs -9, -13, and -22 all bind and

process pre-pro EGF, there are no identical residues between them other than any regions

conserved between the maj ority of other kallikreins. Thus, there seems to be no obvious

critical residues between the three EGF-BPs that would confirm EGF binding ability.

Also, mGK22 and other EGF-BPs may play a crucial role in the regulation of mature

growth factors (Blaber et al., 1987).

The importance of GKs in the progression of Sj S remains to be shown in spite of

the existence of intriguing data that shows GK' s involvement with inflammation and









immune responses. Interestingly, researchers have found mGKl3, also known as EGF-

BP type B, autoantibodies in the sera of another Sj S mouse model (IQI/Jic mice).

Moreover, mGKl3 was shown to cause a proliferative response of splenic T-cells, in

vitro (Takada et al., 2004). This data supports the hypothesis that mGKl3 may act as an

auto-antigen that increases the response of T-cells to organs that commonly express

mGKl3. This hypothesis, if true, may be further strengthened by the fact that mGKs -22,

-9, -13 are expressed exclusively in the salivary glands, the target of the immune response

associated with Sj S (Drinkwater et al., 1987). Thus GKs may play an important role in

the etiology of Sj S in two ways: via lymphoproliferative activity and via autoantibody

generation.

Previous data has suggested that mGK22 may be the enzyme required for PSP

cleavage in the saliva of the NOD mouse model (Day, 2002). This hypothesis is

supported by the fact that GKs are known to have proteolytic effects on various protein

precursors and may be potential regulators of peptide activation in various specific

pathways. However, it must first be shown that mGK22 is capable of cleaving PSP in

isolation. Therefore in an attempt to further solidify mGK22's candidacy as the enzyme

responsible for PSP cleavage in the NOD mouse, mGK22 must be isolated and assessed

for PSP cleaving activity. Since one of the most reliable and accurate methods of PSP

cleavage assessment is via the aforementioned HPLC PSP assay, this method facilitates

detection of PSP proteolysis.

The focus of the current study is to determine if mGK22 is expressed in the

submandibular glands and if this GK is capable of cleaving PSP. At the same time, a

NOD.B10.H2b submandibular tissue cDNA library will be screened by hybridization with









oligonucleotide probes complementary to the N-terminus of mGK22 to determine if other

candidate genes for related proteins may be identified. In order to assess mGK22 PSP

cleaving activity and find other candidates genes responsible for cleaving PSP the

following 4 goals were established:

Construct a cDNA library of NOD.B10.H2b submandibular tissue and clone
mGK22 into a suitable vector for expression.

Probe the cDNA library of NOD.B10.H2b submandibular tissue with primers
complementary to the N-terminus of mGK22 for possible candidate genes and
mutants capable of cleaving PSP.Express mGK22 from NOD.B10.H2b
submandibular tissue cDNA library.

Detect PSP cleavage activity of expressed genes by HPLC-PSP assay.















CHAPTER 2
MATERIALS AND METHODS

Mouse Models

The animal mouse model used in this research was the NOD.B 10.H2b mouse strain.

All of the mice used in this research were approximately 14 weeks of age and purchased

from the University of Florida Department of Pathology Mouse Colony. The mice were

held under SPF conditions, provided food and water, and maintained on a 12-hour dark-

light cycle until euthanized. Studies were carried out under IACUC-approved protocol

CO17.

cDNA Library Construction

Total RNA was isolated and purified from the homogenized submandibular glands

of 14-week old NOD.B 10.H2b miCe USing the RNAeasy RNA extraction kit (Qiagen).

RNA purity and concentration were confirmed by UV absorbance at 260/280 nm and

formaldehyde gel electrophoresis. The cDNA first strand synthesis via reverse

transcription polymerase chain reaction (RT-PCR) was performed on the purified total

RNA from NOD.B 10.H2b USing the SMART (Switching Mechanism At 5' end of RNA

transcript) cDNA Library Construction Kit (Clontech), as outlined in the manufacturer' s

protocol. The SMART protocols (Clontech) employ a 5' SMART IV oligonucleotide and

a 3' CDS III primer (modified oligo (dT) primer) in order to preferentially enrich for full-

length cDNAs during first strand synthesis and subsequent PCR amplification. During

first strand synthesis, the modified oligo (dT) is used to prime the initial reaction while

reverse transcriptase's terminal transferase activity adds extra nucleotides, primarily









deoxycytodine, to the 3' end of the cDNA. The SMART IV oligo has an oligo (G) stretch

which base pairs with the deoxycytodine stretch of the cDNA creating an extended

template. The result is a full-length cDNA that contains the complete 5' end of the mRNA

and a complementary SMART IV oligo sequence that can serve as a universal priming

site for the subsequent LD-PCR amplification. The first strand synthesis cDNA was

amplified using the LD-PCR (Long Distance PCR) method according to the SMART

cDNA Library Construction Kit manual (Clontech) and sub-cloned into pDNR-LIB

plasmid (Clontech). The cDNA-containing pDNR-LIB plasmid was used to transform

25CLL of "Electromax" Topl0 electro-competent cells (Invitrogen) by electroporation

according to the manufacturer' s instructions. The transformed cells were then added to

970 CIL of LB broth and allowed to incubate for 1 hour at 370C with shaking (225 rpm) to

create the original, unamplified cDNA library. The titer of the original, unamplified

cDNA library was calculated to be 6.6 x 105 colony forming units per milliliter

(CFU/mL). The transformed bacterial cells were then plated at a concentration of 100

colonies per plate on 250 mm agar plates containing 1.5% agarose LB-media with

chloramphenicol (30 Cpg/mL). The plates were then allowed to incubate at 370C

overnight.

cDNA Library Screening for mGK22

The transformed bacterial colonies were transferred to a positively charged

nitrocellulose membrane and subjected to DNA hybridization analysis. The colonies were

hybridized to 3' tail labeled digoxigenin-1 1-dUTP/dATP DNA (dig-labeled) probes

whose sequences were identical to the N-terminus of mGK22 (5'-

ATACTTGGAGGATTTAAATGTGAGAAGAATTCCCAACCCTGG-' corresponding

to nucleotides 73-114). The hybridizations were performed according to the Genius









System Users Guide version 3.0 (Roche). Replica-plated nitrocellulose membrane colony

lifts containing cDNA, were generated by placing the membranes on top of a cold agar

plates for 1 minute. After the bacterial colonies were transferred to a positively charged

nitrocellulose membranes, the membranes were placed in series of alkaline washes (0.5N

NaOH, 1.5M NaCl and 0.5N NaOH, 1.5M NaC1, 0. 1% SDS) to lyse the cells and a

neutralization wash (1.0M Tris-HC1, 1.5M NaC1, pH 7.5). Next, the membranes were

baked at a 1200C to fix the colony DNA to the membrane. The nitrocellulose membranes,

with the fixed DNA, were pre-hybridized at 650C overnight in hybridization buffer (5X

SSC, 1.0% blocking reagent, 0. 1%ON-lauroylsarcosine, 0.02% SDS). After washing in 2X

SSC, the membranes were incubated at 680C with the dig-labeled oligonucleotide probes

in hybridization solution (2.0 pmol/mL) for 1 hour. After hybridization, the membranes

were washed in washing buffer (0. 1M maleic acid, 0. 15M NaC1, 0.3% Tween 20) to

remove any non-hybridized probes. Next, the membranes were incubated in blocking

buffer (0. 1M maleic acid, 0. 15M NaC1, 1% blocking reagent) for 30 minutes and then

placed in blocking solution that contained anti-digoxigenin (150 mU/mL) antibody

conjugated to alkaline phosphatase for 1 hour. To remove any unbound antibody the

membranes were washed twice in washing buffer. Detection of the bound antibody was

accomplished by colorimetric development of alkaline phosphatase. Colorimetric

development was performed on the membranes by adding 10 mL color substrate solution,

which contained nitroblue tetrazolium salt (0.3375 mg/mL) and 5-bromo-4-chloro-3-

indolyl phosphate (0. 175 mg/mL) in 10mL of detection buffer (100mM Tris-HC1, 1.5M

NaC1, 50mM MgCl2, pH 9.5). Colorimetric development was allowed to continue









overnight (approximately 12 hours). To prevent over-development, the membranes were

washed twice, in H20, and air-dried for storage.

Analysis of Plasmid DNA by Gel Electrophoresis and Enzyme Digestion

The plasmids of the transformed bacterial cells from the cDNA library and mGK22

expression experiments were analyzed by gel electrophoresis and restriction enzyme

digestion. Colonies from transformed cells were plated on antibiotic selective media.

Next, they were used to inoculate 50ml cultures of LB-media with the appropriate

selective antibiotic and allowed to incubate at 370C overnight. The plasmids from the

transformed cell cultures were isolated and purified using a Maxi-Prep plasmid

purification kit (Qiagen). The protocol was performed as outlined in the Maxi-Prep

plasmid purification kit manual (Qiagen). The purified plasmids were analyzed for UV

absorbance at 260/280 nm using a spec 300 (Bio-Rad) to determine DNA concentration

and purity. The plasmids were also loaded onto a 1% agarose gel containing 4.0%

ethidium bromide and electrophoresed for 2 hours at 50V. The plasmids were then

digested with restriction enzymes Xba I and Hind III (Promega), separately, to analyze

the inserts they may contain. Approximately 10.0plg of purified plasmid DNA was

combined with 10 units of Xba I (approximately 2.0C1) and incubated at 250C for 1 hour.

Separately, another sample of 10.0 Clg of purified plasmid DNA was combined with 10.0

units of Hind III (approximately 2.0C1) and incubated at 370C for 1 hour. Both digestions

were prepared in 50 Cll reactions with enzyme buffer (25.0 Tris acetate, 0.1 Potassium

acetate, 10.0 mM Magnesium acetate, 1.0 DTT, pH 7.80, 0.1 mg/ml Acetylated BSA).

After digestion incubation, both samples were incubated to 650C for 15 minutes to

deactivate the restriction enzymes. The samples were analyzed by gel electrophoresis for









their respective DNA fragment profiles. Finally, the identity and position of the inserts of

the plasmids were confirmed by DNA sequencing.

Expression of Recombinant mGK22

The cDNA from the first strand synthesis was subj ected to polymerase chain

reaction (PCR) amplification using Proofstart high-fidelity DNA polymerase (Qiagen)

and mGK22 specific primers (5 '-CACCGCACCTCCTGTCCAGTCTCGAATAC-3 '

corresponding to nucleotides 52-76 and 5'-

TCAGGGGTTTTTGGCCATAGTGTCTTTT-3 complementary to nucleotides 753-

780). To facilitate directional cloning, the forward primer contained 4 added nucleotides

(CACC) as specified in the pET200 D-topo expression vector manual (Invitrogen). The

PCR conditions were 940C for 3 min followed by 30 cycles of 940C for 45 sec, 650C for

45 sec, 720C for 2 min, followed by a final extension at 720C for 7 min. PCR products of

approximately 750 nucleotide base pairs (bp), which corresponds to the size of mGK22,

were separated from other PCR fragments by gel electrophoresis. The purified cDNA,

which contained the complete region of the mGK22 gene and stop codon, was sub-cloned

into pET200 D-topo expression vector (Invitrogen). The orientation and sequence of the

cDNA in the pET plasmid were confirmed by DNA sequencing. The ligated pET200 D-

topo vector was used to transform Escherichia coli strain BL21 (DE3) cells (Invitrogen)

for expression. The transformed cells were incubated at 370C in 100 mL of LB media

containing 50 Cpg/mL kanamycin until they reached an optical density of A600 The cells

were harvested by centrifugation at 5000g for 10 min. The cell pellets were resuspended

in 1 mL of lysis buffer (50 mM NaH2PO4, 300 mM NaC1, pH 8.0). The cells were treated

to freeze-thawing cycles and sonication (6 cycles of 10second bursts of sonication

followed by 10 seconds cooling) to lyse the cells. To prevent overheating of the samples









and protect the expressed protein, the sonication of the cells was performed on ice. To

separate insoluble cellular debris from the expressed recombinant mGK22 fusion protein,

the crude cell lysates were centrifuged at 10,000g for 30 min. The presence of the fusion

protein, which contains mGK22 residues and extra amino acids originating from the

pET200 D-topo plasmid sequence at the N-terminus of mGK22, was detected in the

lysates by western blot analysis with a 6X anti-histidine antibody. The crude cell lysates

were analyzed for enzymatic activity using the HPLC-PSP assay.

Protein Analysis

SDS-PAGE

Crude cell lysates were analyzed by sodium dodecyl sulfate polyacrylamide gel

electrophoresis (SDS-PAGE) to separate their constituent proteins by molecular weight.

The crude cell lysates were combined with 2X laemmelli buffer (BioRad) in equal

volumes (15Cl~) and incubated at 65 for 15 minutes. The entire sample, 15Cl~ of crude cell

lysates in 15.041~ of laemmelli buffer, was loaded onto a 15.0% Tris-HCI polyacrylamide

gel (Bio-Rad). The gels were electrophoresed for 1 hour at 100V in Tris Glycine-SDS

Buffer (Bio-Rad).After electrophoresis, the gels were washed twice in H20 in

preparation for the next analysis.

Commassie Blue Staining

Visualization of the constituent proteins of the crude cell lysates was accomplished

by commassie blue staining. The polyacrylamide gels from the SDS-PAGE of the crude

cell lysates were placed in commassie blue staining solution (0.125% commassie brilliant

blue R 250, 50.0% methanol, 10.0% acetic acid) and allowed to stain for 30 minutes. The

polyacrylamide gels were then placed in destain solution (50.0% methanol, 10.0% acetic

acid) for 15 minutes. Afterward, the destain solution was discarded and replaced by fresh









destain solution. The polyacrylamide gels were allowed to destain overnight. The

polyacrylamide gels were then washed twice in H20 and photographed.

Membrane Blotting

After separation of proteins, the proteins were transferred to a nitrocellulose

membrane using a Semi-Dry Transfer Cell (Bio-Rad). The proteins from the

polyacrylamide gel were transferred to the membrane at 20V for 30 minutes with

membrane transfer buffer (80% Tris Glycine-SDS buffer and 20% methanol). Following

the transfer, the membranes were washed twice in H20 and used for western blot analysis

with 6X anti-histidine antibody to detect the expressed protein.

Western Blots

The nitrocellulose membranes that contained the transferred proteins from the

crude cell lysates were subj ected to western blot analysis to detect the presence of a 6X

histidine protein. The western blot analysis was performed as outlined in the Western

Breeze western blot kit (Invitrogen) manual with the following specifications. All

incubations were performed at room temperature. After the protein transfer, the

membranes were washed twice with H20 and incubated in blocking buffer (1.0% bovine

serum albumin, 10mM Tris-HC1, 150mM NaC1, 0.05% Tween 20, pH 8.0) for 1 hour.

The membranes were washed 3 times for 5 minutes in antibody wash (10mM Tris-HC1,

150mM NaC1, 0.05% Tween 20, pH 8.0). The membranes were then incubated with 6X

anti-histidine primary antibody in blocking buffer at a concentration of 10 pmol/ml.

Afterward, 3 washes of 5 minutes each in antibody wash were performed to remove

unbound antibodies. Then the membranes were incubated with anti-rabbit IgG secondary

antibody conjugated with alkaline phosphatase (Sigma) at a dilution of 1:7500 in

blocking buffer The membranes were washed three times for 5 minutes in antibody wash









to remove unbound antibodies. After pre washing in water twice, the membranes were

color developed by incubating the membranes in color substrate solution, which

contained nitroblue tetrazolium salt (0.33 mg/mL) substrate and 5-bromo-4-chloro-3-

indolyl phosphate (0. 165 mg/mL) substrate in 10mL of alkaline phosphatase buffer

(100mM Tris-HC1, 100mM NaC1, 5.0mM MgCl2, pH 9.5). To prevent over-development,

after 4 hours the membranes were washed twice for 5 minutes in H20. Membranes were

then air-dried for storage.

HPLC-PSP Assay

PSP Peptide Synthesis

The PSP peptide employed by the HPLC-PSP assay was a custom designed peptide

synthesized by the University of Florida ICBR Protein Chemistry Core Facility,

Gainesville. The PSP peptide was synthesized on an Applied Biosystems Peptide

Synthesizer model 432A using solid phase FMOC chemistry. The PSP peptide was

designed to mimic PSP, since it contains the amino acid residues of the PSP cleavage site

it can act as a substrate for the unknown PSP cleaving protease. The PSP peptide amino

acid sequence was N'-EAVPQNLNLDVELLQ-C'. The PSP peptide amino acid sequence

was identical to the amino acids 20 through 34 of the entire published M~us musculus PSP

sequence (figure 2-1). Also, the PSP peptide contains the PSP protease specific NL-NL

cleavage site corresponding to the 26th and 27th amino acid positions of the M~us musculus

PSP sequence. The PSP peptide molecular weight and purity were confirmed by HPLC

(figure 2-2) and mass spectroscopy (figure 2-3). The University of Florida ICBR Protein

Chemistry Core Facility performed both analyses on the PSP peptide. For the purpose of

PSP cleavage detection by HPLC, 25 mg of the PSP peptide was dissolved in 10 mL of

PSP peptide buffer (10 mM Tris-HC1, pH 8.02). The concentration of PSP peptide in the









PSP peptide solution, used as the enzyme substrate in the HPLC-PSP assay, was 2.5

mg/ml .

Detection of PSP Peptide Cleavage by HPLC

To detect PSP cleavage activity in crude cell lysates, the PSP peptide was used as a

substrate for the unknown PSP cleaving enzyme. 40 Cl1 of crude cell lysates from the

mGK22 expression experiments was added to 40 Cl1 of the PSP peptide solution and

incubated at 42 OC overnight, to ensure complete cleavage. As positive and negative

control samples for PSP cleavage activity, saliva from NOD.B 10.H12b and BALB/C

mouse strains were analyzed for PSP cleavage activity by HPLC. The saliva control

samples (NOD.B10.H2b and BALB/C) were separately incubated with PSP peptide. In

each sample, 40 Cl1 of the PSP peptide solution and 10 Cl~ of saliva were combined along

with 30 Cl1 of PSP peptide buffer and incubated at 420C overnight. All samples, both

crude cell lysates and control saliva samples, were filtered through a 0.45 Cpm

Regenerated Cellulose Micro-Spin filter tube (Alltech) by centrifugation at 5000g for 5

minutes prior to HPLC analyses.

The HPLC-PSP assay used to assess the PSP cleavage activity functions by

detecting the cleavage fragments of the PSP peptide. The HPLC-PSP assay uses a

reverse-phase 5Cpm 300A+ Jupiter Cls column (phenomenex) to separate peptides and

peptide fragments from each other and other compounds based on their binding affinities

to the column. The HPLC-PSP assay uses a linear gradient elution of two buffers: buffer

A (0.1 % Trifluroacetic acid in Acetonitrile) and buffer B (0.1% Trifluroacetic acid in

HPLC grade H120). The HPLC-PSP assay method used in this research specified for a

linear gradient elution of 10% buffer A and 90% of buffer B to 90% buffer A and 10% of

buffer B at a rate of 1.0 ml/min over a duration of 20 minutes for peptide separation. The










peptides were detected by an AD20 wavelength detector (Dionex), which distinguishes

peptides by their respective ultraviolet (UV) absorbance at 214nm as they are eluted from

the column. The elutions were profiled by documentation on a graph of retention time

versus UV absorbance. Also, a linear regression of PSP peptide peak areas was

constructed, enabling relative amounts of PSP peptide in samples to be determined.


-20 -10 +1 10 20
1 HFQI.GSLVVL CGLLIGHSES LLGELGSAVN HLKII.NPPSE AVPQNLHLDV ELL ALTSVP
61 LAKNSILETL NTADLGN1KS FTSINGLLLK INNLKVLDFQ AKISSNGNGI DLTVPLAGEA
121 SLVLPFIGKT VDISYSLDLI NSLSIKTNAQ TGLPEVTIGK CSSNTDKISI SLLGHHLPII
181 NSILDGVSTL LTSTESTYLQ NFICPL1QYV LSTINPSVLQ GLISNLLAGQ VQLAL





Figure2-1.Amino acid sequence of2~us musculus parotid secretary protein. The complete
PSP is shown intact with the 20 amino acid leader sequence. The rectangular
box outlines the synthetic PSP peptide used in the HPLC-PSP assay. The
arrow labels the specific PSP protease NL-NL site.














CHAPTER 3
RESULTS

cDNA Library Screening for mGK22

A NOD.B 10.H2b submandibular gland tissue cDNA library was screened with an

(42-mer) oligonucleotide probe identical to the N-terminal sequence of the mGK22 gene.

The oligonucleotide probe sequence was also identical to the back-translated sequence of

the previously reported unique protein present in the PSP cleaving NOD.B 10.H2b SaliVa

(Day, 2002). The screening of 65 plates of the cDNA library at 100 colonies per plate,

approximately 6500 clones, revealed 35 positive colonies. Since the initial colony titer

was 6.6 X 105 CFU/mL, approximately 1% of the total library was screened. Thus, these

35 colonies probably represent abundant mRNA species. An example of a screened

nitrocellulose membrane is shown in (Figure 3-1).

The cloning region of each plasmid these positive colonies contained was directly

sequenced to identify the cDNA inserts they contained. The sequencing revealed

interesting gene identities of the positive clones including the kallikrein genes mGK9,

mGK26, and mGK6. The alignments and comparisons of these kallikrein genes are

shown in figures 3-2 thru 3-8. Other genes that were identified in the screening were:

salivary protein, salivary protein, cytochrome P450, and PRL-inducible protein. An

interesting result of the cDNA library screening was that mGK6 was found to be the most

frequently expressed in the cDNA library. On the other hand, mGK22 was never

identified in the cDNA library via screening with an mGK22 homologous probe.









However, mGK22 was identified and amplified from the cDNA library using mGK22

specific primers (Figure 3-9).

The products from two separate PCR reactions, using mGK22 specific primers and

the cDNA library as a template, were sequenced. Each PCR reaction, when sequenced,

showed that each reaction produced a different form of the of the mGK22 gene. The

sequencing revealed that one PCR reaction produced a transcript identical to the

published mGK22 sequence while the other PCR reaction produced a transcript with a

single base pair deviation from the published mGK22 sequence. When translated, the

single base pair deviation resulted in an aspartic acid for an asparagine amino acid

substitution at position 34 in mGK22. This substitution was identical to the substitution

found in the N-terminal sequencing of unique protein bands from purified fractions with

PSP cleaving activity from previous research (Day, 2002). Thus, the PCR amplifications

detected the presence of two different forms of the mGK22 gene present in the cDNA

library of NOD.B 10.H12b submandibular gland tissue. Moreover, the different forms of

mGK22 were produced by PCR in isolation and not as a mixture of both forms.

In addition to the cloned mGK genes showing 100.0% homology to the mGK

published sequences, the screening revealed several genes in the cDNA library that

contained minor base pair deviations from the published sequence. These point

mutations, if translated, could result in amino-acid substitutions that deviate from the

published amino acid sequences of the mGK proteins. For example, one positive colony

whose sequence showed homology with mGK26 displayed a single nucleotide

discrepancy resulting in a lysine to proline amino acid substitution at the 112th amino-

acid position (Figures 3-6 and 3-7). This nucleotide discrepancy was identified in









colonies obtained from two separate screenings (Table 1). Also, a positive colony with

sequence homology to mGK6 that displayed another single nucleotide discrepancy

resulting in a proline to lysine amino acid substitution at the 123rd amino-acid position

was identified in colonies obtained from three separate screenings. Thus, the cDNA

library screening of approximately 6500 clones did not identify any specific mGK22

colonies, but did show the presence of several other mGKs and mGK variants with

significant frequency.

Analysis of cDNA Cloned mGK22 and pET200 D-topo Constructs

In order to express mGK22 from the NOD mouse, expression vectors for mGK22

were constructed. The expression of mGK22 was performed using the protocol as

outlined by Matsui et al. (2000). Using NOD.B 10.H12b submandibular tissue cDNA as a

template, PCR amplification, with mGK22 specific primers produced a fragment of

approximately 750 bp (Figure 3-9 lanes 2 and 3). A fragment of 750 bp is consistent with

the published size of mGK22. Moreover, the predicted size of the PCR fragment

generated from the mGK22 specific primers was 704 bp. Indeed the PCR fragment

generated from mGK22 specific primers and mouse submandibular tissue cDNA

displayed a slight shift below the 750 bp molecular weight marker (Figure 3-9 lane 1).

Thus the mass of the nucleotide sequence generated from the PCR cloning of mGK22

was consistent with that of mGK22.

The fragment generated from the PCR reactions was ligated into pET200 D-topo

vector for expression. The resulting plasmid-vector construct was designed to contain the

entire coding region for mGK22 as well as a 6X histidine leader sequence. The plasmid-

vector construct was used to transform E. coli (BL21).









The transformed E. coli (BL21) plasmids were purified and analyzed by gel

electrophoresis to determine their size (Figure 3-10). If mGK22 has correctly ligated into

the plasmid, the resulting plasmid's approximate molecular weight is expected to be 6.5

Kbp. Lane 4 of Eigure 3-10 contains a plasmid with an approximate weight of 6.5 Kbp.

Moreover, the negative control in lane 5 of Eigure 3-10 (an empty pET200 D-topo

plasmid with a molecular weight of 5.6 Kbp) displayed its plasmid at a lower position

than the plasmid in lane 4, thus indicating that the plasmid in lane 5 is of a lower

molecular weight than the plasmid in lane 4. The negative control in lane 5 was

determined to have an approximate molecular weight of 5.6Kbps. Therefore, the

molecular weight of the plasmid in lane 4 of Eigure3 -10 is consistent with expected

molecular weight of the mGK22/pET200 D-topo construct.

The plasmids were analyzed by digestion with the restriction enzyme Hind III to

yield predicted size fragments. Analysis of the nucleotide sequence of the vector

construct showed that mGK22 contained a Hind III restriction enzyme cleaving site.

Also, the pET200 D-topo plasmid, supplied by Invitrogen, has a Hind III restriction

enzyme cleaving site. Therefore, if the mGK22 cDNA insert ligated into pET200 D-topo

vector in the correct orientation, it should generate DNA fragments of approximately 803

bp and 5646 bp when digested with Hind III. On the other hand, the PCR positive control

(lac Z) also contains a Hind III restriction enzyme cleaving site. However, the Hind III

cleaving site in the PCR positive control (lac Z) is at a different position than the Hind III

cleaving site in mGK22 causing it to produce a slightly larger fragment when digested

with Hind III. The positive control lacZ/pET200 D-topo construct should generate DNA

fragments of approximately 845 bp and 5646 bp, slightly larger than mGK22-pET200 D-









topo construct. Figure 3-11 displays the gel electrophoresis of the vector constructs

digested with Hind III. Examination of Eigure 3-11 shows that a small DNA fragment in

lane 4 (lacZ/pET200 D-topo positive control) was slightly larger than the small DNA

fragment in lane 5 (mGK22). Moreover, the large DNA fragments of the 2 separate

digests were relatively the same size, as expected. These DNA fragment profies were

consistent with the predicted DNA fragment profies of mGK22/pET200 D-topo

construct. The actual DNA sequence of the mGK22/pET200 D-topo construct was

confirmed by DNA sequencing. The translation of the DNA sequence showed that the

insert that was contained in the mGK22/pET200 D-topo construct is the complete pro-

mGK22 gene transcript with extra sequences of amino acids on the N-terminus that were

donated by the pET200 D-topo plasmid (Figure 3-8). Thus, the restriction enzyme

digestion analysis and DNA sequencing of the mGK22/pET200 D-topo confirmed that

the construct contained the PCR-cloned mGK22 gene, intact, in frame for expression, and

in the correct orientation.

Analysis of mGK22 Expression

The protein profies of E. coli (BL2 1) transformed with mGK22/pET200 D-topo

plasmid were examined by commassie blue staining (Figure 3-12) and western blot

analysis (Figures 3-13 and 3-14) with an anti-6X histidine antibody. The protein profies

of E. coli (BL2 1) showed the appearance of a protein band with a 6X histidine tag

following induction with IPTG. The recombinant protein from mGK22/pET200 D-topo

should have an approximate molecular weight of 27.7 Kbp. The commassie blue staining

and western blot analysis both showed the appearance of protein at an approximate

molecular weight of 27 Kbp strongly supporting the conclusion that the E. coli (BL21)

are expressing the recombinant protein from mGK22/pET200 D-topo plasmid.









Commassie Blue Staining

As stated above, the protein profies of the E. coli (BL21) obtained from the

commassie blue staining show the appearance of a protein band with the approximate

weight of 27 kD. In Eigure 3-12, lanes 2 and 3 the appearance of a band just above the 25

kD molecular weight marker indicates the expression of the recombinant

mGK22/pET200 D-topo protein. In lanes 1 and 11, the molecular weight markers that

correspond to an approximate molecular weight of 35 kD and 25 kD are the 6th and 7th

band down from the top, respectively. Moreover, the 27 kD protein band in lanes 4, 6, 8,

and 10 of Eigure 3-12 are absent in lanes 2, 3, 5, 7, and 9 of the same Eigure. Lane 2 of

Eigure 3-12 contains the crude lysate of the E. coli (BL21) prior to induction with IPTG

(time point zero). Lanes 3, 5, 7, and 9 contain crude lysate of the E. coli (BL21) that was

not induced with IPTG. Thus, the commassie blue staining shows the presence of an

inducible protein in the crude lysate of E coli (BL21) cells transformed with

mGK22/pET200 D-topo plasmid.

Western Blot Analysis

To detect the expression of the recombinant mGK22 protein in the E. coli (BL21)

transformed with mGK22/pET200 D-topo plasmid, a western blot analysis was

performed on the crude lysates of the transformed E. coli (BL21). A western blot can be

used to detect the presence of a 6X histidine tag via the employment of an anti-6X

histidine antibody. The expression vector mGK22/pET200 D-topo has been designed to

incorporate a 6X histidine tag on the N-terminus of mGK22. Thus the presence of the 6X

histidine tag, at the approximate molecular weight of 27.7 kD, would further confirm the

expression of recombinant mGK22 in the transformed E. coli (BL21).









In figure 3-13, lanes 3 and 4 show the presence of a 6X histidine tag at a position

just above the parallel of the 25 kD molecular weight marker (Bio-Rad) located in lane 1.

The position of the 6X histidine protein bands in lanes 3 and 4 of figure 3-7 corresponds

to an approximate molecular weight of 28 kD. The molecular weight of the 6X histidine-

tagged protein in the crude lysates is consistent with the production of the recombinant

mGK22 protein from the mGK22/pET200 D-topo plasmid.

Figure 3-14 displays the western blot analysis of the crude lysates respective pellet

and supernatant following centrifugation. Lanes 3 and 5, which contain the pellet

component of the crude lysate at one and two hour time points, show the presence of a

6X histidine protein band with an approximate molecular weight of 28 kD. In contrast,

there are no protein bands in lanes 4 and 6 which contain the supernatant component of

the crude lysate at one and two hour time points. The 6X histidine protein present in the

crude lysates appears to be in the insoluble, pellet of the crude lysates.

The western blots also revealed an increase in the concentration of a 6X histidine

tagged protein associated with an increase in incubation time of the crude lysates of the

transformed E. coli (BL21). The intensity of the 6X histidine protein bands increases

from lane 3 to lane 4, which represents the time points 0.5 hours and 1.0 hour

respectively. The time dependent increase of the protein in the crude lysates implies that

it is an inducible protein whose concentration increases with time. Thus, there appears to

be a 6X histidine-tagged inducible protein present in the crude lysates of the transformed

E. coli (BL21). The western blot analysis performed on the crude lysates provides strong

evidence of the presence of a 6X histidine-tagged inducible recombinant protein in the

transformed E. coli (BL21).









HPLC-PSP Analysis to Determine PSP Proteolytic Activity in Crude Lysates

To assess the crude lysates of the transformed E. coli (BL21) for PSP proteolytic

activity, they were subj ected to analysis by the HPLC-PSP peptide assay (Day, 2003).

The HPLC-PSP peptide assay uses a synthetic peptide as a substrate to detect the activity

of a protease that cleaves PSP at a specific amino acid sequence. The 15 amino acid

synthetic peptide that serves as the substrate in the assay is identical to the sequence of

PSP from amino acids 20 to 34. The synthetic peptide's sequence includes the leucine-

asparagine (NL-NL) cleavage site located at amino acid positions 26 and 27 in PSP.

The peptide profiles can reveal PSP cleaving activity by detecting the presence of a

single peak, representing the intact PSP-like peptide synthetic, or split peaks, representing

the fragments of cleaved PSP-like synthetic peptide. The intact peptide and its cleavage

fragments have their respective retention times for further identification.

To determine the retention time and peak area of the uncleaved, intact PSP peptide;

it was subj ected to the HPLC-PSP peptide assay. Exactly 40 CIL of the PSP peptide

solution was added to 40C1L of lysis buffer and incubated at 42 OC overnight. The sample

was then filtered and analyzed by the HPLC-PSP peptide assay. As shown in figure 3-15,

this control sample displayed a single peak with a retention time of 13.46 min. The

retention time of this peptide was used to compare the retention times of the uncleaved,

intact PSP peptide in all future assays. The peptide-alone control sample enabled the

relative peak area for a sample ofPSP peptide, at a concentration of 1.25 mg/ml, to be

determined. The PSP peptide-alone control sample was assayed to detect the effects of

the components of the buffers used in the preparation of the crude lysates on the retention

time and peak area of the PSP peptide during the HPLC-PSP peptide assay.









In an effort to determine the retention time and peak profiles of an enzymatically

cleaved PSP peptide, the saliva of the NOD.B10.H2b mouse was analyzed by the HPLC-

PSP peptide assay. This sample, which includes NOD.B10.H2b saliva, has been

previously shown to cleave PSP peptide and serves as a positive control for the HPLC-

PSP peptide assay. Figure 3-16 displays the HPLC chromatogram of sample containing 5

CIL of NOD.B10.H2b SaliVa COmbined with 40 CIL of PSP peptide solution and 35 CIL of

lysis buffer. The sample was incubated at 42 OC overnight, filtered, and then subj ected to

the HPLC-PSP peptide assay. The two peaks at 9.16 minutes and 12.21 minutes represent

PSP peptide fragments 1 and 2, respectively. This positive control was used to show that

the enzymatic activity of the unknown PSP cleaving enzyme, present in the

NOD.B 10.H12b SaliVa, iS not affected by the components of the lysis buffer used in the

crude lysate preparations.

To assess for PSP cleavage activity in E. coli (BL21) transformed with

mGK22/pET200 D-topo plasmid, the whole cell lysate and the cleared supernatant of the

crude lysate were subj ected to the HPLC-PSP peptide assay. Briefly, 40 CIL of the crude

whole cell lysate and 40 CIL of the cleared supernatant from the mGK22 expression

experiments was added, separately, to 40 CIL of the PSP peptide solution and incubated at

42 OC overnight, to allow sufficient time for complete cleavage. The samples were

filtered and analyzed by the HPLC-PSP assay to determine the retention times and peak

areas of their peptides. In figure 3 -17, the crude whole cell lysate sample incubated with

PSP generated a single peak of interest at a retention time of 13.58 minutes. Moreover,

when the peak area was integrated and applied to a standard curve of a linear regression

of PSP peptide, the peptide concentration in the sample was approximately 1.20 mg/ml,










indicating that PSP cleavage did not occur. Likewise, figure 3-18 shows that the cleared

supernatant sample incubated with PSP peptide solution generated a single peak of

interest at a retention time of 13.45 minutes, which is the approximate retention time for

the uncleaved PSP peptide. Thus, there was no proteolysis of PSP by the cloned,

expressed mGK22. These results may not be unexpected, since the assay to this point was

carried out using the pro-mGK22 form. Moreover, Matsui et al have reported that the

pro-mGK27 form was also inactive.












































Figure 3-1. Photograph of replica-plated nitrocellulose membrane hybridized with
oligonucleotide probe from cDNA library screening. Membranes contain
replica-plated bacterial colonies transformed with plasmids containing cDNA
library inserts from NOD.B 10.H2b mouse submandibular tissue. The
membranes were incubated with a Dig-labeled oligonucleotide probe, whose
sequence was identical to N-terminal of mGK22, and detected with alkaline
phosphatase conjugated antibody. Positive colonies appear as darkened spots
on a heterogeneous background. Colonies were plated at a concentration of
100 CFU/ml.












Gene Retrieved Found mutant Frequency in Frequency of Amino Acid
from form in cDNA cDNA Mutant in Sub stituti on
cDNA Library Library cDNA Library Result of
Library Screening Screening Mutation
MGK6 Yes Yes 10 3 Prol23 4 Lysl23
MGK9 Yes No 3 NA NA
MGK26 Yes Yes 3 2 Lys1124Proll2
MGK22 No No NA NA NA



Tablel. Summary of results from screening of cDNA library made from NOD.B 10.H2b
mouse submandibular tissue. NA= not applicable











mGK22/pET ATGCGGGGTTCTCATCATCATCATCATCATGGTATGGCTAGAACGTAA 56

mGK22/pET GCAAATGGGTCGGGATCTGTACGACGATGACGATAAGGATCACCTAC 108

mGK22/pET GCAC 112
mGK22 ATGAGGTTCCTGATCCTGTTCCTAACCCTGTCCCTAGGAGGGTGTCGA 55
mGK6 ATGAGGTTCCTGATCCTGTTCCTAGCCCTGTCCCTAGGAGGGTGTCGA 55
mGK9 ATGAGGTTCCTGATCCTGTTCCTAGCCCTGTCCCTAGGAGGGTGTCGA 55
mGK26 ATGTGGTTCCTGATCCTGTTCCCAGCCCTGTCCCTAGGAGGGTGTCGA 55

mGK2 2/pET CTCCTGTCCAGTCTCGAATACTTGGAGGATTTAA ATGTGAGAAATCCAC 16 8
mGK2 2 CTCCTGTCCAGTCTCGAATACTTGGAGGATTTAA ATGTGAGAAATCCAC 111
mGK6 CTCCTGTCCAGTCTCGAATTGTTGGAGGATTTAACTGTGAAGATCACC 111
mGK9 CTCCTGTCCACTCTCGAATTGTTGGAGGATTTAA ATGTGAGAAATCCAC 111
mGK2 6 CTCCTCTCCAGTCTCGGGTGGTTGGAGGATTTAACTGTGAAGATCACC 111

GK22/pET TGGCAGGTGGCTGTGTACTACTTAGATGAGTACCTATGCGGGATCTGA 224
mGK22 TGGCAGGTGGCTGTGTACTACTTAGATGAGTACCTATGCGGGATCTGA 167
mGK6 TGGCAAGTGGCTGTGTACCGCTTCACCAAATATCAATGTGGGTCTCGA 167
mGK9 TGGCATGTGGCTGTGTACCGTTACAACGAATATATATGCGGGTCTTGA 167
mGK26 TGGCAGGTGGCTGTGTACTACCAA AAGGAACACATTTGTGGGGTTCGTG 167

mGK2 2 /pE T C CGCAACTGGGTTC TCACAG CTGC CCAC TG CTATGAAGACAAGTATAATATTTGGC 2 80
mGK2 2 C CGCAACTGGGTTC TCACAG CTGC CCAC TG CTATGAAGACAAGTATAATATTTGGC 22 3
mGK6 CGCCAACTGGGTTCTCACAGCTGCCCACTGCCATAATGACATCGTTGC 223
mGK9 TGCCAACTGGGTTCTCACAGCTGCCCACTGCTATTACGAAGACAGTCC 223
mGK26 CCGCAACTGGGTTCTCACAGCTGCCCACTGCTATGTCGACATGGTTGC 223

mGK22/pET TGGGCAAAAACAAGCTATTCCAAGATGAACCCTCTGCTCAGCACATGCG 336
mGK22 TGGGCAAAAACAAGCTATTCCAAGATGAACCCTCTGCTCAGCACATGCG 279
mGK6 TGGGCAAAAACAACTTTTTGGAGGATGAACCCTCTGCCCAACACGTGCG 279
mGK9 TAGGAAAAAACAAC CTATACGAAGAGGAAC CCTCTGCTCAGCAC CGATTGGTCAGC 27 9
mGK26 TGGGCAAAAACAAGTTATTCCAAGAGGAACCCTCTGCTCAGCACATGCG 279

mGK2 2/pET AAAAGCTTCCCTCATCCTGACTTCAACATGAGCCTCCTCCAAGTT CC 392
mGK2 2 AAAAGC TT C CCTCATC CTGACTTCAACATGAG C CTC CT CCAAAGTGTA CC 335
mGK6 AAAGCCATCCCTCACCCTGACTTCAACATGAGCCTCCTGAATACCCCAA 335
mGK9 AAAAGCTTCCTTCACC CTGGCTACAACAGGAGCCTC CATAGAAACCACATCCGACA 33 5
mGK2 6 AAAAGC TT C CCTCAC CCTGG CTTCAACATGAG C CTC CTGATG CTTCAAACAACAC C 33 5

mGK22/pET TACTGGGGCCGACTTAAGCAATGACCTGATGTGCTCGCCTCACAGCCGT 448
mGK22 TACTGGGGCCGACTTAAGCAATGACCTGATG TGCTCGCCTCACAGCCGT 391
mGK6 ACCTGAGGATGACTACAGCAATGACCTGATGTGCTCGCCTCAAAGCCGT 391
mGK9 TCCTGAGTATGACTACAGCAATGACCTGATGTGCTCGCCTCACAGCCGT 391
mGK26 TCCTGGGGCTGACTTCAGCAATGACCTGATGTGCTCGCCTCACAGCCGT 391

mGK22/pET ACATCACAGATGTTGTGAAGCCCATCGACCTGCCCACTACGACAGTGG 504
mGK22 ACATCACAGATGTTGTGAAGCCCATCGACCTGCCCACTACGACAGTGG 447
mGK6 ACATCACAGATGTTGTGAAGCCCATCGACCTGCCCACTGAGACAGTGG 447
mGK9 ACATCACAGATGTTGTGAAGCCCATCGCCCTGCCTACTGAGACAGTGG 447
mGK26 ACATCACAGATGTTGTGAAGCCCATCGCCCTGCCCACAAAGACAGCGG 447

Figure 3-2. Alignment of nucleotide sequences from cDNA library screening. The
nucleotide sequences obtained from the cDNA library screening were aligned
for comparison. The sequences donated by the pET200 D-topo plasmid are in
red. The Pre-pro zymogen peptide nucleotide sequences are represented in
blue. The nucleotide regions with high variability are represented in orange.
The identities of the sequences are: mGK22/pET- PCR product using
glandular kallikrein22 specific primers and cDNA~ from NOD.B 10.H2b
submandibular tissue as a template ligated into pET200/D-topo plasmid,
mGK22- Mus musculus glandular kallikrein 22(AAN78419.1), mGK6- Mus
musculus glandular kallikrein 6(NP_034769.4), mGK9- Mus musculus
glandular kallikrein 9(NP_034246.1), mGK26- Mus musculus glandular
kallikrein 26(NP_034774.1).












mGK22/pET AGCACATGCCTAGCCTCAGGCTGGGGCAGCATTAACCAGTTATCAACC 560
mGK22 AGCACATGCCTAGCCTCAGGCTGGGGCAGCATTAACCAGTTATCAACC 503
mGK6 AGCACATGCCTAGCCTCAGGCTGGGGCAGCATTACACCGTAATAGAAAC 503
mGK9 AGCACATGCCTTGCCTCAGGCTGGGGCAGCACTACACCTTTAGCAATC 503
mGK26 AGCACATGCCTAGCCTCAGGCTGGGGCAGCATTACACCCACATGAAGC 503

mGK2 2 /pET AAATGATCTCCAGTGTGTGTCCATCAAGCTCCATCCTAATGAGCGGGAG 616
mGK2 2 AAATGATCTCCAGTGTGTGTCCATCAAGCTCCATCCTAATGAGCGGGAG 5 59
mGK6 AGATGAGCTCCAGTGTGTGAACCTCAAGCTCCTGCCTAATGGATGCAG 559
mGK9 AAAAGATCTCCAGTGTGTGAACCTCAAGCTCCTGCCTAATGAGCGGCAG 559
mGK26 AGATGATCTTCAGTGTGTGTTCATCACGCTCCTCCCCAATGACTGCAG 559

mGK2 2 /pE T CCCATATACTGAAGGTGACAGATGTCATGC TGTGTG CAGGAGAGATGAATGGAGGC 67 2
mGK2 2 CCCATATACTGAAGGTGACAGATGTCATGC TGTGTG CAGGAGAGATGAATGGAGGC 61 5
mGK6 CCCACATAGAGAAGGTGACAGATGACATGTTGTGTGCAGGATTGTGGC 615
mGK 9 CCCACATAGAGAAGGTGACAGATGTCATGC TGTGTG CAGGAGAGACAGATGGAGGC 61 5
mGK26 TCTACCTACAGAAAGTCACAGATGTCATGCTGTGTGCAGGAGTGTGGC 615

mGK22/pET AAAGACACTTGTAAGGGAGACTCAGGAGGCCCACTGATCTGTGTGGTTC 728
mGK22 AAAGACACTTGTAAGGGAGACTCAGGAGGCCCACTGATCTGTGTGGTTC 671
mGK6 AAAGACACTTGTGCGGGTGACTCAGGAGGCCCACTGATCTGTGTGGTTC 671
mGK9 AAAGACACTTGCAAGGGAGACTCAGGAGGCCCACTGATCTGTGTGGTTC 671
mGK26 AAAGACACTTGTGCGGGTGACTCCGGAGGCCCACTGATTTGTGTGATTC 671

mGK22/pET AGGTATCACATCATGGGGCTCTACCCCATGTGGTGAACCCATCCGCTT 784
mGK22 AGGTATCACATCATGGGGCTCTACCCCATGTGGTGAACCCATCCGCTT 727
mGK6 AGGTATCACATCATGGGGCCCTAGCCCTTGCGGT~AACCCAATTCGGAC 727
mGK9 AGGTATCACATCATGGGGCTTTACCCCATGTGGTGAACCCAAGCGCTT 727
mGK26 AGGAACCACATCAAATGGCCCTGAACCATGCGGTAAACCTGGTACGCTT 727

mGK22/pET ACACCAAACTTATTAAGTTTACCTCCTGGATAAA AGACACTAGCAACCTA 837
mGK22 ACACCAAACTTATTAAGTTTACCTCCTGGATAAA AGACACTAGCAACCTA 780
mGK6 ACACCAGAGTTTTAAATTTCAACACCTGGATAAGAGAA ACTTGCAATATA 786
mGK9 ACACCAAACTTATTAAGTTTACCTCCTGGATCAA AGACACTAGCAACTTA 786
mGK26 ACACCAACCTTATTAAGTTCAACTCCTGGATAAAA GATACTAGTAATCTA 786


Figure 3-2 Continued











mGK22/pET MRGSHHHHHHGMASMTGGQQMGRDLYDDDDKDHPFTAPPVQSIGFCK
mGK22 MRFLILFLTLSLGGIDAAPPVQSRILGGFKCEKN
mGK6 MRFLILFLALSLGGIDAAPPVQSRIVGGFNCEKN
mGK9 MRFLILFLALSLGGIDAAPPVHSRIVGGFKCEKN
mGK26 MWFLILFPALSLGGIDAAPPLQSRVVGGFNCEKN


o3re


mGK22/pET
mGK22
mGK6
mGK9
mGK26

mGK22/pET
mGK22
mGK6
mGK9
mGK26

mGK22/pET
mGK22
mGK6
mGK9
mGK26

mGK22/pET
mGK22
mGK6
mGK9
mGK26

mGK22/pET
mGK22
mGK6
mGK9
mGK26


SQPWQVAVYYLDEYLCGGVLLDRNWVLTAAHCYEDKYNIW LKLFDPA
SQPWQVAVYYLDEYLCGGVLLDRNWVLTAAHCYEDKYNIW LKLFDPA
SQPWQVAVYRFTKYQCGGILLNANWVLTAAHCHNDKYQVWLKNEDPA
SQPWHVAVYRYNEYICGGVLLDANWVLTAAHCYYEENKVSLKNYEPA
SQPWQVAVYYQKEHICGGVLLDRNWVLTAAHCYVDQYEVW LKLFEPA

QHRLVSKSFPHPDFNMSLLQSV PTGADLSNDLMLLRLSKPADITDVVKPID
QHRLVSKSFPHPDFNMSLLQSV PTGADLSNDLMLLRLSKPADITDVVKPID
QHRLVSKAIPHPDFNMSLLNEHTPQPEDDYSNDLMLLRLK KAIDVPD
QHRLVSKSFLHPGYNRSLHRNHIRHPEYDYSNDLMLLRLS KAIDVPA
QHRLVSKSFPHPGFNMSLLMLQTTPPGADFSNDLMLLRLS KAIDVPA

LPTTEPKLGSTCLASGWGSINQLIYQNPNDLQCVSIKLHPNVKAIVT
LPTTEPKLGSTCLASGWGSINQLIYQNPNDLQCVSIKLHPNVKAIVT
LPTEEPKLGSTCLASGWGSITPVKYEYPDELQCVNLKLLPNDAHIKT
LPTEEPKLGSTCLASGWGSTTPFKFQNAKDLQCVNLKLLPNDGAIKT
LPTKEPKPGSTCLASGWGSITPTRWQKSDDLQCVFITLLPNCAVLKT

DVMLCAGEMNGGKDTCKGDSGGPLICDGVLQGITSWGSTPCENPYTL
DVMLCAGEMNGGKDTCKGDSGGPLICDGVLQGITSWGSTPCENPYTL
DDMLCAGDMDGGKDTCAGDSGGPLICDGVLQGITSWGPSPCKNPYTV
DVMLCAGETDGGKDTCKGDSGGPLICDGVLQGITSWGFTPCEKPVTL
DVMLCAGEMGGGKDTCAGDSGGPLICDGILQGTTSNGPEPCKGPYTL

IKFTSWIKDTMAKNP
IKFTSWIKDTMAKNP
LNFNTWIRETMAEND
IKFTSWIKDTMAKNL
IKFNSWIKDTMMKNA


106
87
87
87
87

159
140
140
140
140

212
193
193
193
193

265
246
246
246
246

278
259
261
261
261


Figure3-3. Alignment of translated gene sequences from cDNA library Screening. The
nucleotide sequences obtained from the cDNA library screening were translated into
their corresponding amino acid sequences and aligned. The sequences donated by the
pET200 D-topo plasmid are in red. The Pre-pro zymogen peptide sequences are
represented in blue. The limits of the pre and pro regions are indicated by the arrows
located below the N-terminal region of the sequences. The mature glandular
kallikrein begins at amino acid position 25 (position 44 of mGK22/pET). The green
sequences represent the N-terminal region of the kallikrein proteins with the greatest
homology. The regions with high variability are represented in orange. The identities
of the sequences are: mGK22/pET- PCR product using glandular kallikrein22
specific primers and cDNA from NOD.B 10.H2b Submandibular tissue as a template
ligated into pET200/D-topo plasmid, mGK22- Mlus musculus glandular kallikrein
22(AAN78419.1), mGK6- Mlus musculus glandular kallikrein 6(NP_034769.4),
mGK9- Mlus musculus glandular kallikrein 9(NP_034246.1), mGK26- Mlus musculus
glandular kallikrein 26(NP_034774.1).


a3ro







40





mGK6/cDNA ATGAGGTTCCTGATCCTGTTCCTAGCCCTGTCCCTAGGAGGATTCGACC 58
mGK6 ATGAGGTTCCTGATCCTGTTCCTAGCCCTGTCCCTAGGAGGATTCGACC 58

mGK6/cDNA CTGTCCAGTCTCGAATTGTTGGAGGATTTAACTGTGAGAAATCAGCGCA 116
mGK6 CTGTCCAGTCTCGAATTGTTGGAGGATTTAACTGTGAGAAATCAGCGCA 116

mGK6/cDNA AGTGGCTGTGTACCGCTTCACCAA ATATCAATGTGGGGGTACTCGAGCCAA 174
mGK 6 AGTGGC TGTGTAC CGC TT CAC CAAATAT CAATGTGGGGGTAT C CTG CTGAACGC CAAC 17 4

mGK6/cDNA TGGGTTCTCACAGCTGCCCACTGCCATAATGACAAGTACCAGTGCGCAA 232
mGK6 TGGGTTCTCACAGCTGCCCACTGCCATAATGACAAGTACCAGTGCGCAA 232

mGK6/cDNA ACAACTTTTTGGAGGATGAACCCTCTGCCCAACACCGGCTTCAAAGACC 290
mGK6 ACAACTTTTTGGAGGATGAACCCTCTGCCCAACACCGGCTTCAAAGACC 290

mGK6/cDNA TCACCCTGACTTCAACATGAGCCTCCTGAATGAGCACACCAACTGGTAC 348
mGK6 TCACCCTGACTTCAACATGAGCCTCCTGAATGAGCACACCAACTGGTAC 348

mGK6/cDNA TACAGCAATGACCTGATGCCGCTCCGCCTCAAA AGCCTGCTGCTAAAGTG 406
mGK6 TACAGCAATGACCTGATGCTGCTCCGCCTCAAA AGCCTGCTGCTAAAGTG 406

mGK6/cDNA TGAAGCCCATCGACCTGCCCACTGAGGAGCCCAAGCTGGGGCATCTGCC 464
mGK6 TGAAGCCCATCGACCTGCCCACTGAGGAGCCCAAGCTGGGGCATCTGCC 464

mGK6/cDNA AGGCTGGGGCAGCATTACACCCGTCAAATATGAATACCCAGTACCGGGG 522
mGK6 AGGCTGGGGCAGCATTACACCCGTCAAATATGAATACCCAGTACCGGGG 522

mGK6 /c DNA AAC CTCAAGC TC CTGC CTAATGAGGACTGTGC CAAAGC CCACATAGAGAAGGTGACAG 5 80
mGK 6 AAC CTCAAGC TC CTGC CTAATGAGGACTGTGC CAAAGC CCACATAGAGAAGGTGACAG 5 80

mGK6 /c DNA ATGACATGTTGTGTGCAGGAGATATGGATGGAGG CAAAGACACTTGTG CGGGTGAC TC 63 8
mGK 6 ATGACATGTTGTGTGCAGGAGATATGGATGGAGG CAAAGACACTTGTG CGGGTGAC TC 63 8

mGK6/cDNA AGGAGGCCCACTGATCTGTGATGGTGTTCTCCAAGGTATCATTGGCTGC 696
mGK6 AGGAGGCCCACTGATCTGTGATGGTGTTCTCCAAGGTATCATTGGCTGC 696

mGK6/cDNA CCTTGCGGTAAACCCAATGTGCCGGGTATCTACACCAGAGTTATTCCCT 754
mGK 6 C CTTGCGGTAAAC CCAATGTGC CGGGTATC TACAC CAGAGTTTTAAATTT CAACAC CT 7 54

mGK6/cDNA GGATAAGAGAAACTATGGCTGAAAATGACTGA 786
mGK6 GGATAAGAGAAACTATGGCTGAAAATGACTGA 786

Figure 3-4. Alignment of nucleotide sequences of mGK6 from screening of
NOD.B 10.H2b submandibular tissue cDNA library and Mus musculus mGK6
(BC010754). The nucleotide sequence obtained from the cDNA library
screening was aligned with M~us musculus mGK6 (BC010754) for
comparison. The sequence identities are: mGK6/cDNA- the nucleotide
sequence obtained from the cDNA library screening; mGK6- Mus musculus
glandular kallikrein 6 (BC010754). The sequences are identical with the
exception of a single nucleotide discrepancy, which is represented in red and
indicate~dby an arrovy.













mGK6/cDNA MRFLILFLALSLGGIDAAPPVQSRIVGGFNCEKNSQPWQVAVRTYCG 53
mGK6 MRFLILFLALSLGGIDAAPPVQSRIVGGFNCEKNSQPWQVAVRTYCG 53

mGK6/cDNA LLNANWVLTAAHCHNDKYQVWLGKNNFLEDEPSAQHRLVSKAPPFML 106
mGK6 LLNANWVLTAAHCHNDKYQVWLGKNNFLEDEPSAQHRLVSKAPPFML 106

mGK6/cDNA NEHTPQPEDDYSNDLMPLRLKKPADITDVVKPIDLPTEEPKLSCAGG 159
mGK6 NEHTPQPEDDYSNDLMLLRLKKPADITDVVKPIDLPTEEP KLSCAGG 159

mGK6/cDNA ITPVKYEYPDELQCVNLKLLPNEDCAKAHIEKVTDDMLCA GDDGDCG 212
mGK6 ITPVKYEYPDELQCVNLKLLPNEDCAKAHIEKVTDDMLCA GDDGDCG 212

mGK6/cDNA SGGPLICDGVLQGITSWGPSPCGKPNVPGIYTRVLNFNTWIEMN 260
mGK6 SGGPLICDGVLQGITSWGPSPCGKPNVPGIYTRVLNFNTWIEMN 260

Figure 3-5. Alignment of amino acid translation of the mGK6 nucleotide sequences from
screening of NOD.B10.H2b submandibular tissue cDNA library and amino
acid sequence of Mus musculus mGK6 (BC010754). The nucleotide sequence
obtained from the cDNA library screening was translated into the
corresponding amino acid sequence and aligned with M~us musculus mGK6
(BC010754) amino acid sequence for comparison. The sequence identities
are: mGK6/cDNA- the amino acid translation of the nucleotide sequence
obtained from the cDNA library screening; mGK6- M~us musculus glandular
kallikrein 6(BC010754) amino acid sequence. The sequences are identical
with the exception of a single amino acid discrepancy, which is represented in
red and indicated by an arrow.







42





mGK26/cDNA ATGTGGTTCCTGATCCTGTTCCCAGCCCTGTCCCTAGGAGGATTCGACC 58
mGK26 ATGTGGTTCCTGATCCTGTTCCCAGCCCTGTCCCTAGGAGGATTCGACC 58

mGK2 6/cDNA CTCTCCAGTCTCGGGTGGTTGGAGGATTTAACTGTGAGAAATCACCGCA 116
mGK2 6 CTCTCCAGTCTCGGGTGGTTGGAGGATTTAACTGTGAGAAATCACCGCA 116

mGK26/cDNA GGTGGCTGTGTACTACCAAA AGGAACACATTTGTGGGGTGCTTGGACGCA 174
mGK26 GGTGGCTGTGTACTACCAAA AGGAACACATTTGTGGGGTGCTTGGACGCA 174

mGK26/cDNA TGGGTTCTCACAGCTGCCCACTGCTATGTCGACCAGTATGAGTGCGCAA 232
mGK26 TGGGTTCTCACAGCTGCCCACTGCTATGTCGACCAGTATGAGTGCGCAA 232

mGK26/cDNA ACAAGTTATTCCAAGAGGAACCCTCTGCTCAGCACCGATTGTACAGTCC 290
mGK26 ACAAGTTATTCCAAGAGGAACCCTCTGCTCAGCACCGATTGTACAGTCC 290

mGK26/cDNA TCACCCTGGCTTCAACATGAGCCTCCTGATGCTTCAAACAACTCGGCGC 348
mGK26 TCACCCTGGCTTCAACATGAGCCTCCTGATGCTTCAAACAACTCGGCGC 348

mGK26/cDNA TTCAGCAATGACCTGATGCTGCTCCGCCTCAGCAAGCCTCGACCGTTTG 406
mGK26 TTCAGCAATGACCTGATGCTGCTCCGCCTCAGCAAGCCTCGACCGTTTG 406

mGK26/cDNA TGAAGCCCATCGCCCTGCCCACAAAGGAGCCCAAGCCGGGGAGAAGCACT 464
mGK26 TGAAGCCCATCGCCCTGCCCACAAAGGAGCCCAAGCCGGGGAGAAGCACT 464

mGK26/cDNA AGGCTGGGGCAGCATTACACCCACAAGATGGCAAAA GTCAGAGTTCGGGG 522
mGK26 AGGCTGGGGCAGCATTACACCCACAAGATGGCAAAA GTCAGAGTTCGGGG 522

mGK26/cDNA TTCATCACGCTCCTCCCCAATGAGAACTGTGCCAAAGTCTACCGAGTCG 580
mGK26 TTCATCACGCTCCTCCCCAATGAGAACTGTGCCAAAGTCTACCGAGTCG 580

mGK26/cDNA ATGTCATGCTGTGTGCAGGAGAGATGGGTGGAGGCAAAGACTGGGGGAC 638
mGK26 ATGTCATGCTGTGTGCAGGAGAGATGGGTGGAGGCAAAGACTGGGGGAC 638

mGK26/cDNA CGGAGGCCCACTGATTTGTGATGGTATTCTCCAAGGAACCATCAGCCGA 696
mGK26 CGGAGGCCCACTGATTTGTGATGGTATTCTCCAAGGAACCATCAGCCGA 696

mGK26/cDNA CCATGCGGTAAACCTGGTGTACCAGCCATCTACACCAACCTATGTCCCT 754
mGK26 CCATGCGGTAAACCTGGTGTACCAGCCATCTACACCAACCTATGTCCCT 754

mGK26/cDNA GGATAAAAGATACTATGATGAAAATGCCTGA 786
mGK26 GGATAAAAGATACTATGATGAAAATGCCTGA 786

Figure 3-6. Alignment of nucleotide sequences of mGK26 from screening of
NOD.B 10.H2b submandibular tissue cDNA library and Mus musculus mGK26
(NM_010644). The nucleotide sequence obtained from the cDNA library
screening was aligned with Mus musculus mGK26 (NM_010644) for
comparison. The sequence identities are: mGK26/cDNA- the nucleotide
sequence obtained from the cDNA library screening; mGK26- M~us musculus
glandular kallikrein 26(NM_010644). The sequences are identical, with the
exception of a single nucleotide discrepancy, which is represented in red and
indicate~dby an arrovy.












mGK26/cDNA MWFLILFPALSLGGIDAAPPLQSRVVGGFNCEKNSQPWQVAVYKHCG 53
mGK26 MWFLILFPALSLGGIDAAPPLQSRVVGGFNCEKNSQPWQVAVYKHCG 53

mGK26/cDNA LLDRNWVLTAAHCYVDQYEVWLGKNKLFQEEPSAQHRLVS KSPPFML 106
mGK26 LLDRNWVLTAAHCYVDQYEVWLGKNKLFQEEPSAQHRLVS KSPPFML 106

mGK26/cDNA MLQTTLPGADFSNDLMLLRLSKPADITDVVKPIALPTKEP KPSCAGG 159
mGK26 MLQTTPPGADFSNDLMLLRLSKPADITDVVKPIALPTKEP KPSCAGG 159

mGK26/cDNA ITPTRWQKSDDLQCVFITLLPNENCAKVYLQKVTDVMLCA GEGGDCG 212
mGK26 ITPTRWQKSDDLQCVFITLLPNENCAKVYLQKVTDVMLCA GEGGDCG 212

mGK26/cDNA SGGPLICDGILQGTTSNGPEPCGKPGVPAIYTNLIKFNSWIDMK 260
mGK26 SGGPLICDGILQGTTSNGPEPCGKPGVPAIYTNLIKFNSWIDMK 260

Figure 3-7. Alignment of amino acid translation of the mGK26 nucleotide sequences
from screening of NOD.B10.H2b submandibular tissue cDNA library and
amino acid sequence of2~us musculus mGK26 (NP_034774.1). The
nucleotide sequence obtained from the cDNA library screening was translated
into the corresponding amino acid sequence and aligned with M~us musculus
mGK6 (NP_034774.1) amino acid sequence for comparison. The sequence
identities are: mGK26/cDNA- the amino acid translation of the nucleotide
sequence obtained from the cDNA library screening; mGK26- M~us musculus
glandular kallikrein 26 (NP_034774.1) amino acid sequence. The sequences
are identical with the exception of a single amino acid discrepancy, which is
represented in red and indicated by an arrow.











1 MP.GSHHHHHH GI-LLSMTGGQQ MGRDLYDDDD KDHPF L~PPV CRILGGFKC

51 ERESQPWQVA VYYLDEYLCG G~VLLDPNTWVL TAAHCYEDKY N~I~LGENKLF

101 QDEPSAQHRIL VSKSFPHPDF NH~SLLQSVPT GAD>LSHDLML LRL~SKPADIT

151 DVV~KPIDLPT TEPKLGSTCL ASGWGSINQL IYQNPNDLQC SILILHPITV

201 CVKAHIKIFT DVHMLCAGEHH~ IGKDTCKGDS GGPLICDGVL CGITS~WlGTP

251 CGEPNAPAPIY TKLIEFTSlIl EDT-LLE 277

Figure3-8. Translated amino acid sequence mGK22/pET200 D-topo plasmid construct.
The figure displays the amino acid translation of the DNA sequence of the
pET200 D-topo plasmid that contains mGK22 PCR product as an insert. The
rectangular box outlines the leader sequence donated by the pET200 D-topo
plasmid as indicated by the 6X histidine tag beginning at the 5th amino acid
position. The mGK22 PCR product amino acid sequence begins at amino acid
position 37. Mature mGK22 begins at amino acid position 44 which
corresponds to amino acid position 24 of the published mGK22 amino acid
sequence.


















750 bp-


lane 1 2 3 4 5
Figure 3-9. Results of gel electrophoresis analysis of PCR amplification using mGK22
specific primers. lane 1- 1Kb ladder molecular weight marker; lane2- mGK22
specific primers with first strand synthesis cDNA as a template; lane3-
mGK22 specific primers with transformed cDNA as a template; lane4-
mGK22 specific primers with no DNA(negative control); lane5- control
primers and control template DNA (Invitrogen positive control).


















8 Kp~
7 Khp~
6 Khp~
5 Khp~






lane 1 2 3 4
Figure3-10. Results of gel electrophoresis analysis of plasmid constructs. lanel1- 1Kb
ladder molecular weight marker; lane2- pET200 D-topo plasmid with mGK22
insert; lane3- pET200 D-topo plasmid with PCR (lac Z) positive control;
lane4- empty pET200 D-topo plasmid.





















lane 1 2 3 4 5
Figure3-1 1. Results of gel electrophoresis analysis of hindIII digested plasmid constructs.
lanel- 1Kb ladder molecular weight marker; lane2- mGK22 without hindIII;
lane3- PCR (lac Z) positive control; lane4- mGK22 with hind III; lane5-
empty pET200 D-topo plasmid with hind III.
















lane 1 2345 67 8 91 1
Fiur 312 Rsutso SD-PG alyis. an comsi b ue tiigo rd hl



idcin;lane 3- unnue 1. hor lan 4- induce 1.0 hor;lne5


uninduced 2.0 hour; lane 6- induced 2.0 hour; lane 7- uninduced 3.0 hour;
lane 8- induced 3.0 hour; lane 9- uninduced 5.0 hour; lane 10- induced 5.0
hour; lanell- SDS PAGE molecular weight standards.






49





37kl)~


25k~D-


lane 1 2 3 4 5

Figure 3-13. Analysis of crude whole cell lysates induced with IPTG using western blot
analysis with anti-6Xhistidine antibody. lane 1- 0.0 hours (pre-induction); lane
2- 1.0 hour post induction; lane 2- 2.0 hour post induction; lane 4- empty; lane
5- histidine tagged positive control.







50







37kD- -- -


25kD-






lane 1 2 3 4 5 6

Figure3-14. Analysis of pellet and supernatant of crude lysates induced with IPTG using
western Blot analysis with anti-6Xhistidine antibody. lane 1- 0.0 hour (pre-
induction); lane 2-empty; lane3- pellet 1.0 hour post-induction; lane 4-
supernatant 1.0 hour post-induction; lane 5- pellet 2.0 hour post-induction;
lane 6- supernatant 2.0 hour post-induction.







51


LYSIS BUFFER & PEPTIDE
1 00

0 80


0.60




0 20.I




U 2 0 4 0 6.0 8 0 10 0 12.0 14 0 16 0 18.0 2[
Minutes
Figure3-15. HPLC-PSP assay chromatogram of 40pL of PSP peptide incubated with
40pL of lysis buffer. The uncleaved, intact PSP peptide retention time was
13.46 minutes.










XP52 NOJCD L JIS BUFFERP


0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 1.
Minutes

Figure3-16. HPLC-PSP assay chromatogram of 40CIL of PSP peptide incubated with 5
CIL of NOD.B 10.H2b and 3 5 CL of lysis buffer. The retention times of the
fragments of the cleaved PSP peptide were 9. 16 minutes (peak 6) and 12.21
minutes peakk7.







53















0 .0 4D 6.0 8.0 10D 12.0 14D 16D 18.0
Minutes
Figure3-17. HPLC-PSP assay chromatogram of 40pL of PSP peptide incubated with 40
pL of crude whole cell lysates from the mGK22 expression. This assay was
performed on whole cell lysates prior to centrifugation. The retention time of
the uncleaved, intact PSP peptide was 13.58 minutes.
















1 r I I I I *l l l I I I I I I IIII I 1 | 1 1 I I
0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0
Figure3-18. HPLC-PSP assay chromatogram of 40CIL of PSP peptide incubated with 40
CIL of supernatant of the crude cell lysates from the mGK22 expression. The
retention time of the uncleaved, intact PSP peptide was 13.45 minutes.


0.800
0.600-


4 0.400+


0.0 -
0-"C


I















CHAPTER 4
DISCUSSION

The identity of the unknown PSP proteolytic enzyme present in the salivas of the

NOD mice exhibiting Sj S symptoms remains an elusive target. Although previous data

have pointed to mGK22 as the PSP cleaving culprit in the NOD saliva, the data from the

cDNA library screening of the NOD have revealed the presence of other candidate

proteins with known functional similarity and genetic homology to mGK22 as present in

the submandibular glands of the NOD mouse. A maj ority of these proteins identified in

the cDNA library screening belong to the same kallikrein family of proteins that includes

mGK22 such as: mGK6, mGK9, mGK26. However, there were a few non-kallikrein

genes that were identified by the screening, e.g. salivary protein 1 and 2, PRL-inducible

protein, cytochrome P450.

The cDNA library screening failed to detect mGK22 in the cDNA library. However

that could be the result of a sample screening whose size was too small to be

representative of the complete cDNA library. Calculations of the screening confirm that

approximately 6500 colonies of the 6.6 X 105 independent clones generated in the cDNA

library were screened or about 1% of the library. Possibly, the abundance of mGK22 in

the cDNA library was not great enough for it to be detected by the screening. Thus, the

small fraction of the cDNA library that was screened could have been the reason for the

failure to detect mGK22. Further screening may detect the presence of a gene with a low

copy number such as mGK22.









Although the cDNA library screening failed to detect mGK22, the PCR

amplification, using mGK22 specific primers, did detect mGK22 in the cDNA library.

Moreover, the PCR amplification showed the presence of two different forms of mGK22

in the cDNA library. One of the PCR reactions produced a sequence that was identical to

the published mGK22 sequence while another, separate PCR reaction, produced a

sequence that contained a one base pair deviation from the published mGK22 sequence.

When the sequence was translated to its corresponding amino acid sequence, the base

pair substitution resulted in an aspartic acid for asparagine amino acid substitution. This

amino acid substitution is consistent with the N-terminal amino acid sequence identified

from a unique protein band of a purified saliva fraction shown to have PSP cleaving

activity from previous research (Day, 2002). The two forms could represent alleles of the

mGK22 gene or the existence of an unidentified glandular kallikrein gene. Errors

generated during the PCR process are unlikely, since the protein encoded by this allele

was identified in mice (Day, 2002). Further sequencing of genomic DNA in the NOD

mouse should be performed to detect the presence of both forms of the mGK22 gene.

Genomic DNA sequencing of the NOD mouse would help to understand the nature of the

mGK22 discrepancies that have been found.

The cDNA library screening showed the presence of mGK6, mGK9, and mGK26

in the NOD saliva (Table 1). The N-terminal amino acid sequences of mGK6, mGK9,

and mGK26 share a high degree of homology with mGK22 (figure 3-2 and 3-3). This

homology between these GKs and mGK22 may explain why they were detected with an

oligonucleotide probe that was complementary to the back-translated sequence of the N-

terminus of mGK22.









The functional similarity and homology of mGK9, mGK22, and mGK26 strongly

supports their candidacy as the enzyme responsible for PSP cleavage in the NOD saliva

from previous research. Considering that mGK9, mGK22, and mGK26 share functional

similarities by recognizing, binding, and processing pre-EGF, it is probable that all three

kallikreins could similarly bind and cleave PSP, if indeed mGK22 possesses that ability.

However, it should be noted that mGK22's PSP cleaving activity, at least in the pro-

mGK22 form, has yet to be proven. Therefore, future research should be broadly directed

towards the isolation and assessment of mGK9, mGK22, and mGK26 for PSP cleaving

activity should the mGK22 form not show activity.

The cDNA library screening also revealed the presence of some interesting base

pair polymorphisms of mGK6 (figures 3-4 and 3-5) and mGK26 (figures 3-6 and 3-7)

that make them highly suspect as PSP cleaving enzymes. The mutations were identified

in several different screenings further supporting the conclusion that these nucleotide

substitutions are indeed mutations and not random base pair substitutions due to

transcribing or PCR replication errors. Sequencing of genomic DNA of the NOD mouse

could be used to characterize the base pair polymorphisms of mGK6 and mGK26 found

in the cDNA library screening as mutations. These mutations could result in a gain of

function, resulting in PSP cleaving activity for mGK6 and mGK26. The mutations of

mGK6 and mGK26, identified by the cDNA library screening, resulted in amino acid

substitutions that involved a proline residue, when translated. The cyclic orientation and

nature of a proline residue could be indicative of a beta turn or some other great structural

deviation from the wild type isoform of the protein. Moreover, the mutations could

confer a three-dimensional conformational change in mGK6 and mGK26 such that a PSP









cleaving function is gained. Further characterization by isolation or purification is

required to ascertain the changes in activity, if any, caused by these mutations.

The cDNA library screening results generated from this thesis showed several

candidate genes that may be responsible for the cleaving PSP in the NOD saliva. The

proof of PSP cleaving ability and further binding characterization by these genes or

mutations could help to confirm that one or more of them encode the protein responsible

for PSP cleavage in the NOD saliva observed by researchers.

The data of the expression of mGK22 from this thesis shows that an inducible

6Xhistidine-tagged recombinant protein with a molecular mass equal to the mass of

mGK22 was produced. The crude extract of the cells that were transformed to produce

mGK22 were assessed for PSP cleavage activity after they were shown to produce an

inducible protein with a 6X histidine tag.

Although the crude extract had no PSP cleaving activity, mGK22 may still be the

PSP protease. The inability of the crude extract, which contained mGK22, to cleave PSP

may be the result of interference from some of the transformed bacterial cell constituents.

Further purification could be performed to remove cell constituents that may be

interfering with mGK22 PSP cleaving activity. Another technique would be to analyze

the NOD saliva for PSP cleaving activity while in the presence of mGK22 cell extract

that has been previously shown to be inactive. This technique would show if the secreted

form of mGK22 was affected by some other component present in the cell extract

resulting in inactivity of the PSP cleaving protein. Furthermore, the cell extracts ability to

inhibit PSP cleaving activity in the NOD saliva could serve as a possible explanation for

the crude extracts inability to cleave PSP.









Another possible explanation for the crude extracts inability to cleave PSP could be

that mGK22 was expressed in an inactive form due to the presence of the 7 amino acid

pro sequence on the N-terminus of mGK22. The pro-sequence on the N-terminus of the

mGK22 that was expressed in the transformed bacterial cells could be the cause of the

inactivity of mGK22. Future experiments could be designed to express mGK22 in the

mature form, without the 7 amino acids that comprise the pro-sequence on the N-

terminus of mGK22.

Alternatively, the 6X histidine tag could have interfered with mGK22's activity.

Activation by enzymatic cleavage of the recombinant protein expressed in this research

could confer PSP cleaving activity in the crude extract. Trypsin cleavage at amino acid

position 24 is thought to be important and necessary step in the activation of GKs (Matsui

et al., 2000). Perhaps activation via cleavage with trypsin could confer PSP cleaving

activity in the crude lysates.

Moreover, the reformation of the disulfide bridges present in mGK22 could have

been inhibited by the prokaryotic expression system thus preventing correct refolding and

preventing the formation of required three-dimensional structures. Inappropriate

refolding and incorrect three-dimensional conformation of the protein could induce the

formation of inclusion bodies. The inclusion bodies are insoluble aggregates of proteins

in E. coli. The analysis of the insoluble pellet and soluble supernatant of the crude lysates

showed that the protein expressed was present mostly in the pellet of the crude lysate

(figure 3-14). This suggests that although mGK22 was expressed in the transformed

bacterial cells, much of it was probably in an insoluble form, which could result in

expression of an inactive protein.










Also, post-translational modifications may be required for PSP cleaving activity in

mGK22. Post-translational modifications such as glycosylation, which may be necessary

for PSP cleaving activity, do not occur in a prokaryotic expression system. The

expression of mGK22 in a eukaryotic expression system would produce a complete

mGK22 with great similarity to the secreted form of mGK22 produced in the NOD

saliva.

The data generated by this research has shed some light into the identity of a PSP

cleaving protease present in the saliva of the NOD mouse. Although mGK22 cannot be

ruled out completely, it is less likely that it is the PSP protease, which suggests that the

other kallikrein genes and possibly kallikrein mutants that have been identified to be

present in the NOD cDNA library may be the PSP protease. Further isolation,

purification, and characterization of these other genes and mutants, as well as mGK22,

are required to assess their ability to cleave PSP. However, confirmation ofPSP cleavage

by the candidate genes named in this thesis does not prove that they are solely

responsible for the PSP cleavage in NOD saliva previously observed. The possibility that

another unknown protein is responsible for the observed PSP cleavage in the NOD saliva

cannot be ruled out yet. Purification and isolation, via ammonium sulfate precipitation for

example, directly from the saliva of the NOD mouse could identify the protease

responsible for PSP cleavage in the NOD saliva. Finally, once identified molecular

techniques such as site-directed mutagenesis and gene knockout mice could be developed

to further characterize the nature of the PSP cleavage reaction and its possible role in Sj S

pathogenesis.
















LIST OF REFERENCES


Atkinson JC, Fox PC, Travis WD, Popek E, Katz RW, Balow JE, Pillemar SR. IgA
rheumatoid factor and IgA containing immune complexes in primary Sjiigren's
syndrome. JRheumatol 1989; 16:1205-1210.

Aziz KE, McCluskey PJ, Wakefield D,. Characterisation of follicular dendretic cells in
labial salivary glands of patients with primary Sjiigren' s syndrome: comparison
with tonsillar lymphoid follicles. Ann Rheum 1997; 56: 104-143.

Baum BJ, Dai Y, Hiramatsu Y, Horn VJ, Ambudkar IS. Signaling mechanisms that
regulate saliva formation. Crit Rev Oral Biol2~ed 1993; 4:379-3 84.

Blaber M, Isackson PJ, Bradshaw A, A complete cDNA sequence for the maj or
epidermal growth factor binding protein in the male mouse submandibular gland.
Biochemistry 1987; 26:6742-6749.

Brayer J, Lowry J, Cha S, Robinson CP, Yamachika S, Peck AB, et al. Alleles from
chromosomes 1 and 3 of NOD mice combine to influence Sjiigren's syndrome--ike
autoimmune exocrinopathy. JRheumatol 2000; 27: 1896-1904.

Bymaster FP, McKinzie DL, Felder CC, Wess J. Use of M1-M5 muscuranic receptor
knockout mice as novel tolls to delineate the physiological roles of the muscuranic
cholinergic system. Neurchem Res 2003; 28:437-442.

Cha S, Peck AB, Humpherys-Beher MG. Progress in understanding autoimmune
exocrinopathy using the non-obese diabetic mouse: an update. Crit Rev Oral Biol
M~ed 2002; 13(1):4-16.

Carnaud C, Legrand B, Olivi M, Peterson LB, Wicker LS, Bach JF. Acquired allo-
tolerance to maj or histocompatiblity antigens indifferently contributes to
preventing diabetes development in non-obese diabetic (NOD) mice. JAutoimmun
1992; 5:591-601.

Day, J M. Development of an HPLC based assay used for the characterization and
identification of an unknown protease present in the saliva of the NOD.B10.H2b
mouse model. University ofFlorid'a 2002.

Delaleu N, Jonsson R, Koller MM. Review Sjiigren's syndrome. Eur J Oral Sci 2005;
113:101-113.










Drinkwater CC, Evans BA, Richards R I. Genes encoding epidermal growth factor
binding proteins. Biochemistry 1987; 26:6750-6756.

Evans BA, Drinkwater CC, and Richards RR. Mouse glandular kallikrein genes. JBiol
Chent 1987; 262:8027-8034.

Geetha C, Venkatesh SG, Fasciotto Dunn BH, Gorr SU. Expression and anti-bacterial
activity of human parotid secretary protein (PSP). Biochemical Society
Transactions 2003; 31,815-818.

Gomis-Ruth FX, Bayes A, Soitiropoulou G, Pampalkis G, Tsetsenis T, Villegas V, Aviles
FX, and Coll M. The structure of human prokallikrein 6 reveals a novel activation
mechanism for the kallikrein family. JBiol Chent 2002; 277:27273-27281.

Matsui H, Moriyama A, Takahashi T. Cloning and characterization of mouse klk27, a
novel tissue kallikrein expressed in testicular Leydig cells and exhibiting
chymotrypsin-like specificity. Eur JBiochent 2000; 267:6858-6865.

Olsson AY, Lundwall A. Organization and evolution of the glandular kallikrein locus in
M~us nauscuhts. Biochemical and' Biophysical Research Conanunications 2002;
299:305-311.

Robinson CP, Cornelius J, Bounous DE, Yamamoto H, Humpherys-Beher MG, and Peck
AB. Characterizations of the changing lymphocyte populations and cytokine
expression in the exocrine tissues of autoimmune NOD mice. Autoinanunity 1998;
27:29-44.

Robinson CP, Yuamamoto H, Humpherys-Beher MG, and Peck AB. Genetically
programmed development of the salivary gland abnormalities in the NOD (non-
obese diabetic)- scid mouse in the absence of detectable lymphocytic infiltration: a
potential trigger for sialoadentitis of NOD mice. Clinicallnantunol and'
Inanunopathol 1996; 79:50-59.

Robinson CP, Brayer J, Yamachika S, Esch TR, Peck AB, Stewart CA, Peen E, Johsson
R, Humpherys-Beher MG. Transfer of human serum IgG to non-obese diabetic
IgC1nui mice reveals a role for autoantibodies in the loss of secretary function of
exocrine tissues in Sjoigren' s syndrome. Proc NatlAcad'Sci USA 1998; 95:753 8-
7543.

Robinson CP, Bounous DI, Alford CE, Peck AB, Humpherys-Beher MG. Aberrant
expression and potential function for parotid secretary protein (PSP) in the non-
obese diabetic (NOD) mouse. Adv Exp M~edBiol 1998; 438:925-930.

Robinson CP, Yamachika S, Bounous DI, Brayer J, Jonsson R, Holmdahl R, et al. A
novel NOD-derived murine model of primary Sjoigren' s syndrome. At thliriti Rheuna
1998c; 41:150-156.









Takada K, Takiguchi M, Konno A, Inaba M. Autoimmunity against a tissue kallikrein in
IQI/Jic mice. JBiol Chem 2005; 280:3982-3988.

Voulgarelis M, Moutspoulos HM. Lymphoproliferation in autoimmunity and Sjoigren's
syndrome. Curr RheumtolRep 2003; 5:317-323.

Zeher M, Adany R, Nagy G, Gomez R, Szegedi G. Macrophage containing factor XIII
subunit a in salivary glands of patients with Sjoigren' s syndrome. Jlnvest Allergol
Clin Immunol 1991; 1:261-265.
















BIOGRAPHICAL SKETCH

My name is Javier Brian Alvarado and I am a 36 year old graduate student living in

Gainesville, Florida. I am the only child of Javier and Kathy Alvarado. I attended the

University of South Florida in Tampa, Florida, where I earned a bachelor' s degree in

biology in 1995.

After graduating, I expanded my work experience through employment in private

industry. I have had work experience as a laboratory analyst, researcher, teacher and

athletic coach over a period of 7 years.

In the future, I plan to reside in Gainesville, Florida, where I will raise my family

and endeavor to expand my career as a molecular biologist.




Full Text

PAGE 1

IN SEARCH OF A PAROTID SECRETO RY PROTEIN PROTEASE: A FOCUS ON GLANDULAR KALLIKREIN 22 By JAVIER BRIAN ALVARADO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Javier Brian Alvarado

PAGE 3

iii ACKNOWLEDGMENTS I wish to thank Dr. Ammon Peck for a llowing me to conduct research on this interesting project in his lab. He has taught me the value of good leadership and afforded me the opportunity to explore my own ideas and creative thinking. Al so, I wish to thank Dr. Smurti Killedar for her suppor t and scientific advisement. She is an intelligent, kind and honest person who has provided me w ith the background introduction for this exciting project. Dr. Seunghee Cha has also been a very important resource for SjgrenÂ’s Syndrome throughout this research. She has provided me with clinical insight and information as to the general nature of the SjgrenÂ’s Syndrome. I would like to thank Ms. Joyce Conners fo r her vigilant effort of keeping my administrative affairs on the right track. I woul d also like to thank Janet Cornelius for her management of the laboratory and general labor atory practical advice. I wish to thank Dr. Sally Litherland and her laboratory staff for their helpful advice and kindness. I would like to thank the enti re Peck Lab, both past and present. Specifically, I wish to thank Eric Singson, Jin Wang, Daniel Sa ban, Cuong Nguyen, and Jeff Olpako for their support, friendship, and technical expertise. Finally, I wish to thank my family for th eir support and faith in my abilities. My cousin, D. Chris Mclendon, deserves special thanks for being my best friend and a remarkable scientific mind. And most impor tantly, I wish to thank my mom and dad, Kathy and Javier Alvarado, for their undying support and unconditional love they have provided for me throughout my life

PAGE 4

iv TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF FIGURES...........................................................................................................vi ABSTRACT.....................................................................................................................vi ii CHAPTER 1 INTRODUCTION........................................................................................................1 SjgrenÂ’s Syndrome......................................................................................................1 NOD Mouse Model......................................................................................................4 Temporal Changes in Salivary Composition of NOD Mouse Model...........................6 Glandular Kallikrein...................................................................................................10 2 MATERIALS AND METHODS...............................................................................15 Mouse Models............................................................................................................15 cDNA Library Construction.......................................................................................15 cDNA Library Screening for mGK22........................................................................16 Analysis of Plasmid DNA by Gel Elect rophoresis and Enzyme Digestion...............18 Expression of Recombinant mGK22..........................................................................19 Protein Analysis..........................................................................................................20 SDS-PAGE..........................................................................................................20 Commassie Blue Staining....................................................................................20 Membrane Blotting..............................................................................................21 Western Blots......................................................................................................21 HPLC-PSP Assay.......................................................................................................22 PSP Peptide Synthesis.........................................................................................22 Detection of PSP Peptide Cleavage by HPLC....................................................23 3 RESULTS...................................................................................................................25 cDNA Library Screening for mGK22........................................................................25 Analysis of cDNA Cloned mGK22 and pET200 D-topo Constructs.........................27 Analysis of mGK22 Expression.................................................................................29 Commassie Blue Staining....................................................................................30 Western Blot Analysis.........................................................................................30 HPLC-PSP Analysis to Determine PSP Prot eolytic Activity in Crude Lysates.........32

PAGE 5

v 4 DISCUSSION.............................................................................................................55 LIST OF REFERENCES...................................................................................................61 BIOGRAPHICAL SKETCH.............................................................................................64

PAGE 6

vi LIST OF FIGURES Figure page 2-1 Amino acid sequence of Mus musculus parotid secretory protein...........................24 3-1 Photograph of replica-plated nitr ocellulose membrane hybridized with oligonucleotide probe from cD NA library screening...............................................35 3-2 Alignment of nucleotide sequen ces from cDNA library screening.........................37 3-3 Alignment of translated gene se quences from cDNA library Screening.................39 3-4 Alignment of nucleotide sequences of mGK6 from screening of NOD.B10.H2b submandibular tissue cDNA library and Mus musculus mGK6 (BC010754).........40 3-5 Alignment of amino acid translation of the mGK6 nucleotide sequences from screening of NOD.B10.H2b submandibular tissue cDNA library and amino acid sequence of Mus musculus mGK6 (BC010754)......................................................41 3-6 Alignment of nucleotide sequences of mGK26 from screen ing of NOD.B10.H2b submandibular tissue cDNA library and Mus musculus mGK26 (NM_010644)....42 3-7 Alignment of amino acid translation of the mGK26 nucleotide sequences from screening of NOD.B10.H2b submandibular tissue cDNA library and amino acid sequence of Mus musculus mGK26 (NP_034774.1)...............................................43 3-8 Translated amino acid sequence mG K22/pET200 D-topo plasmid construct.........44 3-9 Results of gel electrophoresis anal ysis of PCR amplification using mGK22 specific primers........................................................................................................45 3-10 Results of gel electrophoresis analysis of plasmid constructs..................................46 3-11 Results of gel electrophoresis analysis of hindIII digested plasmid constructs.......47 3-12 Results of SDS-PAGE analysis and co mmassie blue staining of crude whole cell lysates induced with IPTG with uni nduced crude whole cell lysates .....................48 3-13 Analysis of crude whole cell lysates induced with IPTG using western blot analysis with anti-6 Xhistidine antibody...................................................................49

PAGE 7

vii 3-14 Analysis of pellet and supernatant of crude lysates induced with IPTG using western Blot analysis with anti-6Xhistidine antibody..............................................50 3-15 HPLC-PSP assay chromatogram of 40L of PSP peptide incubated with 40L of lysis buffer...........................................................................................................51 3-16 HPLC-PSP assay chromatogram of 40L of PSP peptide incubated with 5 L of NOD.B10.H2b and 35L of lysis buffer...................................................................52 3-17 HPLC-PSP assay chromatogram of 40L of PSP peptide incubated with 40 L of crude whole cell lysates from the mGK22 expression.........................................53 3-18 HPLC-PSP assay chromatogram of 40L of PSP peptide incubated with 40 L of supernatant of the crude cell ly sates from the mGK22 expression......................54

PAGE 8

viii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science IN SEARCH OF A PAROTID SECRETO RY PROTEIN PROTEASE: A FOCUS ON GLANDULAR KALLIKREIN 22 By Javier Brian Alvarado August, 2006 Chair: Ammon B. Peck Major Department: Oral Biology SjgrenÂ’s Syndrome is an autoimmune disord er that leads to a decrease in saliva and tear production by affecting the saliv ary and lacrimal gland functions. The NOD mouse model is an animal model that exhibi ts an autoimmune e xocrinopathy similar to that of SjgrenÂ’s Syndrome, making it an ex cellent model for the study of SjgrenÂ’s Syndrome in humans. Previous research has observed that the salivas of NOD mice exhibiting SjgrenÂ’s Syndrome contain an unknown protease that ab errantly cleaves paro tid secretory protein at a specific N-terminal point containing the NL-NL amino acid sequence. In an effort to identify the unknown PSP cleaving protease in the NOD saliva, earlie r work determined, using purification and inhibiti on assays, the activity might be caused by an enzyme with strong homology to glandular kallikrein 22. The fo cus of the current research has been to test the idea that mGK22 is capable of cleaving PSP and to screen the cDNA library of

PAGE 9

ix NOD mouse submandibular gland tissue for othe r kallikrein-like proteins that may be capable of cleaving PSP. Screening of cDNA library provided evid ence that many kallikrein and mutant genes known to have similar function and homology with mGK22 ar e expressed in the NOD mouse submandibular gland tissue. The ho mologies and functional similarity of these other kallikreins and mu tant enzymes make them strong candidates as PSP cleaving enzymes. In addition it was possible to successfully clone mGK22 from NOD submandibular tissue and expre ss it in a prokaryotic cell line. However, the crude extract of the prokaryotic cells expressing mGK 22 showed no PSP protease activity, suggesting but not proving that GK22 may not be th e PSP proteolytic entity. The lack of PSP proteolytic activity of mGK22 could be the result of expr essing mGK22 in a bacterial, prokaryotic rather than an eukaryotic system. Further experimentation utilizing eukaryotic expression systems and enzyme purification directly from NOD saliva is probably necessary to further characteriz e mGK22 as well as identify if the PSP proteolytic activity is kallikrein mediated.

PAGE 10

1 CHAPTER 1 INTRODUCTION SjgrenÂ’s Syndrome (SjS) is an autoimmune di sorder that affects the salivary and lacrimal glands resulting in d ecreased saliva and tear produc tion. SjS is also characterized by mononuclear lymphocytic inf iltration of the lacrimal a nd salivary glands. Most SjS patients report the sensation of chroni c dry mouth (xerostomia) and dry eyes (keratoconjunctivitis sicca). Moreover, SjS pa tients report dryne ss of other mucosal surfaces such as lungs, gastrointestinal tract, vagina, and skin. However, these symptoms are subjective and can be the result of causali ties other than SjS such as prescription medication side effects. Theref ore, several objective tests have been used to confirm SjS in patients (Delaleu et al ., 2005). Although there are several te sts for specific aspects of SjS, there are no widely accepted criteria for SjS classification. However, most classification models for SjS include the follo wing manifestations: pr esence of ocular and oral sicca symptoms, measured decreased saliva ry and tear flow, lymphocytic infiltration of salivary glands, and presence of specific autoantibodies. The etiology of the disease remains to be elucidated. As a result, Sj S in patients often goes under diagnosed and untreated. SjgrenÂ’s Syndrome SjS is classified in two ways: primary a nd secondary SjS. The classifications are based upon clinical symptoms and the presen ce of other autoimmune disorders. SjSÂ’s association with other autoimm une disorders and presence of autoantibodies implies that there are both local and systemic aspects of the disease. Primary SjS exists when the

PAGE 11

2 lacrimal and salivary glands are affected and no other connective tissue autoimmune diseases are involved. Secondary SjS occurs when the typical SjS symptoms occur in association with other autoimm une disorders such as rheumato id arthritis, sclerodema, or systemic lupus erythmatosus. SjS displays a sexual dimorphism and ag e preference. Women are diagnosed with SjS more often than men at a ratio greater than 9:1. This is one of the highest female to male ratios when compared to other auto immune rheumatic diseases. Moreover, in women between the age of 40 and 60years old the disease is even more prevalent. Patients with SjS also have a high occurrence of lymphocytic malignancies especially in B-cell lymphocytes (Voulgarelis et al ., 2003). Although the etiology of SjS rema ins unknown, many proposals of SjS pathogenesis have been put fort h. One explanation is that the characteristic decrease of saliva and tear production in Sj S is the result of glandular destruction caused by an inflammatory autoimmune att ack on the acinar cells of la crimal and salivary glands. Disruption of acinar cell apoptotic pathways is thought to play a critic al role in the T-cell mediated glandular destruction. Histopa thological analyses confirms that the lymphocytes infiltrating the minor salivary gla nd of SjS patients consist mostly of CD4+ T-cells and less of B-cells. Macrophages and de ndritic cells have also been found in the salivary gland infiltrate s of SjS patients (Aziz et al ., 1997; Zeher et al ., 1991). A focal score of 1 or greater, where a focus is define d as a cluster of > 50 lymphocytes in a 4mm2 area, from lower lip biopsies are generally c onsidered to be abnormal. The infiltrates are positioned peri-ductal and peri-vascular to the lacrimal and salivary glands (Cha et al ., 2002). Acinar epithelial atrophy and fibrosis have also been observed in histological

PAGE 12

3 evaluations of salivary and lacr imal glands in SjS. It has been proposed that lymphocytic infiltrations cause SjS sympto ms via disruption of apoptotic pathways. However, there have been many discrepancies between gla ndular destruction and hyposalivation in SjS patients observed by researchers (Delaleu et al ., 2005). Histopathogical evidence from labial salivary gland biopsies of lymphocytic infiltration is one of the key components of SjS diagnosis. The manifestations of SjS extend far beyond the sicca symptoms of salivary and lacrimal glands. The extra glandular sympto ms can affect musculoskeletal, pulmonary, vascular, gastrointestinal, hepatobiliary, hematological, dermatological, renal, and nervous systems. This suggests that SjS pa thology has a systemic component affecting other tissues and organ systems. Indeed one of the hallmarks and diagnostic cr iteria of SjS is the presence of organ specific and non-organ specific autoantibodies in SjS patient Â’s sera. Antibodies against nuclear proteins such as SSA (Ro) and SS-B (La) have b een commonly found in the sera of SjS patientÂ’s. Patients with Sj S commonly show high amounts of IgG and hypergammaglobulemia in serum analysis. Intere stingly, rhemeutoid factors (RF) such as IgM-RF and IgA-RF are found in the sera and saliva of SjS patients (Atkinson et al ., 1989). In addition, antibodies reactive against th e type3 muscarinic acetylcholine receptor (M3R) have been detected in the sera of SjS patients. Althoug h the presence of autoantibodies is prevalent in SjS patients, th e role of autoantibodies in the pathogenesis of SjS remains obscure. Antibodies against M3R are thought to play a pivotal role in the pathogenesis of SjS by some researchers. Since secretion of water and electrolytes by acinar cells is

PAGE 13

4 directly induced by acetylcholine and substa nce P, disruption of acetylcholine ligation with M3R would seem to affect saliva production (Baum et al ., 1993). The importance of M3R in saliva production was confirmed via knockout mice experiments (Bymaster et al ., 2003). Moreover, the infusion of monoclona l anti-M3R antibody from NOD mice or SjS patientÂ’s sera into NOD and other mouse strains resulted in a loss of secretory function (Robinson et al., 1998). This suggest s that the sicca sympto ms associated with SjS could be the result of glandular dysfuncti on rather than acinar cell destruction. Thus, B-cell activation may be a cri tical point in SjS pathology i ndicating the importance of a humoral response in the pathology of SjS. Although the presence of lymphocytes and ac inar tissue destruction seem to point to glandular destruction as the main cause of the sicca symptoms in SjS, there are many documented discrepancies between lymphocy tic infiltration of salivary tissue and hyposalivation. Another contributing explanation could be that the si cca symptoms of SjS are caused by an autoantibody mediated gl andular dysfunction. Th e etiology of SjS remains to be elucidated in spite of the considerable data investigating the pathology of SjS. However, many hallmarks and characterist ics of SjS have been identified. Moreover, research has identified many biomarkers that are used to help dia gnose SjS in patients. Due to the fact that SjS symptoms occur in the late stages of the dis ease, the diagnosis of SjS in patients remains a major problem amongst clinicians and researchers. NOD Mouse Model In an effort to further investigate Sj S pathology, many animal models have been used in SjS research. Limitations of usi ng human tissue in SjS research range from ethical issues, environmental and dietar y variance amongst subjects, and genetic diversity. The non-obese diabetic (NOD) mous e strain displays many symptoms with

PAGE 14

5 considerable similarities to human SjS pathology. The NOD mouse develops type I diabetes via a unique H2g7 major histocompatability (MHC) haplotype (Leiter and Atkinson, 1998). The expression of Ag7 with no concomitant surface expression of an I-E molecule in the NOD mouse leads to a differenc e in binding affinity and affects antigen presentation. Also, the NOD mouse contains mu ltiple mutations resu lting in less potent IL-2, low expression of Fc receptor and high expression of prostaglandin synthase 2 in macrophages, lack of complement-d ependent lysis, and functional NK1+ T-cells. The NOD mouse displays different incidence of di abetes rates between male and female in different colonies indicating many interacti ons of endocrine factors, environmental factors, and multiple genes that may confer protection from or susceptibility to the disease. The NOD.B10.H2b is the congenic partner stra in to the NOD mouse whose MHC locus has been replaced with the non-d iabetogenic MHC locus of C57BL/10. The NOD.B10.H2b mouse displays an autoimmune exocrinopathy characterized by autoantibody generation, lymphocytic infiltra tion, and SjS-like sicca symptoms of the lacrimal and salivary glands. However, NOD.B10.H2b displays SjS autoimmune exocrinopathy without developi ng type I diabetes, making it an excellent model for the study of primary SjS in humans (Carnaud et al ., 1992; Robinson et al., 1998c). Additional studies on the C57BL/6 recombin ant strain, which contained the insulin dependent diabetes ( Idd ) susceptibility loci Idd5 derived from chromosome 1 of the NOD mouse, displayed biochemical changes of au toimmune exocrinopathy without the loss of secretory function (Brayer et al ., 2001). Thus, it appears that chromosome 1 of the NOD

PAGE 15

6 mouse controls the biochemical events and other loci critical to immune infiltration and loss of secretory function. Another SjS animal model is the NODscid mouse. The NODscid mouse was developed by breeding a homologous scid (severe combined immunodeficiency) locus into the NOD mouse background. NODscid mice are unable to produce B and T cells. As a result, the NOD-scid mouse has no functional immune system. The NODscid mouse has enabled researchers to explor e the non-autoimmune component of SjS pathology. Indeed, the NODscid mouse shows no lymphocytic infiltration of salivary tissue and no decrease in sa liva production. However, the NODscid mouse shows changes in protein composition and absent proteolytic activity in saliva when compared to the control mouse strain. This indicate s that alteration of the NOD salivary gland occurs independently of lymphocytic inf iltration. Moreover, saliv ary gland dysfunction may cause the immune response that l eads to glandular destruction (Robinson et al ., 1996). Temporal Changes in Salivary Composition of NOD Mouse Model Research on the NODscid mouse has led to the suggestion that SjS pathogenesis occurs in a two stages: an asymptoma tic stage and a symptomatic stage (Cha et al ., 2002). The asymptomatic stage precedes th e symptomatic stage and is hallmarked by several biochemical changes in the glandul ar function of the NOD mouse. One of the biochemical changes of the NOD mouse is the presence of new and or aberrantly altered proteins in the NOD saliva. Speci fically, the internal cleavage of parotid secretory protein (PSP) has been observed in th e saliva of the NOD mouse and is considered a biomarker for the disease (Robinson et al., 1998).

PAGE 16

7 PSP is a secretory glycoprotein found in abundance in the parotid gland of mouse and rat. It is also found in the lacrimal, subl ingual, and submandibular glands at different ages. Although the function of PSP is not know n, one study showed that it does bind to bacterial surfaces in a zinc-dependent ma nner suggesting that it may be capable of controlling bacteria l growth (Robinson et al., 1998). Another study identified human PSP as having anti-bacterial functi on similar to bactericidal/per meability-increasing protein (Geetha et al ., 2003). Researchers also noted an age-dependent variance of the composition of salivary proteins in the NOD -scid mouse. Observations of the saliva composition of the NODscid mouse showed that its salivary protein profile and flow were similar to the BALB/C and prediabetic mice (Robinson et al., 1998). However, the observati ons also showed that the saliva composition of the NODscid mouse changed as the mouse aged. Changes in the salivary composition of the NODscid mouse include: increase in amylase expression, increase in proteolytic activity, and aberrant expression of PSP. The temporal changes in the saliva prof ile of the NOD mouse were resolved by SDS-polyacrylamide gel electrophoresis. The ge l analysis showed a disappearance of a 32kD band and the appearance of a 20kD band in the saliva of NODscid mouse over 15 weeks of age. Coincidentally, a 27kD band appeared in the saliva of the NODscid over 15 weeks of age as well. When the 32kD, 27kD, and 20kD bands were N-terminally sequences they were found to be homologous to the published murine -PSP sequence with interesting distinctions. Th e 32kD and 20kD bands were shown to have N-terminal sequences that were identical to the secreted form of PSP th at is cleaved at the start sequence. The 27kD band was found to be iden tical to the PSP sequence that has been

PAGE 17

8 internally cleaved at a position 27 ami no acids downstream from the protein start sequence indicating the presence of an internally cleaved PSP isoform. The internal cleavage site was determined to be at a specific NL-NL amino acid sequence of PSP located 27 amino acids downstream from the prot ein start sequence. This specific NL-NL site appears to be an unusual cleavage site and does not serve as a substrate for any known proteases. Database searches of know n proteases have been unable to find any proteases that cleave PSP at an NL-NL s ite (Day, 2002). Thus, the cleavage of PSP appears to be caused be an unknown unique pr otease in the saliva of the NOD mouse. Western blot analysis using a polycl onal anti-murine PSP antibody confirmed the presence of aberrantly expressed PSP in mouse saliva of the NOD mouse. The western blots were used to compare saliva from 10-week-old NODscid 20-week-old NODscid 20-week-old BALB/c, and diabetic NOD mi ce. The saliva 32Kd and 20Kd band was found in both the10 week old NODscid and 20 week old BALB/c. On the other hand, the 20-week-old NODscid and the diabetic NOD mouse contained the 27Kd and 20Kd isoforms of PSP. Moreover, the presence of the internally cleaved 27 kD PSP isoform is indicative of PSP cleavage that is present in the saliva of older NODscid and NOD mice and absent in the younger 10 week old NODscid and BALB/c mice. PSP cleavage in the NOD mouse precedes the ap pearance of lymphocytic infiltrates of the salivary glands. Al so, PSP cleavage in the NODscid mouse occurs at the time when lymphocytic infiltrates would begin to appear in the salivary glands of the NOD mouse. This sequential synchronization of proteolytic activity and lymphocytic infiltration of the NOD mouse and NODscid suggest that the changes in the exocrine glands are independent of lymphocytic infi ltration and that the autoimmunity may be

PAGE 18

9 caused by changes in the morphology of the e xocrine glands. Thus PSP cleavage could be one of the hallmark biochemical changes of the asymptomatic phase of SjS in the NOD mouse model. PSP cleavage occurs in the NOD and many of its congenic strains, but not in the normal mice. Thus, the cleavage is presen t in the saliva of the diabetic NOD, NOD-scid and NOD.B10.H2b mice. However, the PSP cleavage does not occur in the saliva of the 10-week old NODscid 8-week old pre-diabetic NOD, C57BL/6 and BALB/c mice. Therefore, this PSP proteolytic activity co uld function as a biomarker for SjS. In 2002, researchers developed a High Performance Liquid Chromatography (HPLC) assay to detect the cleavage frag ments of a PSP-like peptide (Day, 2002). The PSP-like peptide employed by the HPLC assay is a 15 amino acid peptide that corresponds to amino acids 20 to 34 of the published sequence for mouse PSP. The PSPlike peptide also include s the unusual NL-NL cleavage site that serves as a substrate for the unknown PSP cleaving enzyme. In an effort to identify the protein responsible for PSP cleavage, purification methods were used to extract the PSP cleaving protease from NOD.B10.H2b saliva. The HPLC assay was used to detect the PSP cleaving proteases susceptibility to various sp ecific protease inhibitors. The protease inhibitor experi ments conducted on the NOD.B10.H2b saliva revealed some interesting findings and candidate protei ns that could be responsible for the PSP cleavage observed in the NOD.B10.H2b saliva. One interesting discovery from the experiments was total inhibition of the PSP cleavage by serine protease inhibitors. Conversely, the PSP cleavage was not affected by cysteine prot ease inhibitors and chelating agents designed to inhibit zinc-d ependent metalloproteas es as well as other

PAGE 19

10 proteases stabilized by calcium (Day, 2002). Th e protease inhibitor as says suggest that the unknown PSP cleaving protease in the NOD m ouse could be a serine protease. The partial purification and protein analysis of NOD.B10.H2b saliva also suggested candidate genes responsib le for PSP cleavage. The NOD.B10.H2b saliva was purified by Sephadex G-100 gel filtration column, which is used for batch separations of large peptides. The fractions that retained PSP cleaving activity were analyzed by SDSPAGE and commassie blue staining then thei r protein profiles were compared to normal control mice. The protein profiles of the NOD mouse showed a unique protein band that was not present in the normal control mouse. When this protein band was N-terminally sequenced the following amino acid se quence was revealed: ILGXFKXEKDSQPXQ. This amino acid sequence was then search ed against a database of known protein sequences and was shown to have str ong sequence homology to mouse glandular kallikrein 22 (mGK22) (Da y, 2002). Moreover, mGK22 is a known to belong to a serine proteinase gene family whose inhibitor was shown to prevent PSP peptide cleavage. Although the N-terminal fragment show ed near 100% homology with mGK22, it contained a single amino aci d discrepancy with the published mgk22 amino acid sequence. Specifically, mGK22 contains aspara gine residue at posi tion 34 whereas the Nterminal fragment sequenced from previous rese arch displayed an aspartic acid residue at that position. The research performed by Day on the NOD.B10.H2b saliva strongly supports the hypothesis that the PSP cleaving enzyme in the NOD saliva is mGK22 or a protein with high homology to mGK22. Glandular Kallikrein The mouse glandular kallikreins (GK) are a group of biologically active peptides that function as highly speci fic esterases and are encoded by closely linked genes located

PAGE 20

11 on chromosome 7. To date there are 28 GK genes in the mouse, 14 of which code for functional proteins. The kallikrein serine pr oteinase family was originally defined by their ability to release bioactive kinnin from high molecular mass precursors (Olsson and Lundwall, 2002). For example, a major GK found in kidney, pancreas, and salivary glands has been shown to cleave the precursor kininogen to release bradykinin which is a vasoactive peptide thought to pl ay an important role in re gulating blood flow (Schschter, 1980). GK’s are thought to be invo lved in a wide variety of peptide processing pathways and may represent potential regu latory steps in the conversio n of inactive precursors into biologically active peptides (Evans et al ., 1986). GKs, also known as tissue kallikreins, are a family of glycoproteins of varying molecular mass ranging from 25-40 kD. They are related to trypsin, chymotrypsin, and other serine proteases. The GKs all possess a histidine residue at amino acid position 41, an aspartate residue at amino acid position 96, and a serine re sidue at amino acid position 189. These three amino acids form what is known as the catalytic triad, a structure thought to be critical to the formation of the serine protease catalytic site (Young et al ., 1978). Moreover, mouse GKs -22, -9, and –13 cont ain the aspartate residue at amino acid position 183 that is thought to be required for cleavage at basic amino acids (Kreiger et al ., 1974). However, unlike trypsin, GK’s show a high degree of substrate specificity (Evans et al ., 1986). In fact in amino acid sequen ce comparisons, the GK’s display a high degree of homology with each other except in re gions that are thought to be important in determining substrate specificity. Thus the ac tions of GKs are highly specific suggesting that their role (if any) in bioprocessing is exclusive to certain pathways. The high degree of specificity and large multigene family th at encodes GK’s supports the hypothesis that

PAGE 21

12 GK’s have an integral role in the processi ng of a wide variety of hormone and growth factor precursors (Mason et al ., 1983). mGK22 is a 29 kD protein that is expresse d in the salivary glands of mice in a prepro zymogen form. mGK22 is activated by cl eavage of the zymogen peptide at an arginine residue located 24 amino acids downs tream from the peptide start site. Although mGK22 has characteristics similar to tryps in, it lacks the trypsin calcium-binding loop and fails to form trypsin’s six disulfide bridge s. In contrast, mGK22 has the characteristic kallikrein loop beginning at amino acid pos ition 77 and forms five disulfide bridges (Blaber et al ., 1987). mGK22, also known as epiderma l growth factor binding protein (EGF-BP) type A, is one three GKs known to bind and cleave the mouse 9 kD epidermal growth factor (EGF) precursor at the car boxy terminus to produce the mature growth factor. Interestingly, EGF is a major protein produced by the salivary glands and secreted in saliva. Other kallikrein EGF-BPs are EG F-BP type B and type C coded by mGK13 and mGK9 genes respectively. Research has s hown that mGK13 and mGK26 have a 99% homology leading researchers to conclude that they represen t allelic variations of the same gene (Olsson and Lundwall, 2002). A lthough mGKs –9, -13, and -22 all bind and process pre-pro EGF, there are no identical re sidues between them other than any regions conserved between the majority of other kall ikreins. Thus, there seems to be no obvious critical residues between the three EGF-BP s that would confirm EGF binding ability. Also, mGK22 and other EGF-BPs may play a crucial role in the regulation of mature growth factors (Blaber et al ., 1987). The importance of GKs in the progression of SjS remains to be shown in spite of the existence of intriguing data that s hows GK’s involvement with inflammation and

PAGE 22

13 immune responses. Interestingly, research ers have found mGK13, also known as EGFBP type B, autoantibodies in the sera of another SjS mouse model (IQI/Jic mice). Moreover, mGK13 was shown to cause a prolif erative response of splenic T-cells, in vitro (Takada et al ., 2004). This data supports the hypot hesis that mGK13 may act as an auto-antigen that increases the response of T-cells to organs th at commonly express mGK13. This hypothesis, if true, may be furthe r strengthened by the fact that mGKs –22, -9, -13 are expressed exclusivel y in the salivary glands, the target of the immune response associated with SjS (Drinkwater et al ., 1987). Thus GKs may play an important role in the etiology of SjS in two ways: via lym phoproliferative activity and via autoantibody generation. Previous data has suggested that mG K22 may be the enzyme required for PSP cleavage in the saliva of the NOD mous e model (Day, 2002). This hypothesis is supported by the fact that GKs are known to have proteolytic eff ects on various protein precursors and may be potential regulators of peptide activation in various specific pathways. However, it must first be shown that mGK22 is capable of cleaving PSP in isolation. Therefore in an attempt to further solidify mGK22’s candidacy as the enzyme responsible for PSP cleavage in the NOD mouse, mGK22 must be isolated and assessed for PSP cleaving activity. Since one of the mo st reliable and accurate methods of PSP cleavage assessment is via the aforementione d HPLC PSP assay, this method facilitates detection of PSP proteolysis. The focus of the current study is to de termine if mGK22 is expressed in the submandibular glands and if this GK is cap able of cleaving PSP. At the same time, a NOD.B10.H2b submandibular tissue cDNA library will be screened by hybridization with

PAGE 23

14 oligonucleotide probes complementary to the Nterminus of mGK22 to determine if other candidate genes for related proteins may be identified. In order to assess mGK22 PSP cleaving activity and find other candidate s genes responsible for cleaving PSP the following 4 goals were established: Construct a cDNA library of NOD.B10.H2b submandibular tissue and clone mGK22 into a suitable vector for expression. Probe the cDNA library of NOD.B10.H2b submandibular tissue with primers complementary to the N-terminus of mGK22 for possible candidate genes and mutants capable of cleaving PSP.Express mGK22 from NOD.B10.H2b submandibular tissue cDNA library. Detect PSP cleavage activity of e xpressed genes by HPLC-PSP assay.

PAGE 24

15 CHAPTER 2 MATERIALS AND METHODS Mouse Models The animal mouse model used in this research was the NOD.B10.H2b mouse strain. All of the mice used in this research were approximately 14 weeks of age and purchased from the University of Florida Departme nt of Pathology Mouse Colony. The mice were held under SPF conditions, provided food and water, and maintained on a 12-hour darklight cycle until euthanized. Studies were carried out under IACUC-approved protocol CO17. cDNA Library Construction Total RNA was isolated and purified from the homogenized submandibular glands of 14-week old NOD.B10.H2b mice using the RNAeasy RNA extraction kit (Qiagen). RNA purity and concentration were confir med by UV absorbance at 260/280 nm and formaldehyde gel electrophoresis. The c DNA first strand synthesis via reverse transcription polymerase chain reaction (RTPCR) was performed on the purified total RNA from NOD.B10.H2b using the SMART (Switching Me chanism At 5' end of RNA transcript) cDNA Library Construction Kit (Clo ntech), as outlined in the manufacturerÂ’s protocol. The SMART protocols (Clontech) employ a 5' SMART IV oligonucleotide and a 3' CDS III primer (modified oligo (dT) prime r) in order to preferen tially enrich for fulllength cDNAs during first stra nd synthesis and subsequent PCR amplification. During first strand synthesis, the modified oligo (dT) is used to prime the initial reaction while reverse transcriptaseÂ’s terminal transferase activity adds extra nucleotides, primarily

PAGE 25

16 deoxycytodine, to the 3' end of the cDNA. Th e SMART IV oligo has an oligo (G) stretch which base pairs with the deoxycytodine st retch of the cDNA creating an extended template. The result is a full-length cDNA that contains the complete 5' end of the mRNA and a complementary SMART IV oligo sequence that can serve as a universal priming site for the subsequent LD-PCR amplifica tion. The first strand synthesis cDNA was amplified using the LD-PCR (Long Distan ce PCR) method according to the SMART cDNA Library Construction Kit manual (Clo ntech) and sub-cloned into pDNR-LIB plasmid (Clontech). The cDNA-containing p DNR-LIB plasmid was used to transform 25 L of “Electromax” Top10 electro-compet ent cells (Invitrogen) by electroporation according to the manufacturer’s instructions. The transformed cells were then added to 970 L of LB broth and allowed to incubate for 1 hour at 37C with shaking (225 rpm) to create the original, unamplified cDNA librar y. The titer of the or iginal, unamplified cDNA library was calculated to be 6.6 x 105 colony forming units per milliliter (CFU/mL). The transformed bacterial cells we re then plated at a concentration of 100 colonies per plate on 250 mm agar plates containing 1.5% agarose LB-media with chloramphenicol (30 g/mL). The plates were then allowed to incubate at 37C overnight. cDNA Library Screening for mGK22 The transformed bacterial colonies were transferred to a positively charged nitrocellulose membrane and subjected to DNA hybridization analysis The colonies were hybridized to 3 tail labeled digoxigenin-11-dU TP/dATP DNA (dig-labeled) probes whose sequences were identical to the N-terminus of mGK22 (5 ATACTTGGAGGATTTAAATGTGAG AAGAATTCCCAACCCTGG-3 corresponding to nucleotides 73-114). The hybridizations we re performed according to the Genius

PAGE 26

17 System Users Guide version 3.0 (Roche). Repl ica-plated nitrocellulose membrane colony lifts containing cDNA, were generated by placi ng the membranes on top of a cold agar plates for 1 minute. After the bacterial colo nies were transferred to a positively charged nitrocellulose membranes, the membranes were placed in series of alkaline washes (0.5N NaOH, 1.5M NaCl and 0.5N NaOH, 1.5M NaCl 0.1% SDS) to lyse the cells and a neutralization wash (1.0M Tris-HCl, 1.5M NaCl, pH 7.5). Next, the membranes were baked at a 120C to fix the colony DNA to the membrane. The nitrocellulose membranes, with the fixed DNA, were pre-hybridized at 65C overnight in hybridization buffer (5X SSC, 1.0% blocking reagent, 0.1%N-lauroylsar cosine, 0.02% SDS). After washing in 2X SSC, the membranes were incubated at 68C wi th the dig-labeled oligonucleotide probes in hybridization solution (2.0 pmol/mL) fo r 1 hour. After hybridiz ation, the membranes were washed in washing buffer (0.1M ma leic acid, 0.15M NaCl, 0.3% Tween 20) to remove any non-hybridized probes. Next, the membranes were incubated in blocking buffer (0.1M maleic acid, 0.15M NaCl, 1% bl ocking reagent) for 30 minutes and then placed in blocking solution that contai ned anti-digoxigenin (150 mU/mL) antibody conjugated to alkaline phosphatase for 1 hour. To remove any unbound antibody the membranes were washed twice in washing buffer. Detection of the bound antibody was accomplished by colorimetric development of alkaline phosphatase. Colorimetric development was performed on the membranes by adding 10 mL color substrate solution, which contained nitroblue tetrazolium salt (0.3375 mg/mL) and 5bromo-4-chloro-3indolyl phosphate (0.175 mg/mL) in 10mL of detection buffer (100mM Tris-HCl, 1.5M NaCl, 50mM MgCl2, pH 9.5). Colorimetric development was allowed to continue

PAGE 27

18 overnight (approximately 12 hours). To preven t over-development, the membranes were washed twice, in H2O, and air-dried for storage. Analysis of Plasmid DNA by Gel El ectrophoresis and Enzyme Digestion The plasmids of the transformed bacteria l cells from the cDNA library and mGK22 expression experiments were analyzed by ge l electrophoresis and restriction enzyme digestion. Colonies from transformed cells we re plated on antibiotic selective media. Next, they were used to i noculate 50ml cultures of LB-m edia with the appropriate selective antibiotic and allowed to incubate at 37C overnight. The plasmids from the transformed cell cultures were isolated and purified using a Maxi-Prep plasmid purification kit (Qiagen). The protocol was performed as outlined in the Maxi-Prep plasmid purification kit manual (Qiagen). The purified plasmids were analyzed for UV absorbance at 260/280 nm using a spec 300 (B io-Rad) to determine DNA concentration and purity. The plasmids were also load ed onto a 1% agarose gel containing 4.0% ethidium bromide and electrophoresed for 2 hours at 50V. The plasmids were then digested with restriction enzymes Xba I and Hind III (Promega), separately, to analyze the inserts they may contain. Approximately 10.0 g of purified plasmid DNA was combined with 10 units of Xba I (approximately 2.0 l) and incubated at 25C for 1 hour. Separately, another sample of 10.0 g of purified plasmid DNA was combined with 10.0 units of Hind III (approximately 2.0 l) and incubated at 37C for 1 hour. Both digestions were prepared in 50 l reactions with enzyme buffer (25.0 Tris acetate, 0.1 Potassium acetate, 10.0 mM Magnesium acetate, 1.0 D TT, pH 7.80, 0.1 mg/ml Acetylated BSA). After digestion incubation, both samples we re incubated to 65C for 15 minutes to deactivate the restriction enzy mes. The samples were analyzed by gel electrophoresis for

PAGE 28

19 their respective DNA fragment profiles. Finally, the identity and positio n of the inserts of the plasmids were confirmed by DNA sequencing. Expression of Recombinant mGK22 The cDNA from the first strand synthesi s was subjected to polymerase chain reaction (PCR) amplification using Proofst art high-fidelity DNA polymerase (Qiagen) and mGK22 specific primers (5 -CACCGCACCTCCTGTCCAGTCTCGAATAC-3 corresponding to nucle otides 52-76 and 5 TCAGGGGTTTTTGGCCA TAGTGTCTTTT-3 complementary to nucleotides 753780). To facilitate directional cloning, the forward primer containe d 4 added nucleotides (CACC) as specified in the pET200 D-topo expression vector manual (Invitrogen). The PCR conditions were 94C for 3 min followed by 30 cycles of 94C for 45 sec, 65C for 45 sec, 72C for 2 min, followed by a final ex tension at 72C for 7 min. PCR products of approximately 750 nucleotide base pairs (bp) which corresponds to the size of mGK22, were separated from other PCR fragment s by gel electrophoresis. The purified cDNA, which contained the complete region of th e mGK22 gene and stop codon, was sub-cloned into pET200 D-topo expression vector (Invitr ogen). The orientati on and sequence of the cDNA in the pET plasmid were confirmed by DNA sequencing. The ligated pET200 Dtopo vector was used to transform Escherichia coli strain BL21 (DE3) cells (Invitrogen) for expression. The transformed cells were incubated at 37C in 100 mL of LB media containing 50 g/mL kanamycin until they reached an optical density of A600 The cells were harvested by centrifugation at 5000g for 10 min. The cell pellets were resuspended in 1 mL of lysis buffer (50 mM NaH2PO4, 300 mM NaCl, pH 8.0). The cells were treated to freeze-thawing cycles and sonication (6 cycles of 10second bursts of sonication followed by 10 seconds cooling) to lyse the ce lls. To prevent overheating of the samples

PAGE 29

20 and protect the expressed protein, the sonica tion of the cells was performed on ice. To separate insoluble cellular debris from th e expressed recombinant mGK22 fusion protein, the crude cell lysates were centrifuged at 10,000g for 30 min. The presence of the fusion protein, which contains mGK22 residues and extra ami no acids originating from the pET200 D-topo plasmid sequence at the N-te rminus of mGK22, was detected in the lysates by western blot analysis with a 6X anti-histidine an tibody. The crude cell lysates were analyzed for enzymatic ac tivity using the HPLC-PSP assay. Protein Analysis SDS-PAGE Crude cell lysates were analyzed by s odium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) to separate thei r constituent proteins by molecular weight. The crude cell lysates were combined with 2X laemmelli buffer (BioRad) in equal volumes (15 l) and incubated at 65 for 15 minutes. The entire sample, 15 l of crude cell lysates in 15.0 l of laemmelli buffer, was loaded onto a 15.0% Tris-HCl polyacrylamide gel (Bio-Rad). The gels were electrophoresed for 1 hour at 100V in Tris Glycine-SDS Buffer (Bio-Rad).After electrophoresis, th e gels were washed twice in H2O in preparation for the next analysis. Commassie Blue Staining Visualization of the constituent proteins of the crude cell lysates was accomplished by commassie blue staining. The polyacrylamide gels from the SDS-PAGE of the crude cell lysates were placed in commassie blue staining solution (0.125% commassie brilliant blue R 250, 50.0% methanol, 10.0% acetic acid) a nd allowed to stain for 30 minutes. The polyacrylamide gels were then placed in de stain solution (50.0% methanol, 10.0% acetic acid) for 15 minutes. Afterward, the destain solution was discarded and replaced by fresh

PAGE 30

21 destain solution. The polyacrylamide gels were allowed to destain overnight. The polyacrylamide gels were then washed twice in H2O and photographed. Membrane Blotting After separation of proteins, the proteins were transferred to a nitrocellulose membrane using a Semi-Dry Transfer Ce ll (Bio-Rad). The proteins from the polyacrylamide gel were transferred to th e membrane at 20V for 30 minutes with membrane transfer buffer (80% Tris Glycin e-SDS buffer and 20% methanol). Following the transfer, the membranes were washed twice in H2O and used for western blot analysis with 6X anti-histidine antibody to detect the expressed protein. Western Blots The nitrocellulose membranes that contai ned the transferred proteins from the crude cell lysates were subjected to western blot analysis to detect the presence of a 6X histidine protein. The western blot analysis was performed as outlined in the Western Breeze western blot kit (Invitrogen) manual with the following specifications. All incubations were performed at room temp erature. After the protein transfer, the membranes were washed twice with H2O and incubated in bloc king buffer (1.0% bovine serum albumin, 10mM Tris-HCl, 150mM NaCl, 0.05% Tween 20, pH 8.0) for 1 hour. The membranes were washed 3 times for 5 minutes in antibody wash (10mM Tris-HCl, 150mM NaCl, 0.05% Tween 20, pH 8.0). The memb ranes were then incubated with 6X anti-histidine primary antibody in blocking bu ffer at a concentration of 10 pmol/ml. Afterward, 3 washes of 5 minutes each in antibody wash were performed to remove unbound antibodies. Then the membranes were in cubated with anti-rabbit IgG secondary antibody conjugated with alkaline phosphata se (Sigma) at a dilution of 1:7500 in blocking buffer The membranes were washed three times for 5 minutes in antibody wash

PAGE 31

22 to remove unbound antibodies. After pre washin g in water twice, the membranes were color developed by incubating the membrane s in color substrate solution, which contained nitroblue tetrazolium salt (0.33 mg /mL) substrate and 5-bromo-4-chloro-3indolyl phosphate (0.165 mg/mL) substrat e in 10mL of alkaline phosphatase buffer (100mM Tris-HCl, 100mM NaCl, 5.0mM MgCl2, pH 9.5). To prevent over-development, after 4 hours the membranes were washed twice for 5 minutes in H2O. Membranes were then air-dried for storage. HPLC-PSP Assay PSP Peptide Synthesis The PSP peptide employed by the HPLC-PSP assay was a custom designed peptide synthesized by the University of Florid a ICBR Protein Chemistry Core Facility, Gainesville. The PSP peptide was synthesized on an Applied Biosystems Peptide Synthesizer model 432A using solid phase FMOC chemistry. The PSP peptide was designed to mimic PSP, since it contains the amino acid residues of the PSP cleavage site it can act as a substrate for the unknown PSP cleaving protease. The PSP peptide amino acid sequence was N -EAVPQNLNLDVELLQ-C The PSP peptide amino acid sequence was identical to the amino acids 20 through 34 of the entire published Mus musculus PSP sequence (figure 2-1). Also, the PSP peptid e contains the PSP pr otease specific NL-NL cleavage site corresponding to the 26th and 27th amino acid positions of the Mus musculus PSP sequence. The PSP peptide molecular weig ht and purity were confirmed by HPLC (figure 2-2) and mass spectro scopy (figure 2-3). The Univers ity of Florida ICBR Protein Chemistry Core Facility performed both an alyses on the PSP peptid e. For the purpose of PSP cleavage detection by HPLC, 25 mg of th e PSP peptide was dissolved in 10 mL of PSP peptide buffer (10 mM Tris-HCl, pH 8.02). The concentration of PSP peptide in the

PAGE 32

23 PSP peptide solution, used as the enzyme substrate in the HPLC-PSP assay, was 2.5 mg/ml. Detection of PSP Peptide Cleavage by HPLC To detect PSP cleavage activity in crude cel l lysates, the PSP peptide was used as a substrate for the unknown PSP cleaving enzyme. 40 l of crude cell lysates from the mGK22 expression experiments was added to 40 l of the PSP peptide solution and incubated at 42 C overnight, to ensure complete cleavage. As positive and negative control samples for PSP cleavage activity, saliva from NOD.B10.H2b and BALB/C mouse strains were analyzed for PSP cleavag e activity by HPLC. The saliva control samples (NOD.B10.H2b and BALB/C) were separately in cubated with PSP peptide. In each sample, 40 l of the PSP peptide solution and 10 l of saliva were combined along with 30 l of PSP peptide buffer and incubated at 42C overnight. All samples, both crude cell lysates and control saliva samples, were filtered through a 0.45 m Regenerated Cellulose Micro-Spin filter tube (Alltech) by centrifugation at 5000g for 5 minutes prior to HPLC analyses. The HPLC-PSP assay used to asse ss the PSP cleavage activity functions by detecting the cleavage fragments of the PSP peptide. The HPLC-PSP assay uses a reverse-phase 5 m 300 Jupiter C18 column (phenomenex) to separate peptides and peptide fragments from each other and othe r compounds based on their binding affinities to the column. The HPLC-PSP assay uses a line ar gradient elution of two buffers: buffer A (0.1 % Trifluroacetic acid in Acetonitrile) and buffer B (0.1% Trifluroacetic acid in HPLC grade H2O). The HPLC-PSP assay method used in this research specified for a linear gradient eluti on of 10% buffer A and 90% of buffer B to 90% buffer A and 10% of buffer B at a rate of 1.0 ml/min over a durat ion of 20 minutes for peptide separation. The

PAGE 33

24 peptides were detected by an AD20 wavele ngth detector (Dionex) which distinguishes peptides by their respective ultraviolet (UV) absorbance at 214nm as they are eluted from the column. The elutions were profiled by documentation on a graph of retention time versus UV absorbance. Also, a linear re gression of PSP peptide peak areas was constructed, enabling relative amounts of PSP peptide in samples to be determined. Figure2-1.Amino acid sequence of Mus musculus parotid secretory protein. The complete PSP is shown intact with the 20 amino acid leader sequence. The rectangular box outlines the synthetic PSP peptide used in the HPLC-PSP assay. The arrow labels the specific PSP protease NL-NL site.

PAGE 34

25 CHAPTER 3 RESULTS cDNA Library Screening for mGK22 A NOD.B10.H2b submandibular gland tissue cDNA library was screened with an (42-mer) oligonucleotide probe id entical to the N-terminal sequence of the mGK22 gene. The oligonucleotide probe sequence was also id entical to the back-t ranslated sequence of the previously reported unique protei n present in the PSP cleaving NOD.B10.H2b saliva (Day, 2002). The screening of 65 plates of th e cDNA library at 100 colonies per plate, approximately 6500 clones, revealed 35 positive colonies. Since the initial colony titer was 6.6 X 105 CFU/mL, approximately 1% of the to tal library was screened. Thus, these 35 colonies probably represent abundant mR NA species. An example of a screened nitrocellulose membrane is shown in (Figure 3-1). The cloning region of each plasmid these pos itive colonies cont ained was directly sequenced to identify the cDNA inserts they contained. The sequencing revealed interesting gene identities of the positive clones including the kallikrein genes mGK9, mGK26, and mGK6. The alignmen ts and comparisons of th ese kallikrein genes are shown in figures 3-2 thru 3-8. Other genes th at were identified in the screening were: salivary protein1, salivary protein2, cytochro me P450, and PRL-inducible protein. An interesting result of the cDNA lib rary screening was that mGK6 was found to be the most frequently expressed in the cDNA libra ry. On the other hand, mGK22 was never identified in the cDNA library via scr eening with an mGK22 homologous probe.

PAGE 35

26 However, mGK22 was identified and amplif ied from the cDNA library using mGK22 specific primers (Figure 3-9). The products from two separate PCR reac tions, using mGK22 specific primers and the cDNA library as a template, were seque nced. Each PCR reaction, when sequenced, showed that each reaction produced a differe nt form of the of the mGK22 gene. The sequencing revealed that one PCR reaction produced a transcript identical to the published mGK22 sequence while the other PCR reaction produced a transcript with a single base pair deviation from the publis hed mGK22 sequence. When translated, the single base pair deviation resulted in an aspartic acid for an asparagine amino acid substitution at position 34 in mGK22. This substitution was identical to the substitution found in the N-terminal sequencing of unique protein bands from purified fractions with PSP cleaving activity from previous research (Day, 2002). Thus, the PCR amplifications detected the presence of two different fo rms of the mGK22 gene present in the cDNA library of NOD.B10.H2b submandibular gland tissue. Moreover, the different forms of mGK22 were produced by PCR in isolation and not as a mixture of both forms. In addition to the cloned mGK ge nes showing 100.0% homology to the mGK published sequences, the screening revealed several genes in the cDNA library that contained minor base pair deviations from the published sequence. These point mutations, if translated, coul d result in amino-acid substitu tions that deviate from the published amino acid sequences of the mGK proteins. For example, one positive colony whose sequence showed homology with mGK26 displayed a single nucleotide discrepancy resulting in a lysine to proline amino acid substitution at the 112th aminoacid position (Figures 3-6 and 3-7). This nucleotide discrepancy was identified in

PAGE 36

27 colonies obtained from two se parate screenings (Table 1) Also, a positive colony with sequence homology to mGK6 that displayed another single nucleotide discrepancy resulting in a proline to lysine amino acid substitution at the 123rd amino-acid position was identified in colonies obtained from three separate screenings. Thus, the cDNA library screening of approximately 6500 cl ones did not identify any specific mGK22 colonies, but did show the presence of se veral other mGKs and mGK variants with significant frequency. Analysis of cDNA Cloned mGK22 and pET200 D-topo Constructs In order to express mGK22 from the NOD mouse, expression vectors for mGK22 were constructed. The expression of mGK 22 was performed using the protocol as outlined by Matsui et al (2000). Using NOD.B10.H2b submandibular tissue cDNA as a template, PCR amplification, with mGK22 sp ecific primers produced a fragment of approximately 750 bp (Figure 3-9 lanes 2 and 3) A fragment of 750 bp is consistent with the published size of mGK22. Moreover, th e predicted size of the PCR fragment generated from the mGK22 specific primer s was 704 bp. Indeed the PCR fragment generated from mGK22 specific primers and mouse submandibular tissue cDNA displayed a slight shift below the 750 bp mol ecular weight marker (Figure 3-9 lane 1). Thus the mass of the nucleotide sequence generated from the PCR cloning of mGK22 was consistent with that of mGK22. The fragment generated from the PCR r eactions was ligated into pET200 D-topo vector for expression. The resulting plasmid-v ector construct was designed to contain the entire coding region for mGK22 as well as a 6X histidine leader sequence. The plasmidvector construct was used to transform E. coli (BL21).

PAGE 37

28 The transformed E. coli (BL21) plasmids were purified and analyzed by gel electrophoresis to determine thei r size (Figure 3-10). If mGK2 2 has correctly ligated into the plasmid, the resulting plasmidÂ’s approxima te molecular weight is expected to be 6.5 Kbp. Lane 4 of figure 3-10 cont ains a plasmid with an appr oximate weight of 6.5 Kbp. Moreover, the negative control in lane 5 of figure 3-10 (an empty pET200 D-topo plasmid with a molecular weight of 5.6 Kbp) displayed its plasmid at a lower position than the plasmid in lane 4, thus indicating that the plasmid in lane 5 is of a lower molecular weight than the plasmid in la ne 4. The negative control in lane 5 was determined to have an approximate mol ecular weight of 5.6Kbps. Therefore, the molecular weight of the plasmid in lane 4 of figure3-10 is consistent with expected molecular weight of the m GK22/pET200 D-topo construct. The plasmids were analyzed by digesti on with the restricti on enzyme Hind III to yield predicted size fragments. Analysis of the nucleotide sequence of the vector construct showed that mGK22 contained a Hind III restriction enzyme cleaving site. Also, the pET200 D-topo plasmid, supplied by Invitrogen, has a Hind III restriction enzyme cleaving site. Therefore, if the m GK22 cDNA insert ligated into pET200 D-topo vector in the correct orient ation, it should generate DNA fr agments of approximately 803 bp and 5646 bp when digested with Hind III. On the other hand, the PCR positive control (lac Z) also contains a Hind III restriction enzyme cleaving site. However, the Hind III cleaving site in the PCR positive control (lac Z) is at a different position than the Hind III cleaving site in mGK 22 causing it to produce a slightly la rger fragment when digested with Hind III. The positive control lacZ /pET200 D-topo construct should generate DNA fragments of approximately 845 bp and 5646 bp, slightly larger than mGK22-pET200 D-

PAGE 38

29 topo construct. Figure 3-11 displays the gel electrophoresis of the vector constructs digested with Hind III. Examination of figur e 3-11 shows that a small DNA fragment in lane 4 (lacZ/pET200 D-topo positive control) was slightly larger than the small DNA fragment in lane 5 (mGK22). Moreover, th e large DNA fragments of the 2 separate digests were relatively the same size, as expected. These DNA fragment profiles were consistent with the predicted DNA frag ment profiles of mGK22/pET200 D-topo construct. The actual DNA sequence of the mGK22/pET200 D-topo construct was confirmed by DNA sequencing. The translati on of the DNA sequence showed that the insert that was contained in the mGK22/p ET200 D-topo construct is the complete promGK22 gene transcript with extra sequences of amino acids on the N-terminus that were donated by the pET200 D-topo plasmid (Figure 3-8). Thus, the restriction enzyme digestion analysis and DNA sequencing of the mGK22/pET200 D-topo confirmed that the construct contained the PCR-cloned mGK22 gene, intact, in frame for expression, and in the correct orientation. Analysis of mGK22 Expression The protein profiles of E. coli (BL21) transformed with mGK22/pET200 D-topo plasmid were examined by commassie blue staining (Figure 3-12) and western blot analysis (Figures 3-13 and 3-14) with an an ti-6X histidine antibody. The protein profiles of E. coli (BL21) showed the appearance of a protein band with a 6X histidine tag following induction with IPTG. The recomb inant protein from mGK22/pET200 D-topo should have an approximate molecular wei ght of 27.7 Kbp. The commassie blue staining and western blot analysis both showed the appearance of protein at an approximate molecular weight of 27 Kbp strongl y supporting the conclusion that the E. coli (BL21) are expressing the recombinant protei n from mGK22/pET200 D-topo plasmid.

PAGE 39

30 Commassie Blue Staining As stated above, the pr otein profiles of the E. coli (BL21) obtained from the commassie blue staining show the appearance of a protein band with the approximate weight of 27 kD. In figure 3-12, lanes 2 and 3 the appearance of a band just above the 25 kD molecular weight marker indicate s the expression of the recombinant mGK22/pET200 D-topo protein. In lanes 1 a nd 11, the molecular weight markers that correspond to an approximate molecular weight of 35 kD and 25 kD are the 6th and 7th band down from the top, respectively. Moreover, the 27 kD protein band in lanes 4, 6, 8, and 10 of figure 3-12 are absent in lanes 2, 3, 5, 7, and 9 of the same figure. Lane 2 of figure 3-12 contains the crude lysate of the E. coli (BL21) prior to induction with IPTG (time point zero). Lanes 3, 5, 7, a nd 9 contain crude lysate of the E. coli (BL21) that was not induced with IPTG. Thus, the commassie blue staining shows the presence of an inducible protein in the crude lysate of E. coli (BL21) cells transformed with mGK22/pET200 D-topo plasmid. Western Blot Analysis To detect the expression of the recombinant mGK22 protein in the E. coli (BL21) transformed with mGK22/pET200 D-topo pl asmid, a western blot analysis was performed on the crude lysates of the transformed E. coli (BL21). A western blot can be used to detect the presence of a 6X histid ine tag via the employment of an anti-6X histidine antibody. The expression vector m GK22/pET200 D-topo has been designed to incorporate a 6X histidine tag on the N-terminus of mGK22. Thus the presence of the 6X histidine tag, at the approxi mate molecular weight of 27.7 kD, would further confirm the expression of recombinant mGK22 in the transformed E. coli (BL21).

PAGE 40

31 In figure 3-13, lanes 3 and 4 show the pres ence of a 6X histidine tag at a position just above the parallel of the 25 kD molecular weight marker (Bio-Rad) located in lane 1. The position of the 6X histidine protein bands in lanes 3 and 4 of figure 3-7 corresponds to an approximate molecular weight of 28 kD The molecular weight of the 6X histidinetagged protein in the crude lysates is consis tent with the producti on of the recombinant mGK22 protein from the m GK22/pET200 D-topo plasmid. Figure 3-14 displays the wester n blot analysis of the crude lysates respective pellet and supernatant following centrifugation. La nes 3 and 5, which contain the pellet component of the crude lysate at one and tw o hour time points, show the presence of a 6X histidine protein band with an approxima te molecular weight of 28 kD. In contrast, there are no protein bands in lanes 4 and 6 which contain th e supernatant component of the crude lysate at one and two hour time point s. The 6X histidine protein present in the crude lysates appears to be in the in soluble, pellet of the crude lysates. The western blots also revealed an increas e in the concentration of a 6X histidine tagged protein associated with an increase in incubation time of the crude lysates of the transformed E. coli (BL21). The intensity of the 6X histidine protein bands increases from lane 3 to lane 4, which represents the time points 0.5 hours and 1.0 hour respectively. The time dependent increase of the protein in the crude lysates implies that it is an inducible protein whos e concentration increases with time. Thus, there appears to be a 6X histidine-tagged inducible protein pres ent in the crud e lysates of the transformed E. coli (BL21). The western blot analysis pe rformed on the crude lysates provides strong evidence of the presence of a 6X histidine-tagged inducible recombinant protein in the transformed E. coli (BL21).

PAGE 41

32 HPLC-PSP Analysis to Determine PSP Pr oteolytic Activity in Crude Lysates To assess the crude lysates of the transformed E. coli (BL21) for PSP proteolytic activity, they were subjected to analysis by the HPLC-PSP peptide assay (Day, 2003). The HPLC-PSP peptide assay uses a synthetic pe ptide as a substrate to detect the activity of a protease that cleaves PSP at a speci fic amino acid sequence. The 15 amino acid synthetic peptide that serves as the substrate in the assay is identical to the sequence of PSP from amino acids 20 to 34. The syntheti c peptideÂ’s sequence includes the leucineasparagine (NL-NL) cleavage site located at amino acid positions 26 and 27 in PSP. The peptide profiles can reveal PSP cleaving activity by detecting the presence of a single peak, representing the intact PSP-like pe ptide synthetic, or split peaks, representing the fragments of cleaved PSP-like synthetic pe ptide. The intact peptide and its cleavage fragments have their respective reten tion times for furthe r identification. To determine the retention time and peak area of the uncleaved, intact PSP peptide; it was subjected to the HPLC-PSP peptide a ssay. Exactly 40 L of the PSP peptide solution was added to 40L of lysis buffer a nd incubated at 42 C overnight. The sample was then filtered and analyzed by the HPLC -PSP peptide assay. As shown in figure 3-15, this control sample displayed a single p eak with a retention time of 13.46 min. The retention time of this peptide was used to compare the retention times of the uncleaved, intact PSP peptide in all future assays. Th e peptide-alone control sample enabled the relative peak area for a sample of PSP peptide, at a concentration of 1.25 mg/ml, to be determined. The PSP peptide-alone control samp le was assayed to de tect the effects of the components of the buffers used in the pr eparation of the crude lysates on the retention time and peak area of the PSP peptid e during the HPLC-PSP peptide assay.

PAGE 42

33 In an effort to determine the retention time and peak profiles of an enzymatically cleaved PSP peptide, the saliva of the NOD.B10.H2b mouse was analyzed by the HPLCPSP peptide assay. This sample, which includes NOD.B10.H2b saliva, has been previously shown to cleave PSP peptide and serves as a positive control for the HPLCPSP peptide assay. Figure 3-16 displays the HP LC chromatogram of sample containing 5 L of NOD.B10.H2b saliva combined with 40 L of PSP peptide solution and 35 L of lysis buffer. The sample was incubated at 42 C overnight, filtered, and then subjected to the HPLC-PSP peptide assay. The two peaks at 9.16 minutes and 12.21 minutes represent PSP peptide fragments 1 and 2, respectively. This positive control was used to show that the enzymatic activity of the unknown PSP cleaving enzyme, present in the NOD.B10.H2b saliva, is not affected by the compone nts of the lysis buffer used in the crude lysate preparations. To assess for PSP cleavage activity in E. coli (BL21) transformed with mGK22/pET200 D-topo plasmid, the whole cell ly sate and the cleared supernatant of the crude lysate were subjected to the HPLCPSP peptide assay. Briefly, 40 L of the crude whole cell lysate and 40 L of the clear ed supernatant from the mGK22 expression experiments was added, separately, to 40 L of the PSP peptide solution and incubated at 42 C overnight, to allow sufficient time fo r complete cleavage. The samples were filtered and analyzed by the HPLC-PSP assay to determine the retention times and peak areas of their peptides. In figure 3-17, the cr ude whole cell lysate sample incubated with PSP generated a single peak of interest at a retention time of 13.58 minutes. Moreover, when the peak area was integrated and applied to a standard curve of a linear regression of PSP peptide, the peptide concentration in the sample was approximately 1.20 mg/ml,

PAGE 43

34 indicating that PSP cleavage did not occur. Likewise, figure 3-18 shows that the cleared supernatant sample incubated with PSP pep tide solution generated a single peak of interest at a retention time of 13.45 minutes, which is the approximate retention time for the uncleaved PSP peptide. Thus, there was no proteolysis of PSP by the cloned, expressed mGK22. These results may not be une xpected, since the assa y to this point was carried out using the pro-mGK22 form. Moreover, Matsui et al have reported that the pro-mGK27 form was also inactive.

PAGE 44

35 Figure 3-1. Photograph of rep lica-plated nitrocellulose membrane hybridized with oligonucleotide probe from cDNA libra ry screening. Membranes contain replica-plated bacterial colonies transformed with plasmids containing cDNA library inserts from NOD.B10.H2b mouse submandibular tissue. The membranes were incubated with a Di g-labeled oligonucleotide probe, whose sequence was identical to N-terminal of mGK22, and detect ed with alkaline phosphatase conjugated antibody. Positive co lonies appear as darkened spots on a heterogeneous background. Colonies we re plated at a concentration of 100 CFU/ml.

PAGE 45

36 Gene Retrieved from cDNA Library Found mutant form in cDNA Library Frequency in cDNA Library Screening Frequency of Mutant in cDNA Library Screening Amino Acid Substitution Result of Mutation MGK6 Yes Yes 10 3 Pro123 Lys123 MGK9 Yes No 3 NA NA MGK26 Yes Yes 3 2 Lys112 Pro112 MGK22 No No NA NA NA Table1. Summary of results from screen ing of cDNA library made from NOD.B10.H2b mouse submandibular tissue. NA= not applicable

PAGE 46

37 mGK22/pET ATGCGGGGTTCTCATCATCATCATCATCATGGTATGGCTAGCATGACTGGTGGACA 56 mGK22/pET GCAAATGGGTCGGGATCTGTACGACGATGACGATAAGGATCATCCCTTCACC 108 mGK22/pET GCAC 112 mGK22 ATGAGGTTCCTGATCCTGTTCCTAACCCTGTCCCTAGGAGGGATTGATGCTGCAC 55 mGK6 ATGAGGTTCCTGATCCTGTTCCTAGCCCTGTCCCTAGGAGGGATTGATGCTGCAC 55 mGK9 ATGAGGTTCCTGATCCTGTTCCTAGCCCTGTCCCTAGGAGGGATTGATGCTGCAC 55 mGK26 ATGTGGTTCCTGATCCTGTTCCCAGCCCTGTCCCTAGGAGGGATTGATGCTGCAC 55 mGK22/pET CTCCTGTCCAGTCTCGA ATACTTGGAGGATTTAAATGTGAGAAGAATTCCCAACCC 168 mGK22 CTCCTGTCCAGTCTCGA ATACTTGGAGGATTTAAATGTGAGAAGAATTCCCAACCC 111 mGK6 CTCCTGTCCAGTCTCGA ATTGTTGGAGGATTTAACTGTGAGAAGAATTCCCAGCCC 111 mGK9 CTCCTGTCCACTCTCGA ATTGTTGGAGGATTTAAATGTGAGAAGAATTCCCAACCC 111 mGK26 CTCCTCTCCAGTCTCGG GTGGTTGGAGGATTTAACTGTGAGAAGAATTCCCAACCC 111 GK22/pET TGGCAGGTGGCTGTGTACTACTTAGATGAGTACCTATGCGGGGGAGTCCTGTTGGA 224 mGK22 TGGCAGGTGGCTGTGTACTACTTAGATGAGTACCTATGCGGGGGAGTCCTGTTGGA 167 mGK6 TGGCAAGTGGCTGTGTACCGCTTCACCAAATATCAATGTGGGGGTATCCTGCTGAA 167 mGK9 TGGCATGTGGCTGTGTACCGTTACAACGAATATATATGCGGGGGTGTCCTGTTGGA 167 mGK26 TGGCAGGTGGCTGTGTACTACCAAAAGGAACACATTTGTGGGGGTGTCCTGTTGGA 167 mGK22/pET CCGCAACTGGGTTCTCACAGCTGCCCACTGCTATGAAGACAAGTATAATATTTGGC 280 mGK22 CCGCAACTGGGTTCTCACAGCTGCCCACTGCTATGAAGACAAGTATAATATTTGGC 223 mGK6 CGCCAACTGGGTTCTCACAGCTGCCCACTGCCATAATGACAAGTACCAGGTGTGGC 223 mGK9 TGCCAACTGGGTTCTCACAGCTGCCCACTGCTATTACGAAGAGAACAAGGTTTCCC 223 mGK26 CCGCAACTGGGTTCTCACAGCTGCCCACTGCTATGTCGACCAGTATGAGGTTTGGC 223 mGK22/pET TGGGCAAAAACAAGCTATTCCAAGATGAACCCTCTGCTCAGCACCGATTGGTCAGC 336 mGK22 TGGGCAAAAACAAGCTATTCCAAGATGAACCCTCTGCTCAGCACCGATTGGTCAGC 279 mGK6 TGGGCAAAAACAACTTTTTGGAGGATGAACCCTCTGCCCAACACCGGCTTGTCAGC 279 mGK9 TAGGAAAAAACAACCTATACGAAGAGGAACCCTCTGCTCAGCACCGATTGGTCAGC 279 mGK26 TGGGCAAAAACAAGTTATTCCAAGAGGAACCCTCTGCTCAGCACCGATTGGTCAGC 279 mGK22/pET AAAAGCTTCCCTCATCCTGACTTCAACATGAGCCTCC TCCAAAGTGTA CC 392 mGK22 AAAAGCTTCCCTCATCCTGACTTCAACATGAGCCTCC TCCAAAGTGTA CC 335 mGK6 AAAGCCATCCCTCACCCTGACTTCAACATGAGCCTCC TGAATGAGCACACCCCACA 335 mGK9 AAAAGCTTCCTTCACCCTGGCTACAACAGGAGCCTCC ATAGAAACCACATCCGACA 335 mGK26 AAAAGCTTCCCTCACCCTGGCTTCAACATGAGCCTCC TGATGCTTCAAACAACACC 335 mGK22/pET TACTGGGGCC GACTTAAGCAATGACCTGATGCTGCTCCGCCTCAGCAAGCCTGCTG 448 mGK22 TACTGGGGCC GACTTAAGCAATGACCTGATGCTGCTCCGCCTCAGCAAGCCTGCTG 391 mGK6 ACCTGAGGAT GACTACAGCAATGACCTGATGCTGCTCCGCCTCAAAAAGCCTGCTG 391 mGK9 TCCTGAGTAT GACTACAGCAATGACCTGATGCTGCTCCGCCTCAGCAAGCCTGCTG 391 mGK26 TCCTGGGGCT GACTTCAGCAATGACCTGATGCTGCTCCGCCTCAGCAAGCCTGCTG 391 mGK22/pET ACATCACAGATGTTGTGAAGCCCATCGACCTGCCCACTACGGAGCCCAAGCTGGGG 504 mGK22 ACATCACAGATGTTGTGAAGCCCATCGACCTGCCCACTACGGAGCCCAAGCTGGGG 447 mGK6 ACATCACAGATGTTGTGAAGCCCATCGACCTGCCCACTGAGGAGCCCAAGCTGGGG 447 mGK9 ACATCACAGATGTTGTGAAGCCCATCGCCCTGCCTACTGAGGAGCCCAAGCTGGGG 447 mGK26 ACATCACAGATGTTGTGAAGCCCATCGCCCTGCCCACAAAGGAGCCCAAGCCGGGG 447 Figure 3-2. Alignment of nucleotide seque nces from cDNA library screening. The nucleotide sequences obtained from th e cDNA library screening were aligned for comparison. The sequences donated by the pET200 D-topo plasmid are in red. The Pre-pro zymogen peptide nucleo tide sequences are represented in blue. The nucleotide regions with high variability are represented in orange. The identities of the sequences are: mGK22/pETPCR product using glandular kallikrein22 specific pr imers and cDNA from NOD.B10.H2b submandibular tissue as a template ligated into pET200/D-topo plasmid, mGK22Mus musculus glandular kallikrein 22(AAN78419.1), mGK6Mus musculus glandular kallikrein 6(NP_034769.4), mGK9Mus musculus glandular kallikrein 9(NP_034246.1), mGK26Mus musculus glandular kallikrein 26(NP_034774.1).

PAGE 47

38 mGK22/pET AGCACATGCCTAGCCTCAGGCTGGGGCAGCATTAACCAGTTAATATACCAAAACCC 560 mGK22 AGCACATGCCTAGCCTCAGGCTGGGGCAGCATTAACCAGTTAATATACCAAAACCC 503 mGK6 AGCACATGCCTAGCCTCAGGCTGGGGCAGCATTACACCCGTCAAATATGAATACCC 503 mGK9 AGCACATGCCTTGCCTCAGGCTGGGGCAGCACTACACCTTTCAAGTTCCAAAATGC 503 mGK26 AGCACATGCCTAGCCTCAGGCTGGGGCAGCATTACACCCACAAGATGGCAAAAGTC 503 mGK22/pET AAATGATCTCCAGTGTGTGTCCATCAAGCTCCATCCTAATGAGGTCTGTGTGAAAG 616 mGK22 AAATGATCTCCAGTGTGTGTCCATCAAGCTCCATCCTAATGAGGTCTGTGTGAAAG 559 mGK6 AGATGAGCTCCAGTGTGTGAACCTCAAGCTCCTGCCTAATGAGGACTGTGCCAAAG 559 mGK9 AAAAGATCTCCAGTGTGTGAACCTCAAGCTCCTGCCTAATGAGGACTGTGGCAAAG 559 mGK26 AGATGATCTTCAGTGTGTGTTCATCACGCTCCTCCCCAATGAGAACTGTGCCAAAG 559 mGK22/pET CCCATATACTGAAGGTGACAGATGTCATGCTGTGTGCAGGAGAGATGAATGGAGGC 672 mGK22 CCCATATACTGAAGGTGACAGATGTCATGCTGTGTGCAGGAGAGATGAATGGAGGC 615 mGK6 CCCACATAGAGAAGGTGACAGATGACATGTTGTGTGCAGGAGATATGGATGGAGGC 615 mGK9 CCCACATAGAGAAGGTGACAGATGTCATGCTGTGTGCAGGAGAGACAGATGGAGGC 615 mGK26 TCTACCTACAGAAAGTCACAGATGTCATGCTGTGTGCAGGAGAGATGGGTGGAGGC 615 mGK22/pET AAAGACACTTGTAAGGGAGACTCAGGAGGCCCACTGATCTGTGATGGTGTTCTACA 728 mGK22 AAAGACACTTGTAAGGGAGACTCAGGAGGCCCACTGATCTGTGATGGTGTTCTACA 671 mGK6 AAAGACACTTGTGCGGGTGACTCAGGAGGCCCACTGATCTGTGATGGTGTTCTCCA 671 mGK9 AAAGACACTTGCAAGGGAGACTCAGGAGGCCCACTGATCTGTGATGGTGTTCTCCA 671 mGK26 AAAGACACTTGTGCGGGTGACTCCGGAGGCCCACTGATTTGTGATGGTATTCTCCA 671 mGK22/pET AGGTATCACATCATGGGGCTCTACCCCATGTGGTGAACCCAATGCACCGGCCATCT 784 mGK22 AGGTATCACATCATGGGGCTCTACCCCATGTGGTGAACCCAATGCACCGGCCATCT 727 mGK6 AGGTATCACATCATGGGGCCCTAGCCCTTGCGGTAAACCCAATGTGCCGGGTATCT 727 mGK9 AGGTATCACATCATGGGGCTTTACCCCATGTGGTGAACCCAAAAAGCCGGGCGTCT 727 mGK26 AGGAACCACATCAAATGGCCCTGAACCATGCGGTAAACCTGGTGTACCAGCCATCT 727 mGK22/pET ACACCAAACTTATTAAGTTTACCTCCTGGATAAAAGACACTATGGCCAAAAACCCCTGA 837 mGK22 ACACCAAACTTATTAAGTTTACCTCCTGGATAAAAGACACTATGGCCAAAAACCCCTGA 780 mGK6 ACACCAGAGTTTTAAATTTCAACACCTGGATAAGAGAAACTATGGCTGAAAATGACTGA 786 mGK9 ACACCAAACTTATTAAGTTTACCTCCTGGATCAAAGACACTATGGCAAAAAACCTCTGA 786 mGK26 ACACCAACCTTATTAAGTTCAACTCCTGGATAAAAGATACTATGATGAAAAATGCCTGA 786 Figure 3-2 Continued

PAGE 48

39 mGK22/pET MRGSHHHHHHGMASMTGGQQMGRDLYDDDDKDHPFT APPVQSR IL GGFKCEKN 53 mGK22 MRFLILFLTLSLGGIDAAPPVQSR IL GGFKCEKN 34 mGK6 MRFLILFLALSLGGIDAAPPVQSR IV GGFNCEKN 34 mGK9 MRFLILFLALSLGGIDAAPPVHSR IV GGFKCEKN 34 mGK26 MWFLILFPALSLGGIDAAPPLQSR VV GGFNCEKN 34 pre • pro mGK22/pET SQPW QVAVYYLDEYLCGGVLLDRNWVLTAAHCYEDKYNIWLGKNKLFQDEPSA 106 mGK22 SQPW QVAVYYLDEYLCGGVLLDRNWVLTAAHCYEDKYNIWLGKNKLFQDEPSA 87 mGK6 SQPW QVAVYRFTKYQCGGILLNANWVLTAAHCHNDKYQVWLGKNNFLEDEPSA 87 mGK9 SQPW HVAVYRYNEYICGGVLLDANWVLTAAHCYYEENKVSLGKNNLYEEEPSA 87 mGK26 SQPW QVAVYYQKEHICGGVLLDRNWVLTAAHCYVDQYEVWLGKNKLFQEEPSA 87 mGK22/pET QHRLVSKSFPHPDFNMS LLQSV PTGA DLSNDLMLLRLSKPADITDVVKPID 159 mGK22 QHRLVSKSFPHPDFNMS LLQSV PTGA DLSNDLMLLRLSKPADITDVVKPID 140 mGK6 QHRLVSKAIPHPDFNMS LLNEHTPQPED DYSNDLMLLRLKKPADITDVVKPID 140 mGK9 QHRLVSKSFLHPGYNRS LHRNHIRHPEY DYSNDLMLLRLSKPADITDVVKPIA 140 mGK26 QHRLVSKSFPHPGFNMS LLMLQTTPPGA DFSNDLMLLRLSKPADITDVVKPIA 140 mGK22/pET LPTTEPKLGSTCLASGWGSINQLIYQNPNDLQCVSIKLHPNEVCVKAHILKVT 212 mGK22 LPTTEPKLGSTCLASGWGSINQLIYQNPNDLQCVSIKLHPNEVCVKAHILKVT 193 mGK6 LPTEEPKLGSTCLASGWGSITPVKYEYPDELQCVNLKLLPNEDCAKAHIEKVT 193 mGK9 LPTEEPKLGSTCLASGWGSTTPFKFQNAKDLQCVNLKLLPNEDCGKAHIEKVT 193 mGK26 LPTKEPKPGSTCLASGWGSITPTRWQKSDDLQCVFITLLPNENCAKVYLQKVT 193 mGK22/pET DVMLCAGEMNGGKDTCKGDSGGPLICDGVLQGITSWGSTPCGEPNAPAIYTKL 265 mGK22 DVMLCAGEMNGGKDTCKGDSGGPLICDGVLQGITSWGSTPCGEPNAPAIYTKL 246 mGK6 DDMLCAGDMDGGKDTCAGDSGGPLICDGVLQGITSWGPSPCGKPNVPGIYTRV 246 mGK9 DVMLCAGETDGGKDTCKGDSGGPLICDGVLQGITSWGFTPCGEPKKPGVYTKL 246 mGK26 DVMLCAGEMGGGKDTCAGDSGGPLICDGILQGTTSNGPEPCGKPGVPAIYTNL 246 mGK22/pET IKFTSWIKDTMAKNP 278 mGK22 IKFTSWIKDTMAKNP 259 mGK6 LNFNTWIRETMAEND 261 mGK9 IKFTSWIKDTMAKNL 261 mGK26 IKFNSWIKDTMMKNA 261 Figure3-3. Alignment of translated gene sequences from cDNA library Screening. The nucleotide sequences obtained from the cDNA library screening were translated into their corresponding amino acid sequences a nd aligned. The sequences donated by the pET200 D-topo plasmid are in red. The Pre-pro zymogen peptide sequences are represented in blue. The limits of the pre a nd pro regions are indicated by the arrows located below the N-terminal region of the sequences. The mature glandular kallikrein begins at amino acid position 25 (position 44 of mGK22/pET). The green sequences represent the N-terminal region of the kallikrein proteins with the greatest homology. The regions with high variabilit y are represented in orange. The identities of the sequences are: mGK22/pETPC R product using glandular kallikrein22 specific primers and cDNA from NOD.B10.H2b submandibular tissue as a template ligated into pET200/D-topo plasmid, mGK22Mus musculus glandular kallikrein 22(AAN78419.1), mGK6Mus musculus glandular kallikrein 6(NP_034769.4), mGK9Mus musculus glandular kallikrein 9(NP_034246.1), mGK26Mus musculus glandular kallikrein 26(NP_034774.1).

PAGE 49

40 mGK6/cDNA ATGAGGTTCCTGATCCTGTTCCTAGCCCTGTCCCTAGGAGGGATTGATGCTGCACCTC 58 mGK6 ATGAGGTTCCTGATCCTGTTCCTAGCCCTGTCCCTAGGAGGGATTGATGCTGCACCTC 58 mGK6/cDNA CTGTCCAGTCTCGAATTGTTGGAGGATTTAACTGTGAGAAGAATTCCCAGCCCTGGCA 116 mGK6 CTGTCCAGTCTCGAATTGTTGGAGGATTTAACTGTGAGAAGAATTCCCAGCCCTGGCA 116 mGK6/cDNA AGTGGCTGTGTACCGCTTCACCAAATATCAATGTGGGGGTATCCTGCTGAACGCCAAC 174 mGK6 AGTGGCTGTGTACCGCTTCACCAAATATCAATGTGGGGGTATCCTGCTGAACGCCAAC 174 mGK6/cDNA TGGGTTCTCACAGCTGCCCACTGCCATAATGACAAGTACCAGGTGTGGCTGGGCAAAA 232 mGK6 TGGGTTCTCACAGCTGCCCACTGCCATAATGACAAGTACCAGGTGTGGCTGGGCAAAA 232 mGK6/cDNA ACAACTTTTTGGAGGATGAACCCTCTGCCCAACACCGGCTTGTCAGCAAAGCCATCCC 290 mGK6 ACAACTTTTTGGAGGATGAACCCTCTGCCCAACACCGGCTTGTCAGCAAAGCCATCCC 290 mGK6/cDNA TCACCCTGACTTCAACATGAGCCTCCTGAATGAGCACACCCCACAACCTGAGGATGAC 348 mGK6 TCACCCTGACTTCAACATGAGCCTCCTGAATGAGCACACCCCACAACCTGAGGATGAC 348 mGK6/cDNA TACAGCAATGACCTGATGC C GCTCCGCCTCAAAAAGCCTGCTGACATCACAGATGTTG 406 mGK6 TACAGCAATGACCTGATGC T GCTCCGCCTCAAAAAGCCTGCTGACATCACAGATGTTG 406 mGK6/cDNA TGAAGCCCATCGACCTGCCCACTGAGGAGCCCAAGCTGGGGAGCACATGCCTAGCCTC 464 mGK6 TGAAGCCCATCGACCTGCCCACTGAGGAGCCCAAGCTGGGGAGCACATGCCTAGCCTC 464 mGK6/cDNA AGGCTGGGGCAGCATTACACCCGTCAAATATGAATACCCAGATGAGCTCCAGTGTGTG 522 mGK6 AGGCTGGGGCAGCATTACACCCGTCAAATATGAATACCCAGATGAGCTCCAGTGTGTG 522 mGK6/cDNA AACCTCAAGCTCCTGCCTAATGAGGACTGTGCCAAAGCCCACATAGAGAAGGTGACAG 580 mGK6 AACCTCAAGCTCCTGCCTAATGAGGACTGTGCCAAAGCCCACATAGAGAAGGTGACAG 580 mGK6/cDNA ATGACATGTTGTGTGCAGGAGATATGGATGGAGGCAAAGACACTTGTGCGGGTGACTC 638 mGK6 ATGACATGTTGTGTGCAGGAGATATGGATGGAGGCAAAGACACTTGTGCGGGTGACTC 638 mGK6/cDNA AGGAGGCCCACTGATCTGTGATGGTGTTCTCCAAGGTATCACATCATGGGGCCCTAGC 696 mGK6 AGGAGGCCCACTGATCTGTGATGGTGTTCTCCAAGGTATCACATCATGGGGCCCTAGC 696 mGK6/cDNA CCTTGCGGTAAACCCAATGTGCCGGGTATCTACACCAGAGTTTTAAATTTCAACACCT 754 mGK6 CCTTGCGGTAAACCCAATGTGCCGGGTATCTACACCAGAGTTTTAAATTTCAACACCT 754 mGK6/cDNA GGATAAGAGAAACTATGGCTGAAAATGACTGA 786 mGK6 GGATAAGAGAAACTATGGCTGAAAATGACTGA 786 Figure 3-4. Alignment of nucleotide seque nces of mGK6 from screening of NOD.B10.H2b submandibular tissue cDNA library and Mus musculus mGK6 (BC010754). The nucleotide sequence obtained from the cDNA library screening was aligned with Mus musculus mGK6 (BC010754) for comparison. The sequence identities are: mGK6/cDNAthe nucleotide sequence obtained from the cDNA library screening; mGK6Mus musculus glandular kallikrein 6 (BC010754). The sequences are identical with the exception of a single nucleotide discrepa ncy, which is represented in red and indicated by an arrow.

PAGE 50

41 mGK6/cDNA MRFLILFLALSLGGIDAAPPVQSRIVGGFNCEKNSQPWQVAVYRFTKYQCGGI 53 mGK6 MRFLILFLALSLGGIDAAPPVQSRIVGGFNCEKNSQPWQVAVYRFTKYQCGGI 53 mGK6/cDNA LLNANWVLTAAHCHNDKYQVWLGKNNFLEDEPSAQHRLVSKAIPHPDFNMSLL 106 mGK6 LLNANWVLTAAHCHNDKYQVWLGKNNFLEDEPSAQHRLVSKAIPHPDFNMSLL 106 mGK6/cDNA NEHTPQPEDDYSNDLM P LRLKKPADITDVVKPIDLPTEEPKLGSTCLASGWGS 159 mGK6 NEHTPQPEDDYSNDLM L LRLKKPADITDVVKPIDLPTEEPKLGSTCLASGWGS 159 mGK6/cDNA ITPVKYEYPDELQCVNLKLLPNEDCAKAHIEKVTDDMLCAGDMDGGKDTCAGD 212 mGK6 ITPVKYEYPDELQCVNLKLLPNEDCAKAHIEKVTDDMLCAGDMDGGKDTCAGD 212 mGK6/cDNA SGGPLICDGVLQGITSWGPSPCGKPNVPGIYTRVLNFNTWIRETMAEN 260 mGK6 SGGPLICDGVLQGITSWGPSPCGKPNVPGIYTRVLNFNTWIRETMAEN 260 Figure 3-5. Alignment of amino acid translati on of the mGK6 nucleotide sequences from screening of NOD.B10.H2b submandibular tissue cDNA library and amino acid sequence of Mus musculus mGK6 (BC010754). The nucleotide sequence obtained from the cDNA library sc reening was translated into the corresponding amino acid se quence and aligned with Mus musculus mGK6 (BC010754) amino acid sequence for comparison. The sequence identities are: mGK6/cDNAthe amino acid translation of the nucleotide sequence obtained from the cDNA library screening; mGK6Mus musculus glandular kallikrein 6(BC010754) amino acid seque nce. The sequences are identical with the exception of a single amino aci d discrepancy, which is represented in red and indicated by an arrow.

PAGE 51

42 mGK26/cDNA ATGTGGTTCCTGATCCTGTTCCCAGCCCTGTCCCTAGGAGGGATTGATGCTGCACCTC 58 mGK26 ATGTGGTTCCTGATCCTGTTCCCAGCCCTGTCCCTAGGAGGGATTGATGCTGCACCTC 58 mGK26/cDNA CTCTCCAGTCTCGGGTGGTTGGAGGATTTAACTGTGAGAAGAATTCCCAACCCTGGCA 116 mGK26 CTCTCCAGTCTCGGGTGGTTGGAGGATTTAACTGTGAGAAGAATTCCCAACCCTGGCA 116 mGK26/cDNA GGTGGCTGTGTACTACCAAAAGGAACACATTTGTGGGGGTGTCCTGTTGGACCGCAAC 174 mGK26 GGTGGCTGTGTACTACCAAAAGGAACACATTTGTGGGGGTGTCCTGTTGGACCGCAAC 174 mGK26/cDNA TGGGTTCTCACAGCTGCCCACTGCTATGTCGACCAGTATGAGGTTTGGCTGGGCAAAA 232 mGK26 TGGGTTCTCACAGCTGCCCACTGCTATGTCGACCAGTATGAGGTTTGGCTGGGCAAAA 232 mGK26/cDNA ACAAGTTATTCCAAGAGGAACCCTCTGCTCAGCACCGATTGGTCAGCAAAAGCTTCCC 290 mGK26 ACAAGTTATTCCAAGAGGAACCCTCTGCTCAGCACCGATTGGTCAGCAAAAGCTTCCC 290 mGK26/cDNA TCACCCTGGCTTCAACATGAGCCTCCTGATGCTTCAAACAACAC T TCCTGGGGCTGAC 348 mGK26 TCACCCTGGCTTCAACATGAGCCTCCTGATGCTTCAAACAACAC C TCCTGGGGCTGAC 348 mGK26/cDNA TTCAGCAATGACCTGATGCTGCTCCGCCTCAGCAAGCCTGCTGACATCACAGATGTTG 406 mGK26 TTCAGCAATGACCTGATGCTGCTCCGCCTCAGCAAGCCTGCTGACATCACAGATGTTG 406 mGK26/cDNA TGAAGCCCATCGCCCTGCCCACAAAGGAGCCCAAGCCGGGGAGCACATGCCTAGCCTC 464 mGK26 TGAAGCCCATCGCCCTGCCCACAAAGGAGCCCAAGCCGGGGAGCACATGCCTAGCCTC 464 mGK26/cDNA AGGCTGGGGCAGCATTACACCCACAAGATGGCAAAAGTCAGATGATCTTCAGTGTGTG 522 mGK26 AGGCTGGGGCAGCATTACACCCACAAGATGGCAAAAGTCAGATGATCTTCAGTGTGTG 522 mGK26/cDNA TTCATCACGCTCCTCCCCAATGAGAACTGTGCCAAAGTCTACCTACAGAAAGTCACAG 580 mGK26 TTCATCACGCTCCTCCCCAATGAGAACTGTGCCAAAGTCTACCTACAGAAAGTCACAG 580 mGK26/cDNA ATGTCATGCTGTGTGCAGGAGAGATGGGTGGAGGCAAAGACACTTGTGCGGGTGACTC 638 mGK26 ATGTCATGCTGTGTGCAGGAGAGATGGGTGGAGGCAAAGACACTTGTGCGGGTGACTC 638 mGK26/cDNA CGGAGGCCCACTGATTTGTGATGGTATTCTCCAAGGAACCACATCAAATGGCCCTGAA 696 mGK26 CGGAGGCCCACTGATTTGTGATGGTATTCTCCAAGGAACCACATCAAATGGCCCTGAA 696 mGK26/cDNA CCATGCGGTAAACCTGGTGTACCAGCCATCTACACCAACCTTATTAAGTTCAACTCCT 754 mGK26 CCATGCGGTAAACCTGGTGTACCAGCCATCTACACCAACCTTATTAAGTTCAACTCCT 754 mGK26/cDNA GGATAAAAGATACTATGATGAAAAATGCCTGA 786 mGK26 GGATAAAAGATACTATGATGAAAAATGCCTGA 786 Figure 3-6. Alignment of nucleotide seque nces of mGK26 fr om screening of NOD.B10.H2b submandibular tissue cDNA library and Mus musculus mGK26 (NM_010644). The nucleotide sequence obtained from the cDNA library screening was aligned with Mus musculus mGK26 (NM_010644) for comparison. The sequence identities are: mGK26/cDNAthe nucleotide sequence obtained from the cDNA library screening; mGK26Mus musculus glandular kallikrein 26(NM_010644). The sequences are identical, with the exception of a single nucleotide discrepa ncy, which is represented in red and indicated by an arrow.

PAGE 52

43 mGK26/cDNA MWFLILFPALSLGGIDAAPPLQSRVVGGFNCEKNSQPWQVAVYYQKEHICGGV 53 mGK26 MWFLILFPALSLGGIDAAPPLQSRVVGGFNCEKNSQPWQVAVYYQKEHICGGV 53 mGK26/cDNA LLDRNWVLTAAHCYVDQYEVWLGKNKLFQEEPSAQHRLVSKSFPHPGFNMSLL 106 mGK26 LLDRNWVLTAAHCYVDQYEVWLGKNKLFQEEPSAQHRLVSKSFPHPGFNMSLL 106 mGK26/cDNA MLQTT L PGADFSNDLMLLRLSKPADITDVVKPIALPTKEPKPGSTCLASGWGS 159 mGK26 MLQTT P PGADFSNDLMLLRLSKPADITDVVKPIALPTKEPKPGSTCLASGWGS 159 mGK26/cDNA ITPTRWQKSDDLQCVFITLLPNENCAKVYLQKVTDVMLCAGEMGGGKDTCAGD 212 mGK26 ITPTRWQKSDDLQCVFITLLPNENCAKVYLQKVTDVMLCAGEMGGGKDTCAGD 212 mGK26/cDNA SGGPLICDGILQGTTSNGPEPCGKPGVPAIYTNLIKFNSWIKDTMMKN 260 mGK26 SGGPLICDGILQGTTSNGPEPCGKPGVPAIYTNLIKFNSWIKDTMMKN 260 Figure 3-7. Alignment of amino acid transl ation of the mGK26 nucleotide sequences from screening of NOD.B10.H2b submandibular tissue cDNA library and amino acid sequence of Mus musculus mGK26 (NP_034774.1). The nucleotide sequence obtained from the cDNA library screening was translated into the corresponding amino acid sequence and aligned with Mus musculus mGK6 (NP_034774.1) amino acid sequen ce for comparison. The sequence identities are: mGK26/cDNAthe ami no acid translation of the nucleotide sequence obtained from the cDNA library screening; mGK26Mus musculus glandular kallikrein 26 (NP_034774.1) amino acid sequence. The sequences are identical with the exception of a single amino acid discrepancy, which is represented in red and indicated by an arrow.

PAGE 53

44 Figure3-8. Translated amino acid sequence mGK22/pET200 D-topo plasmid construct. The figure displays the amino acid tr anslation of the DNA sequence of the pET200 D-topo plasmid that contains mGK22 PCR product as an insert. The rectangular box outlines the leader sequence donated by the pET200 D-topo plasmid as indicated by the 6X histidine tag beginning at the 5th amino acid position. The mGK22 PCR product amino aci d sequence begins at amino acid position 37. Mature mGK22 begins at amino acid position 44 which corresponds to amino acid position 24 of the published mGK22 amino acid sequence.

PAGE 54

45 lane 1 2 3 4 5 Figure 3-9. Results of gel elect rophoresis analysis of P CR amplification using mGK22 specific primers. lane 11Kb ladder mo lecular weight marker; lane2mGK22 specific primers with first strand sy nthesis cDNA as a template; lane3mGK22 specific primers with transformed cDNA as a template; lane4mGK22 specific primers with no DNA( negative control); lane5control primers and control template DNA (Invitrogen pos itive control).

PAGE 55

46 lane 1 2 3 4 Figure3-10. Results of gel elect rophoresis analysis of plas mid constructs. lane11Kb ladder molecular weight marker; lane 2pET200 D-topo plasmid with mGK22 insert; lane3pET200 D-topo plasmid with PCR (lac Z) positive control; lane4empty pET200 D-topo plasmid.

PAGE 56

47 lane 1 2 3 4 5 Figure3-11. Results of gel electr ophoresis analysis of hindIII digested plasmid constructs. lane11Kb ladder molecular weight ma rker; lane2mGK 22 without hindIII; lane3PCR (lac Z) positive control; lane4mGK22 with hind III; lane5empty pET200 D-topo plasmid with hind III.

PAGE 57

48 lane 1 2 3 4 5 6 7 8 9 10 11 Figure 3-12. Results of SDS-PAGE analysis and commassie blue staining of crude whole cell lysates induced with IPTG with uninduced cr ude whole cell lysates lane1SDS PAGE molecular weight standards; lane 20.0 hours (preinduction); lane 3uninduced 1.0 hour; lane 4induced 1.0 hour; lane 5uninduced 2.0 hour ; lane 6induced 2.0 hour ; lane 7uninduced 3.0 hour; lane 8induced 3.0 hour; lane 9uni nduced 5.0 hour; lane 10induced 5.0 hour; lane11SDS PAGE mol ecular weight standards.

PAGE 58

49 lane 1 2 3 4 5 Figure 3-13. Analysis of crude whole cell lysa tes induced with IPTG using western blot analysis with anti-6Xhis tidine antibody. lane 10 .0 hours (pre-induction); lane 21.0 hour post induction; lane 22.0 hour post induction; lane 4empty; lane 5histidine tagged positive control.

PAGE 59

50 lane 1 2 3 4 5 6 Figure3-14. Analysis of pellet and supernatant of crude lysa tes induced with IPTG using western Blot analysis with anti-6Xhis tidine antibody. lane 10.0 hour (preinduction); lane 2-empty; lane3pe llet 1.0 hour post-induction; lane 4supernatant 1.0 hour post-induction; lane 5pellet 2.0 hour post-induction; lane 6supernatant 2.0 hour pos t-induction.

PAGE 60

51 Minutes Figure3-15. HPLC-PSP assay chromatogram of 40L of PSP peptide incubated with 40L of lysis buffer. The uncleaved, in tact PSP peptide retention time was 13.46 minutes.

PAGE 61

52 Minutes Figure3-16. HPLC-PSP assay chromatogram of 40L of PSP peptide incubated with 5 L of NOD.B10.H2b and 35L of lysis buffer. The retention times of the fragments of the cleaved PSP peptide were 9.16 minutes (peak 6) and 12.21 minutes (peak7).

PAGE 62

53 Minutes Figure3-17. HPLC-PSP assay chromatogram of 40L of PSP peptide incubated with 40 L of crude whole cell lysates from the mGK22 expression. This assay was performed on whole cell lysates prior to centrifugation. The re tention time of the uncleaved, intact PSP peptide was 13.58 minutes.

PAGE 63

54 Figure3-18. HPLC-PSP assay chromatogram of 40L of PSP peptide incubated with 40 L of supernatant of the crude cell lysates from the mGK22 expression. The retention time of the uncleaved, intact PSP peptide was 13.45 minutes.

PAGE 64

55 CHAPTER 4 DISCUSSION The identity of the unknown PSP proteolytic enzyme present in the salivas of the NOD mice exhibiting SjS symptoms remains an elusive target. Although previous data have pointed to mGK22 as the PSP cleaving culp rit in the NOD saliva, the data from the cDNA library screening of the NOD have re vealed the presence of other candidate proteins with known functional similarity a nd genetic homology to mGK22 as present in the submandibular glands of the NOD mouse. A majority of these proteins identified in the cDNA library screening belong to the same ka llikrein family of pr oteins that includes mGK22 such as: mGK6, mGK9, mGK26. Howe ver, there were a few non-kallikrein genes that were identified by the screening, e.g. salivary protein 1 and 2, PRL-inducible protein, cytochrome P450. The cDNA library screening failed to dete ct mGK22 in the cDNA library. However that could be the result of a sample screening whose size was too small to be representative of the complete cDNA library. Calculations of the screening confirm that approximately 6500 colonies of the 6.6 X 105 independent clones generated in the cDNA library were screened or about 1% of th e library. Possibly, the abundance of mGK22 in the cDNA library was not great enough for it to be detected by the screening. Thus, the small fraction of the cDNA library that was screened could have been the reason for the failure to detect mGK22. Further screening ma y detect the presence of a gene with a low copy number such as mGK22.

PAGE 65

56 Although the cDNA library screening failed to detect mGK22, the PCR amplification, using mGK22 specific primers, did detect mGK22 in the cDNA library. Moreover, the PCR amplificati on showed the presence of tw o different forms of mGK22 in the cDNA library. One of the PCR reactions produced a sequence that was identical to the published mGK22 sequence while anothe r, separate PCR reaction, produced a sequence that contained a one base pair de viation from the published mGK22 sequence. When the sequence was translated to its corresponding amino acid sequence, the base pair substitution resulted in an aspartic acid for asparagine amino acid substitution. This amino acid substitution is consistent with th e N-terminal amino acid sequence identified from a unique protein band of a purified saliva fraction shown to have PSP cleaving activity from previous research (Day, 2002). Th e two forms could represent alleles of the mGK22 gene or the existence of an unide ntified glandular kallik rein gene. Errors generated during the PCR process are unlikely, since the protein enc oded by this allele was identified in mice (Day, 2002). Further sequencing of genomic DNA in the NOD mouse should be performed to detect the pr esence of both forms of the mGK22 gene. Genomic DNA sequencing of the NOD mouse woul d help to understand the nature of the mGK22 discrepancies th at have been found. The cDNA library screening showed the presence of mGK 6, mGK9, and mGK26 in the NOD saliva (Table 1). The N-termin al amino acid sequences of mGK6, mGK9, and mGK26 share a high degree of homology with mGK22 (figure 3-2 and 3-3). This homology between these GKs and mGK22 may e xplain why they were detected with an oligonucleotide probe that was complementary to the back-translate d sequence of the Nterminus of mGK22.

PAGE 66

57 The functional similarity and homol ogy of mGK9, mGK22, and mGK26 strongly supports their candidacy as the enzyme re sponsible for PSP cleavage in the NOD saliva from previous research. Considering that mGK9, mGK22, and mGK26 share functional similarities by recognizing, bindi ng, and processing pre-EGF, it is probable that all three kallikreins could similarly bind and cleave PSP, if indeed mGK22 possesses that ability. However, it should be noted that mGK22Â’s PSP cleaving activity, at least in the promGK22 form, has yet to be proven. Therefore, future research should be broadly directed towards the isolation and assessment of mGK9, mGK22, and mGK 26 for PSP cleaving activity should the mGK22 form not show activity. The cDNA library screening al so revealed the presence of some interesting base pair polymorphisms of mGK6 (figures 3-4 and 3-5) and mGK26 (figures 3-6 and 3-7) that make them highly suspect as PSP cleav ing enzymes. The mutations were identified in several different screenings further s upporting the conclusion that these nucleotide substitutions are indeed mutations and not random base pair substitutions due to transcribing or PCR replica tion errors. Sequenc ing of genomic DNA of the NOD mouse could be used to characterize the base pa ir polymorphisms of mGK6 and mGK26 found in the cDNA library screening as mutations. These mutations could result in a gain of function, resulting in PSP cleaving activity for mGK6 and mGK26. The mutations of mGK6 and mGK26, identified by the cDNA libra ry screening, resulted in amino acid substitutions that involved a proline residue, when translated. The cy clic orientation and nature of a proline residue coul d be indicative of a beta turn or some other great structural deviation from the wild type isoform of the protein. Moreover, the mutations could confer a three-dimensional conformational ch ange in mGK6 and mGK26 such that a PSP

PAGE 67

58 cleaving function is gained. Further characterization by is olation or purification is required to ascertain the changes in activit y, if any, caused by these mutations. The cDNA library screening results genera ted from this thesis showed several candidate genes that may be responsible for the cleaving PSP in the NOD saliva. The proof of PSP cleaving ability and further binding characterization by these genes or mutations could help to confir m that one or more of them encode the protein responsible for PSP cleavage in the NOD saliv a observed by researchers. The data of the expression of mGK22 from this thesis shows that an inducible 6Xhistidine-tagged recombinant protein with a molecular mass equal to the mass of mGK22 was produced. The crude extract of th e cells that were transformed to produce mGK22 were assessed for PSP cleavage activity after they were shown to produce an inducible protein with a 6X histidine tag. Although the crude extract had no PSP cl eaving activity, mGK22 may still be the PSP protease. The inability of the crude ex tract, which contained mGK22, to cleave PSP may be the result of interference from some of the transformed bacterial cell constituents. Further purification could be performed to remove cell constituents that may be interfering with mGK22 PSP cl eaving activity. Another techni que would be to analyze the NOD saliva for PSP cleaving activity while in the presence of mGK22 cell extract that has been previously shown to be inactive. This technique would show if the secreted form of mGK22 was affected by some othe r component present in the cell extract resulting in inactivity of the PSP cleaving protein. Furthermore, the cell extrac ts ability to inhibit PSP cleaving activity in the NOD saliva could serve as a possible explanation for the crude extracts inab ility to cleave PSP.

PAGE 68

59 Another possible explanation fo r the crude extracts inabil ity to cleave PSP could be that mGK22 was expressed in an inactive fo rm due to the presence of the 7 amino acid pro sequence on the N-terminus of mGK22. Th e pro-sequence on the N-terminus of the mGK22 that was expressed in the transformed bacterial cells could be the cause of the inactivity of mGK22. Future experiments co uld be designed to express mGK22 in the mature form, without the 7 amino acids th at comprise the pro-sequence on the Nterminus of mGK22. Alternatively, the 6X histidine tag could have interfered with mGK22Â’s activity. Activation by enzymatic cleavage of the recomb inant protein expressed in this research could confer PSP cleaving activity in the crude extract. Tr ypsin cleavage at amino acid position 24 is thought to be important and neces sary step in the activ ation of GKs (Matsui et al ., 2000). Perhaps activation via cleavage with trypsin could confer PSP cleaving activity in the crude lysates. Moreover, the reformation of the disulfid e bridges present in mGK22 could have been inhibited by the prokaryotic expression sy stem thus preventing correct refolding and preventing the formation of required thre e-dimensional structures. Inappropriate refolding and incorrect three-dimensional c onformation of the protein could induce the formation of inclusion bodies. The inclusion bo dies are insoluble aggregates of proteins in E. coli The analysis of the insoluble pellet an d soluble supernatant of the crude lysates showed that the protein expressed was present mostly in the pellet of the crude lysate (figure 3-14). This sugges ts that although mGK22 was e xpressed in the transformed bacterial cells, much of it was probably in an insoluble form, which could result in expression of an inactive protein.

PAGE 69

60 Also, post-translational modi fications may be required for PSP cleaving activity in mGK22. Post-translational modi fications such as glycosylat ion, which may be necessary for PSP cleaving activity, do not occur in a prokaryotic expression system. The expression of mGK22 in a e ukaryotic expression system would produce a complete mGK22 with great similarity to the secreted form of mGK22 produced in the NOD saliva. The data generated by this research has shed some light into the identity of a PSP cleaving protease present in the saliva of the NOD mouse. Although mGK22 cannot be ruled out completely, it is less likely that it is the PSP proteas e, which suggests that the other kallikrein genes and possi bly kallikrein mutants that have been identified to be present in the NOD cDNA library may be the PSP protease. Further isolation, purification, and characterization of these ot her genes and mutants, as well as mGK22, are required to assess their ability to cleav e PSP. However, confirmation of PSP cleavage by the candidate genes named in this thesis does not prove that they are solely responsible for the PSP cleavage in NOD saliva pr eviously observed. The possibility that another unknown protein is responsible for the observed PSP cleavage in the NOD saliva cannot be ruled out yet. Puri fication and isolation, via ammo nium sulfate precipitation for example, directly from the saliva of the NOD mouse could identify the protease responsible for PSP cleavage in the NOD saliv a. Finally, once identified molecular techniques such as site-directed mutagenesi s and gene knockout mice could be developed to further characterize the nature of the PSP cleavage reac tion and its possible role in SjS pathogenesis.

PAGE 70

61 LIST OF REFERENCES Atkinson JC, Fox PC, Travis WD, Popek E, Katz RW, Balow JE, Pillemar SR. IgA rheumatoid factor and IgA containing immune complexes in primary Sjgren’s syndrome. J Rheumatol 1989; 16:1205-1210. Aziz KE, McCluskey PJ, Wakefield D,. Charact erisation of follicular dendretic cells in labial salivary glands of patients with primary Sjgren’s syndrome: comparison with tonsillar lymphoid follicles. Ann Rheum 1997; 56:104-143. Baum BJ, Dai Y, Hiramatsu Y, Horn VJ, Ambudkar IS. Signaling mechanisms that regulate saliva formation. Crit Rev Oral Biol Med 1993; 4:379-384. Blaber M, Isackson PJ, Bradshaw A, A complete cDNA sequence for the major epidermal growth factor binding protein in the male mouse submandibular gland. Biochemistry 1987; 26:6742-6749. Brayer J, Lowry J, Cha S, Robinson CP, Yamachika S, Peck AB, et al Alleles from chromosomes 1 and 3 of NOD mice combine to influence Sjgren’s syndrome-–ike autoimmune exocrinopathy. J Rheumatol 2000; 27:1896-1904. Bymaster FP, McKinzie DL, Felder CC, We ss J. Use of M1-M5 muscuranic receptor knockout mice as novel tolls to delineate th e physiological roles of the muscuranic cholinergic system. Neurchem Res 2003; 28:437-442. Cha S, Peck AB, Humpherys-Beher MG Progress in understanding autoimmune exocrinopathy using the non-obese diabetic mouse: an update. Crit Rev Oral Biol Med 2002; 13(1):4-16. Carnaud C, Legrand B, Olivi M, Peterson LB, Wicker LS, Bach JF. Acquired allotolerance to major histocompatiblity antigens indifferently contributes to preventing diabetes development in non-obese diabetic (NOD) mice. J Autoimmun 1992; 5:591-601. Day, J M. Development of an HPLC based assay used for the characterization and identification of an unknown protease pr esent in the saliva of the NOD.B10.H2b mouse model. University of Florida 2002. Delaleu N, Jonsson R, Koller MM. Review Sjgren’s syndrome. Eur J Oral Sci 2005; 113:101-113.

PAGE 71

62 Drinkwater CC, Evans BA, Richards R I. Genes encoding epidermal growth factor binding proteins. Biochemistry 1987; 26:6750-6756. Evans BA, Drinkwater CC, and Richards RR. Mouse glandular kallikrein genes. J Biol Chem 1987; 262:8027-8034. Geetha C, Venkatesh SG, Fasciotto Dunn BH, Gorr SU. Expressi on and anti-bacterial activity of human parotid secretory protein (PSP). Biochemical Society Transactions 2003; 31,815-818. Gomis-Ruth FX, Bayes A, Soitiropoulou G, Pamp alkis G, Tsetsenis T, Villegas V, Aviles FX, and Coll M. The structure of human prokallikrein 6 reveals a novel activation mechanism for the kallikrein family J Biol Chem 2002; 277:27273-27281. Matsui H, Moriyama A, Takahashi T. Cl oning and characteriza tion of mouse klk27, a novel tissue kallikrein expr essed in testicular Leydig cells and exhibiting chymotrypsin-like specificity. Eur J Biochem 2000; 267:6858-6865. Olsson AY, Lundwall A. Organization and evolut ion of the glandular kallikrein locus in Mus musculus Biochemical and Biophysical Research Communications 2002; 299:305-311. Robinson CP, Cornelius J, Bounous DE, Yamamoto H, Humpherys-Beher MG, and Peck AB. Characterizations of the changing lymphocyte populations and cytokine expression in the exocrine ti ssues of autoimmune NOD mice. Autoimmunity 1998; 27:29-44. Robinson CP, Yuamamoto H, Humpherys-Beher MG, and Peck AB. Genetically programmed development of the saliv ary gland abnormalities in the NOD (nonobese diabetic)scid mouse in the absence of detect able lymphocytic infiltration: a potential trigger for sial oadentitis of NOD mice. Clinical Immunol and Immunopathol 1996; 79:50-59. Robinson CP, Brayer J, Yamachika S, Esch TR, Peck AB, Stewart CA, Peen E, Johsson R, Humpherys-Beher MG. Transfer of human serum IgG to non-obese diabetic Ig null mice reveals a role for autoantibodies in the loss of secr etory function of exocrine tissues in SjgrenÂ’s syndrome. Proc Natl Acad Sci USA 1998; 95:75387543. Robinson CP, Bounous DI, Alford CE, Peck AB, Humpherys-Beher MG. Aberrant expression and potential function for paro tid secretory protein (PSP) in the nonobese diabetic (NOD) mouse. Adv Exp Med Biol 1998; 438:925-930. Robinson CP, Yamachika S, Bounous DI, Brayer J, Jonsson R, Holmdahl R, et al. A novel NOD-derived murine model of primary SjgrenÂ’s syndrome. Arthritis Rheum 1998c; 41:150-156.

PAGE 72

63 Takada K, Takiguchi M, Konno A, Inaba M. Au toimmunity against a tissue kallikrein in IQI/Jic mice. J Biol Chem 2005; 280:3982-3988. Voulgarelis M, Moutspoulos HM. Lymphoprolif eration in autoimmunity and SjgrenÂ’s syndrome. Curr Rheumtol Rep 2003; 5:317-323. Zeher M, Adany R, Nagy G, Gomez R, Szeg edi G. Macrophage containing factor XIII subunit a in salivary gl ands of patients with SjgrenÂ’s syndrome. J Invest Allergol Clin Immunol 1991 ; 1:261-265.

PAGE 73

64 BIOGRAPHICAL SKETCH My name is Javier Brian Alvarado and I am a 36 year old graduate student living in Gainesville, Florida. I am the only child of Javier and Kathy Alvarado. I attended the University of South Florida in Tampa, Flor ida, where I earned a bachelorÂ’s degree in biology in 1995. After graduating, I expanded my work e xperience through employment in private industry. I have had work experience as a laboratory analyst, researcher, teacher and athletic coach over a period of 7 years. In the future, I plan to reside in Gainesvi lle, Florida, where I will raise my family and endeavor to expand my car eer as a molecular biologist.