Functional and inhibitory studies on cystathionine gamma lyase

An expression and purification system for the human CSE enzyme was also developed to enable a more reliable method of screening of the inhibitor candidates via a purified protein assay,

FUNCTIONAL AND INHIBITORY STUDIES ON

ACKNOWLEDGEMENTS

I would like to thank my supervisor, Dr Tan Choon Hong, as well as Dr Deng Lih-Wen,

Dr Jayaraman Sivaraman and Prof Philip Keith Moore for the use of their laboratory facilities and their invaluable guidance and support in this project My appreciations also

go towards Miss Guan Yanyi, Mr Zhao Yujun, Mr Fu Xiao, Mr Tan Yaw Sing and Miss Chua Jia Hui for their effort in synthesizing the inhibitor candidates, Mr Sun Qingxiang for his guidance on X-ray crystallography studies, Ms Wang Xiaoning, Ms Liu Jie and

Mr Cheng Fei for their patience in guiding me on molecular cloning and immunoblotting techniques, as well as members of Dr Tan’s, Dr Deng’s, Dr Sivaraman and Prof Moore’s labs for their help in this project Last but not least, I would like to express my gratitude towards my husband, Mr Bryan Lim for his constant encouragement and emotional support in my pursuance of the Masters degree

TABLE OF CONTENTS

Acknowledgements i 

Table of Contents ii 

Summary v 

List of Publications vii 

List of Tables viii 

List of Figures ix 

List of Symbols xiv 

1 Introduction 1 

2 Tissue H2S assay for screening inhibitors of H2S production 5 

2.1 Objectives 5 

2.2 Experimental 5 

2.3 Results and discussion 7 

2.4 Conclusion 13

3 Cloning and expression of recombinant human CSE 14 

3.1 Objectives 14 

3.2 Experimental 15 

3.2.1 Preparation of recombinant human CSE plasmids 15 

3.2.2 Mammalian expression of FLAG-tagged CSE 16 

3.2.3 Western blotting 16 

3.2.4 Optimization of bacterial expression of human CSE 17 

3.2.5 Miniaturized assay of H2S production 18 

3.2.6 Bacterial expression of human CSE enzyme 19 

3.2.7 Purification of human CSE enzyme 19 

3.3 Results and discussion 22 

3.3.1 Construction of recombinant human CSE plasmids 22 

3.3.2 Expression and determination of H2S synthesizing activity of recombinant human CSE 23 

3.3.3 Purification of human CSE enzyme 26 

3.4 Conclusion 32

4 Purified protein assay screen for inhibitors of H2S production 33 

4.1 Objectives 33 

4.2 Experimental 33 

4.2.1 Kinetics of H2S production from purified recombinant human CSE 33 

4.2.2 Protein H2S assay screen for inhibitors of H2S production 34 

4.2.3 Determination of H2S standard curve 35 

4.2.4 Optimized protein H2S assay screen for inhibitors of H2S production 36 

4.3 Results and discussion 37 

4.3.1 Kinetics of H2S production from purified recombinant human CSE 37 

4.3.2 Protein H2S assay screen of various commercially available inhibitor candidates 39 

4.3.2 Protein H2S assay screen of various chemically synthesized inhibitor candidates 42 

4.3.4 Optimization of protein H2S assay screen 44 

4.3.5 Redetermination of H2S standard curve 49 

4.3.6 Redetermination of inhibition potency of inhibitor candidates 50 

4.4 Conclusion 54 

5 Elucidation of the three-dimensional structure of human CSE 55 

5.1 Objectives 55 

5.2 Experimental 55 

5.2.1 Determination of protein homogeneity via DLS experiments 55 

5.2.2 Screening and optimization of crystallizing conditions 56 

5.2.3 X-ray diffraction and structure determination 56 

5.2.4 Proteolytic cleavage of CSE 57 

5.3 Results and discussion 59 

5.3.1 Optimization of protein concentration for crystallization studies 59 

5.3.2 Screening and optimization of crystallizing conditions for CSE 60 

5.3.3 Screening and optimization of crystallizing conditions for CSE-inhibitor complexes 64 

5.3.4 Proteolytic cleavage of CSE 69 

5.4 Conclusion 71 

6 Mechanism of H2S production 73 

Objectives 73 

Experimental 76 

6.2.1 Assay of H2S synthesis from various in vivo sulfur-containing compounds 76 

6.2.2 Cloning of pET-22b(+)_CSE 76 

6.2.3 Bacterial expression and purification of polyhistidine-tagged (His-tagged) CSE 76 

6.2.4 Optimization of bacterial induction conditions for His-tagged CSE 77 

6.2.5 Preparation of mutant CSE clones 78 

6.2.6 Bacterial expression of mutant GST-tagged CSE proteins 78 

6.2.7 Optimized procedure for purification of GST-tagged mutant and wild-type

CSE 79 

6.2.8 Analysis of protein secondary structure via circular dichroism (CD) measurements 80 

6.2.9 Comparison of the H2S synthesizing activities of the CSE mutant proteins 80 

6.2.10 Kinetics of H2S production under varying exogenous PLP concentrations 80  6.3 Results and discussion 82 

6.3.1 Assay of H2S synthesis from various in vivo sulfur-containing compounds 82 

6.3.2 Cloning of pET-22b(+)-CSE 84 

6.3.3 Bacterial expression and purification of His-tagged CSE 85 

6.3.4 Preparation of mutant CSE clones 86 

6.3.5 Bacterial expression and purification of mutant GST-tagged CSE proteins 87 

6.3.6 Analysis of protein secondary structure via CD measurements 89 

6.3.7 Comparison of the H2S synthesizing activities of the CSE mutant proteins 91 

Mutant CSE proteins affecting the binding of PLP cofactor 92 

Mutant CSE proteins affecting the activation of L-cysteine substrate 95 

Mutant CSE proteins affecting the affinity of the enzyme for L-cysteine 99 

6.3.8 Kinetics of H2S production in the presence of varying PLP concentrations 102  6.3.9 Proposed mechanism for catalysis of H2S production by human CSE 107 

6.4 Conclusion 112 

7 Development of a polyclonal antibody specific towards human CSE 114 

7.1 Objectives 114 

7.2 Experimental 115 

7.2.2 Immunoprecipitation (IP) of endogenous CSE using rabbit antibody serum 115  7.2.3 Purification of anti-hCSE 1366 116 

7.2.4 Probing for endogenous CSE levels in various cell lysates 116 

7.2.5 Immunoprecipitation of endogenous CSE using purified anti-hCSE antibody 117 

7.3 Results and discussion 118 

7.3.1 Testing of anti-hCSE sera 118 

7.3.2 Purification of antibody serum from rabbit 1366 119 

7.3.3 Characterization of purified anti-hCSE antibody 120 

7.4 Conclusion 123 

8 Concluding remarks 125 

Biblography 127 

Appendix 1: Forward and reverse primers used for PCR amplification of CSE 132 

Appendix 2: Mutagenic primers used for thermal cycling of mutant strands 133 

Appendix 3: Mechanism for H2S production as proposed in the Honors project 135 

SUMMARY

In recent years, increased interest has been directed towards hydrogen sulfide (H2S) as a third gasotransmitter and its role in various neurodegenerative and cardiovascular diseases Cystathionine-γ-lyase (CSE) is one of the two enzymes believed to be responsible for the endogenous production of H2S Research has also shown that inhibitors of H2S production are effective in reducing the severity of certain diseases caused by increased endogenous H2S levels However, these established inhibitors of CSE exhibit low potency, low selectivity and poor cell-membrane permeability As such,

we aimed to develop more specific and potent inhibitors of CSE towards H2S production

To achieve this, various inhibitor candidates were synthesized and tested using a previously established rat liver homogenate assay An expression and purification system for the human CSE enzyme was also developed to enable a more reliable method of screening of the inhibitor candidates via a purified protein assay, which was optimized for more efficient trapping of evolved H2S in this work The X-ray crystal structures of the enzyme in its apo- and holoenzyme forms, as well as in complex with one of its inhibitors have also been determined to aid in future rational design of inhibitors Initial attempts to co-crystallize the enzyme with some of our inhibitor candidates were also performed in this work

Although CSE has been well-known for its role in the transsulfuration pathway, the biochemical role of the enzyme in production of H2S is currently not well understood Hence, we were also interested in the mechanism for CSE-mediated H2S production This was achieved via site-directed mutagenesis and kinetic studies on the enzyme The in

vitro release of H2S from various sulfur-containing compounds present in our bodies was also tested using our purified protein assay Through these studies, not only were we able

to propose a more detailed mechanism for the catalysis of H2S production, we were also able to identify crucial residues which may directly affect the binding of inhibitors as well as certain key functional groups and their distribution within the inhibitor to allow for increased binding affinity to the enzyme Lastly, a polyclonal rabbit antibody that is specific towards human CSE was also developed to serve as a platform for future studies

of the function of the enzyme at the cellular level

LIST OF PUBLICATIONS

Sun, Q., Collins, R., Huang, S., Holmberg-Schiavone, L., Anand, G S., Tan, C H., den-Berg, S., Deng, L W., Moore, P K., Karlberg, T., and Sivaraman, J (2009) Structural Basis for the inhibition mechanism of human cystathionine-gamma-lyase:

van-An enzyme responsible for the production of H2S Journal of Biological Chemistry ,

284 (5), 3076-3085

Huang, S., Chua, J H., Sivaraman, J., Tan, C H., & Deng, L W (2009) Site-directed mutagenesis and kinetic studies on human cystathionine-gamma-lyase reveal interesting insights into the mechanism of H2S production Paper in preparation

LIST OF TABLES

Table 1 Percentage inhibition levels for various commercially available compounds assayed at 10 mM L-cysteine, 2 mM PLP and 5 mM test compound concentrations unless otherwise stated 8Table 2 Percentage inhibition levels for various chemically synthesized test compounds assayed at 10 mM L-cysteine substrate, 2 mM PLP and 5 mM test compound concentrations unless otherwise stated 8Table 3 Effect of various buffers on the polydispersity index of the protein solution, as measured by DLS at a protein concentration of 1mg/mL and a temperature of 20 °C 29Table 4 Effect of increasing sodium chloride concentrations on the polydispersity index

of the protein solution 29Table 5 A comparison of the kinetic parameters of human CSE when utilizing L-cysteine

Table 9 Correlation between logP values and production of H2S for the various mutated amino acids at 339th position of human CSE 101Table 10 Relative levels of endogenous CSE in various cell lines 122 

LIST OF FIGURES

Figure 1 Inhibition profiles and IC50 values of (A) PAG, (B) BCA, (C) N-Boc-L-cysteine

and (D) N-Cbz-D-cysteine determined in the presence of 10 mM L-cysteine substrate 12

Figure 2 (A) PCR amplification of human CSE for subsequent cloning into pGEX-4T-3, pcDNA3.1(+) and p3xFLAG-CMV-10 (B) DNA gel electrophoresis of restriction

enzyme cleaved recombinant CSE plasmids 22

Figure 3 Western blot analysis of the expression of FLAG-tagged human CSE in 293T cells transfected with recombinant pcDNA3.1(+)-FLAG-CSE and p3xFLAG-CMV-10-CSE plasmids 23

Figure 4 (A) 10 % SDS-PAGE gel showing expression of GST-CSE fusion protein (~66 kDa) under different induction conditions (B) 10 % SDS-PAGE of total (T), soluble (S)

and insoluble (I) fractions of cell lysates from bacteria induced for 3 h at 30 °C or 18 h at

18 °C 24

Figure 5 H2S synthesizing activities (expressed as nmol H2S produced per mg total protein) of rat liver homogenate, lysates of 293T cells transfected with pcDNA3.1(+)-FLAG-CSE or p3xFLAG-CMV-10-CSE and lysates of bacterial cells transformed with pGEX-4T-3-CSE induced under various conditions 25Figure 6 10 % SDS-PAGE analysis of affinity purification and thrombin cleavage of GST-CSE 27

Figure 7 (A) Anion exchange profile of the affinity pure CSE enzyme (B) Gel filtration

profile of the protein after ion exchange chromatography 31

Figure 8 10 % SDS-PAGE (A) and 6 % native-PAGE (B) gels of the peak gel filtration

fractions and the final purified CSE enzyme after gel filtration chromatography 32

Figure 9 Relationship between amount of H2S produced in 30 min against amount of purified recombinant CSE added in the presence of 10 mM L-cysteine substrate 37

Figure 10 (A) Graph of initial reaction velocity against L-cysteine substrate

concentration in the presence of 2 mM PLP (B) Logarithmic Hill plot of lg(V/(Vmax-V))

against lg[S] 38Figure 11 Average percentage inhibition values of various L-cysteine analogues, BCA and PAG assayed at 10 mM concentration in the presence of 10 mM L-cysteine substrate,

2 mM PLP and 10 µg purified CSE 40

Figure 12 Inhibition profiles and IC50 values of (A) N-acetyl-L-cysteine, (B)

N-isobutyryl-L-cysteine, (C) BCA and (D) PAG determined in the presence of 5 mM Lcysteine substrate, 2 mM PLP and 5 µg purified CSE 42Figure 13 Averaged percentage inhibition values of various synthesized inhibitor candidates assayed at 0.1 mM, 1 mM or 5 mM concentration in the presence of 2.75 mM

-L-cysteine substrate, 2 mM PLP and 5 µg purified CSE (candidates 1 to 6) or 7.5 µg purified GST-CSE (candidates 7 to 9) 44

Figure 14 Net A670 readings reflecting the distribution of trapped sulfides when different amounts of NaOH were added together with ZnAc 47Figure 15 Relationship between absorbance at 670 nm and amount of H2S produced 50Figure 16 Average percentage inhibition values of various synthesized inhibitor candidates assayed at 2.5 mM concentration in the presence of 2.75 mM L-cysteine substrate, 0.5 mM PLP and 7.5 µg GST-CSE 51Figure 17 DLS profile and parameters for the purified CSE protein at 5.0 mg/mL 60

Figure 18 Optimization of CSE crystallizing condition for X-ray diffraction and subsequent structure determination 61

Figure 19 Asymmetric units of human CSE determined in this work (A) and by our collaborator (B) 62

Figure 20 (A) Electron density map around PLP in the CSE holoenzyme Significant differences in stereo-overlay of peptide chains around the Tyr-114 (B) and Lys-212 (C)

residues, shown in green (our structure) and yellow (collaborator’s structure) 63Figure 21 Absorbance spectra of our purified CSE enzyme before and after L-cysteine incubation, and upon a readdition of equimolar amount of PLP 64Figure 22 A closed-up view of the active site region of the CSE-PAG complex 65Figure 23 Proposed mechanism for the inhibition of CSE by PAG 66

Figure 24 IC50 analysis on the inhibition of H2S production from the Y114F mutant CSE protein by PAG 67

Figure 25 (A) Spherulites of CSE complexed with 5 mM N-isobutyryl-D-cysteine

formed in 0.1 M BICINE pH 9, 20 % PEG 6000; (B, C) Needles formed around

spherulites 2 weeks later 68

Figure 26 (A) Crystal of CSE complexed with 5 mM N-isobutyryl-L-cysteine in 0.1 M BICINE pH 9, 20 % PEG 6000, 10 mM ZnCl2; (B, C) ZnCl2 crystals at bottom of well.69

Figure 27 10 % SDS-PAGE gel analysis on the proteolytic cleavage of CSE 70Figure 28 Optimization of the proteolysis of CSE by chymotrypsin 71

Figure 29 Alignment of the amino acid sequences of mouse, rat, human, Dictyostelium

(slime mold), yeast and Streptomyces CSE as well as E coli cystathionine-γ-synthase

(CGS) and cystathionine-β-lyase (CBL) 74Figure 30 Active site of the human CSE enzyme showing the location of crucial amino acids (only side chains shown) which would be studied by site-directed mutagenesis 75

Figure 31 A comparison of the net amount of H2S produced over 30 min by 5 mM of various sulfur-containing compounds 82

Figure 32 (A) PCR amplification of CSE for cloning into pET-22b(+) (B) Restriction

enzyme cleaved plasmids indicating the presence of CSE insert which was determined to

be correct upon sequencing 85Figure 33 10 % SDS-PAGE gel analysis showing attempted expression of His-tag CSE

in the presence of 0.1 mM IPTG at 20 °C for 20 h (A) and optimization of bacterial expression conditions (B) 86

Figure 34 0.8 % agarose gel showing PCR amplification of various mutant CSE plasmids 87Figure 35 10 % SDS-PAGE analysis of the induction of GST-tagged mutant and wild-type CSE proteins 88Figure 36 Proportion of α-helices, β-sheets, turns and unordered regions of GST-tagged mutant and wild-type CSE proteins 89Figure 37 Distances (in angstroms) between the polar contacts of the carboxylic acid side

pGEX-4T-3-chain of Glu-157 and amino group of Tyr-114 in the CSE holoenzyme (A) and apoenzyme (B) 91

Figure 38 A comparison of the H2S synthesizing abilities from 5 µg of various tagged CSE alanine mutants against wild-type GST-CSE 91

GST-Figure 39 H2S synthesizing activities of mutant CSE proteins with mutagenic alterations

to the Lys-212, Tyr-114, Asn-161 and Phe-190 residues, assayed in the presence of 5 µg GST-tagged protein, 0.5 mM PLP and 2.75 mM L-cysteine 94

Figure 40 H2S synthesizing activities of mutant CSE proteins with mutagenic alterations

to the Tyr-60, Arg-62, Ser-340 and Thr-189 residues, assayed in the presence of 5 µg GST-tagged protein, 0.5 mM PLP and 2.75 mM L-cysteine 96 

Figure 41 (left) The human CSE tetramer made up of a dimer of dimers (right)

Magnification of the interactions between PLP and Tyr-60 and Arg-62 from the adjacent monomer in subunits C and D of the enzyme 97

Figure 42 (left) Schematic representation of the hydrogen bonding network involving Thr-189, Asp-187 and the PLP cofactor (right) H2S synthesizing activities of mutant CSE proteins with mutagenic alterations to the Asp-187 residue, assayed in the presence

of 5 µg GST-tagged protein, 0.5 mM PLP and 2.75 mM L-cysteine 98

Figure 43 H2S synthesizing activities of mutant CSE proteins with mutagenic alterations

to the Glu-339 residue 101

Figure 44 Ionic interactions involving Arg-375 before (left) and after (right) the binding

of L-cysteine substrate 102

Figure 45 Amount of H2S produced in 30 min under different exogenous concentrations

of PLP, assayed in the presence of 7.9 µg GST-tagged CSE and 2.75 mM L-cysteine 103

Figure 46 (A) Graphs of initial reaction velocity, V against concentration of exogenous

PLP determined under various concentrations of L-cysteine substrate (B) Graphs of

initial reaction velocity, V against L-cysteine substrate concentration for the various concentrations of PLP that was added in the assay 105

Figure 47 (A) Double reciprocal plots for the various concentrations of L-cysteine

substrate that was added in the assay (B) Secondary plot for determination of the true

Vmax and KM values for L-cysteine 106

Figure 48 (A) Double reciprocal plots for the various concentrations of exogenous PLP that was added in the assay, up to 75 µM (B) Secondary plot for determination of the

true Vmax and KM values for PLP 107

Figure 49 Proposed mechanism for the catalysis of H2S production from L-cysteine by human CSE 109

Figure 50 Amount of H2S produced in 30 min under different exogenous concentrations

of PLP, assayed in the presence of 5 µg GST-tagged Y114F mutant CSE and 2.75 mM

L-cysteine 112

Figure 51 Probing of different amounts of pure CSE (A) and endogenous CSE from HepG2, 293T and 5 % w/v rat liver homogenate (B) with anti-hCSE sera from either rabbit 1365 or rabbit 1366 (C) Immunoprecipitation of endogenous CSE from 293T cell

lysates by utilizing antibody serum of either rabbit 1365 or 1366 119

Figure 52 Chromatograph of eluted anti-hCSE from HiTrap Protein A column and volume of 1 M Tris pH 9.0 base needed to neutralize various fractions from the blank run 120 

Figure 53 Western blot on purified CSE and endogenous CSE levels in various homogenates or lysates utilizing anti-hCSE antibody from different immunization batches 121Figure 54 A comparison of the relative endogenous CSE levels among various cell lysates, normalized against β-actin 122Figure 55 Immunoprecipitation of endogenous CSE from 293T, HepG2, K562 and U937 cells 123 

CSE Cystathionine gamma lyase

DLS Dynamic light scattering

SDS-PAGE Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

TBS Tris buffered saline

TCA Trichloroacetic acid

V Initial reaction velocity

1 Introduction

Cystathionine-γ-lyase (CSE, EC 4.4.1.1), an enzyme found in mammals and some fungi,

is involved in the reverse transsulfuration pathway (Scheme 1) where L-methionine is converted to L-cysteine through a series of metabolic interconversions (Rose & Wixom, 1955) Specifically, the role of CSE in this reaction pathway is to convert L-cystathionine

to L-cysteine whilst generating α-ketobutyrate and ammonia The reaction proceeds via

an α, γ-elimination mechanism where the C-γ-S bond of L-cystathionine is specifically cleaved to yield L-cysteine (Flavin, 1971) A defect in this metabolic pathway has been found to be associated with cystathioninuria as well as L-cysteine deficiency and subsequent impairment of glutathione metabolism (Uren, Ragin, & Chaykovsky, 1978, Vina, et al., 1995) In humans, CSE activity was detected in adult liver tissue but not that

in fetal liver (Sturman, Gaull, & Raiha, 1970) This was however attributed to translational regulation of CSE gene expression in the developing human liver (Levonen,

post-et al., 2000) Studies by Levonen post-et al (2000) had also identified two isoforms of CSE as splice variants of one another; the longer being enzymatically more active than the shorter Structurally, CSE is composed of four identical monomers of approximately 45kDa with a covalently bound pyridoxal 5’-phosphate (PLP) cofactor in each monomer The crystal structure of the enzyme was first depicted from yeast CSE by X-ray crystallography (Messerschmidt, et al., 2003) In Messerschmidt’s study, factors affecting the enzymatic specificity of the various transsulfuration enzymes had also been discussed Recently, studies on the single nucleotide polymorphic variant of human CSE (c.1364G>T, Ser403ÆIle) which had previously been found to be correlated with higher plasma homocysteine concentrations under homozygous conditions (Wang, Huff,

Spence, & Hegele, 2004) had shown that the PLP content and steady-state kinetic properties of the polymorphic enzyme was similar to that for the normal Ser403 variant (Zhu, Lin, & Banerjee, 2008) Experiments on the Thr67ÆIle and Gln240ÆGlu mutant CSE proteins which had previously been identified to lead to cystathioninuria (Wang & Hegele, 2003) had also revealed that the affinity of the mutant enzymes for PLP was weakened and that the enzyme activity could be restored by exogenous PLP in this study (Zhu, Lin, & Banerjee, 2008)

Scheme 1 Reverse transsulfuration pathway present in mammals and fungi

Besides the primary role of the enzyme in the conversion of L-cystathionine to L-cysteine, studies have also shown that rat liver CSE can utilize L-cysteine as a substrate for producing H2S gas (Stipanuk & Beck, 1982; Braunstein & Goryachenkova, 1984) This gas which had previously been primarily regarded as an environmental hazard and toxic gas, has recently been recognized as a third gasotransmitter besides carbon monoxide and nitric oxide (Wang, 2002) Other than CSE which accounts for endogenous production of

H2S in the liver, kidney, intestine and vascular smooth muscle cells, the in vivo production of H2S has also been attributed to cystathionine-β-synthase (CBS) and 3-mercaptopyruvate sulfurtransferase for brain and heart tissues respectively (Kamoun, 2004) Studies by Yang et al (2008) had specifically shown H2S as a physiologic

L-methionine

COO

NH3

S OOC

α β γ

COO

NH3

HS

Cystathionine γ-lyase (CSE)

COO O

L-cysteine α-ketobutyrate

+ NH3+

H2O

vasorelaxant and that prononced hypertension was triggered in CSE knockout mice models due to the absence of in vivo H2S production Characterization of a novel, slow-releasing H2S compound, GYY4137 under in vivo conditions had also shown that the vasorelaxant effect of H2S occurs via the opening of vascular smooth muscle KATP

channels (Li, et al., 2008) Imbalances in the endogenous H2S levels have therefore been associated with various diseases such as Alzheimer’s disease (Eto, et al., 2002), pulmonary hypertension (Li, et al., 2005), haemorrhagic shock (Mok, et al., 2004), carrageenan-induced hindpaw oedema (Bhatia, et al., 2005a), acute pancreatitis (Bhatia,

et al., 2005b) and endotoxemia (Collin, et al., 2005) In addition, H2S donors such as sodium hydrosulfide or GYY4137 as well as inhibitors of H2S production such as DL-

propargylglycine (PAG) and β-cyanoalanine (BCA) have been found to exhibit therapeutic potential in various disease models where the severity of the diseases were found to be alleviated upon administration of these compounds (Szabó, 2007; Li, et al., 2008) H2S donor compounds and inhibitors of H2S biosynthesis may hence provide insights into the mechanisms underlying various diseases, or function as therapeutic drugs Currently, the two commercially available inhibitors of H2S production, PAG and BCA possess low potency, low selectivity and limited cell-membrane permeability characteristics (Szabó, 2007) Therefore, there is a need to develop more effective inhibitors of H2S production

To achieve our aim, various L-cysteine and L-cystine analogues would first be tested using a rat liver homogenate assay However, due to many problems with this assay, we proceeded to develop a pure protein assay by cloning, expressing and purifying the

human CSE enzyme for subsequent screening of inhibitor candidates The purified protein would also be utilized for X-ray crystallography studies for elucidation of the three-dimensional structure of human CSE so as to aid in the rational design of inhibitors

of H2S production in future In addition, the expressed protein would also enable us to further explore the functional role of CSE in the catalysis of H2S production which is currently not well understood, as well as gain further insights into the mechanism for production of H2S These would be achieved via site-directed mutagenesis and kinetic studies Lastly, a polyclonal antibody which is specific towards the human CSE enzyme would be developed and characterized so as to serve as a platform for future functional studies on this protein

2 Tissue H2S assay for screening inhibitors of H2S production

2.1 Objectives

As mentioned in the introduction, there lies a need in developing more selective and potent inhibitors of H2S production since the current commercially available inhibitors, PAG and BCA are relatively weak and less selective In this section which had been accomplished during the UROPS project, various commercially available and chemically synthesized L-cysteine analogues would be tested for their inhibition levels towards H2S production from rat liver homogenates Drawbacks of the strategy used for inhibitor design as well as the tissue H2S assay screen would also be discussed

2.2 Experimental

A spectrophotometric assay modified from that described by Stpanuk and Beck (Stipanuk

& Beck, 1982) was used for assaying the production of H2S from rat liver homogenates All experiments on intact animals were undertaken with adherence to guidelines from the local National University of Singapore Institutional Animal Care and Use Committee (IACUC) Upon killing the rats, the livers were removed, cut into small pieces and kept frozen at -80 °C prior to the assay For each assay, a small portion of the rat liver was thawed and homogenized in appropriate amounts of ice-cold 100 mM KHPO4 pH 7.4 buffer Stock solutions of PLP and L-cysteine were prepared in 100 mM KHPO4 pH 7.4 buffer For each test, a duplicate and a baseline control were performed in 1.5 mL

cryovial tubes A negative control experiment without addition of any test compound was also performed Test compounds were either purchased from commercial sources or

chemically synthesized Trichloroacetic acid (TCA, 10 % w/v, 250 µL) was first added to

only the baseline control tubes to stop enzymatic reactions immediately by denaturing protein once the liver homogenate was added This was followed by the sequential addition of saline (10 μL for test compounds dissolved in 100 mM KHPO4 pH 7.4 buffer;

25 µL for test compounds dissolved in DMSO), PLP (50 mM, 20 μL), rat liver homogenate (5 % w/v, 430 μL), and the test compound (20 μL for compounds dissolved

in KHPO4 buffer; 5 µL for compounds dissolved in DMSO) to each tube For the negative control experiment, the same volume of solvent in which the test compound was dissolved was added instead of the test compound The tubes were then vortexed and preincubated on ice for 30 min, after which L-cysteine substrate (10 mM, 20 μL) was added The tubes were parafilmed tightly, gently vortexed and incubated in a 37 °C water bath for 30 min After incubation, the tubes were cooled on ice Zinc acetate (ZnAc, 1 % w/v, 250 μL) was added via needle to trap evolved H2S and all enzymatic reactions were stopped by addition of TCA (250 μL) via needle After centrifuging at 4 °C, 10000 rpm

for 2 min, N,N-dimethyl-p-phenylenediamine dihydro-chloride dye (NNDPD, 20 mM,

133 μL) in 7.2 M HCl and FeCl3 (30 mM, 133 μL) in 1.2 M HCl were added for development of methylene blue The tubes were centrifuged at 4 °C again, at 12000 rpm for 4 min 300 μL of the supernatant from each tube was loaded into a 96-well microplate, and the absorbance at 670 nm (A670) was measured The amount of H2S produced was calculated against a calibration curve of sodium hydrosulfide (NaHS: 0-100 μM) and the percentage inhibition of each test compound was then determined For determination of the IC50 values of potential inhibitors, the assay was performed in varying concentrations

of the inhibitor The IC50 value was then estimated from the graph of percentage inhibition versus inhibitor concentration

2.3 Results and discussion

From the commercially available compounds which were tested, N-isobutyryl-L-cysteine, N-isobutyryl-D-cysteine and N-acetyl-L-cysteine were some of the better inhibitors besides PAG and BCA which are the two established inhibitors of CSE (Table 1) Our strategy was hence to synthesize L-cysteine or D-cysteine analogues with modifications

to the amino group Compounds with substituents attached to the sulfhydryl group of Lcysteine were also synthesized and tested, but these were found to be poorer inhibitors compared to L-cysteine analogues with only their amino groups modified by the same or

-a simil-ar group (T-able 2) Modific-ation to both -amino groups of cystine -also led to -a decrease in inhibition potency These results suggest that a free sulfhydryl group may play a crucial role in the binding of the compound to the enzyme’s active site Although PAG and BCA do not possess this sulfhydryl functionality, they bind to the enzyme mechanistically through their amino group unlike these test compounds, which we believe would bind to the enzyme through other reversible or non-mechanistic means As for the N-substituted urea or thiourea L-cysteine derivatives, the urea derivatives generally fared better than their corresponding thiourea derivatives The inhibition levels were observed to increase when the electron-withdrawing property of the thiourea group was increased (from phenyl-thiourea to (3,5-difluoro)-thiourea to (3,5-bis-trifluoromethyl-phenyl)-thiourea), though such a trend could not be established for the corresponding urea groups Initially, it was also postulated that D-cysteine rather than L-

cysteine analogues, could be better inhibitors of H2S production since both

N-isobutyryl-D-cysteine and the Cbz-protected D-cysteine analogue exhibited higher inhibition levels than their corresponding enantiomer However, this was not observed for the Boc-

protected D- and L-cysteine analogues Due to the low availability and high cost of D

-cysteine as a starting material for synthesis of D-cysteine analogues, subsequent cysteine analogues were all synthesized from L-cysteine

Table 1 Percentage inhibition levels for various commercially available compounds

assayed at 10 mM L-cysteine, 2 mM PLP and 5 mM test compound concentrations unless

otherwise stated

OH

NH2O

DL - propargylglycine

NH2O

S-β-(4-pyridylethyl)- L -cysteine

HS

OH O

N-isobutyryl- L -cysteine

O

NH O N-isobutyryl- D -cysteine

O

NH O N-acetyl- L-cysteine

* Compounds assayed at 10 mM concentration

Table 2 Percentage inhibition levels for various chemically synthesized test compounds

assayed at 10 mM L-cysteine substrate, 2 mM PLP and 5 mM test compound

concentrations unless otherwise stated

O

NH O

N-Boc- D -Cysteine

NH O

O

N-pivaloyl- L-cysteine

S OH

O

NH O

S NH O HO O

N,N-dipivaloyl- L -cystine

O

NH O

N-Cbz- L -cysteine

O

NH O O

S NH

S NH O HO O

O Cl

N-(2-naphthoyl)- L-Cysteine

O NH S

NH O HO S

S O

O O

O

N,N-ditosyl- L -cystine

HS OH

NH O

OMe

thiourea)- L -cysteine

NH O

O NH

N,N-(diphenyl-urea)- L -cystine

NH O

S NH

S N H

F

F

F F

thiourea)- L -cysteine

NH O

S NH

thiourea)- L -cysteine

NH O

O NH

phenyl)-urea)- L -cysteine

NH O

S NH

F3C CF3

phenyl)-thiourea)- L -cysteine

HS OH

NH O

S N H

thiourea)- L -cysteine O

N,S-Bis((4-acetyl-phenyl)-O

^ Compounds assayed at 2.5 mM L- cysteine substrate and test compound concentrations

† Compounds assayed at 10 mM and 2.5 mM L- cysteine substrate and test compound concentrations respectively

Two of the more potent analogues that were synthesized, N-Boc-L-cysteine and

N-Cbz-D-cysteine, besides the commercially available inhibitors PAG and BCA, were selected for determination of their inhibition profiles (Fig 1) Although there were compounds which were more potent than these two analogues at 5 mM concentration (for example N-((3,5-Bis-trifluoromethyl-phenyl)-thiourea)-L-cysteine and N-(phenyl-urea)-L-cysteine), their inhibition profiles could not be obtained since their solubilities became rather poor when more than 5 mM of these samples were assayed From the IC50 values, N-Boc-L-

cysteine is still a relatively weak inhibitor compared to PAG (IC50 = 0.2 mM) and BCA (IC50 = 0.1 mM) The IC50 value for N-Cbz-D-cysteine could not be determined since precipitation became eminent for inhibitor concentrations beyond 8 mM

Figure 1 Inhibition profiles and IC50 values of (A) PAG, (B) BCA, (C) N-Boc-L-cysteine

and (D) N-Cbz-D-cysteine determined in the presence of 10 mM L-cysteine substrate

A major problem encountered during the assay was attributed to the poor solubility of most inhibitor candidates Although this could be alleviated by lowering the concentrations of the test compounds and substrate, the sensitivity of the assay would become compromised since only very little amounts of H2S was produced for the negative control tubes to which the inhibition levels of the compounds would be subsequently computed As such, an optimized substrate and test compound concentration of 10 mM and 5 mM respectively was utilized for most experiments Although the dissolution of the poorly soluble compounds in DMSO could also alleviate

IC 50 : Cannot be determined; Rsq: 0.983

Concentration of Cbz-D-cysteine (mM) D

the problem, some precipitation was still evident upon addition to the rat liver homogenate, particularly for those hydrophobic L-cysteine analogues As there may be a possibility of inaccurate results due to any undesirable effects of DMSO, the volume of DMSO was kept to a minimum of 1 % v/v in the assay Nevertheless, a negative control experiment where DMSO was added to this final concentration was included in each assay to ensure that the production of H2S had not been compromised

2.4 Conclusion

The random synthesis of L-cysteine analogues did not serve as a strategic way for the design of inhibitors of H2S production Moreover, the tissue H2S assay that was used for screening had required large amounts of substrate and test compounds to yield large enough responses for accurate results This was undesirable not only due to high costs but also false positives which may result due to non-specific inhibition caused by the aggregation of excessive amounts of the test compound to the target protein (McGovern

et al., 2002) There was also a possibility of degradation of the substrate or test samples

by other enzymes present in the rat liver homogenate, thus affecting the inhibition levels

A more efficient approach would be to base the design of the inhibitor upon the dimensional structure of the protein, as well as to utilize pure protein instead of rat liver tissue homogenates in the assay To achieve this, the human CSE gene would be cloned into various vectors for subsequent production of the purified enzyme

three-3 Cloning and expression of recombinant human CSE

3.1 Objectives

Due to the intrinsic drawbacks of the rat liver homogenate assay that was utilized in the preliminary screen for inhibitors of H2S production, an expression and purification system for the human CSE protein had been developed during the UROPS and Honors projects Besides developing a better assay for the production of H2S and screening of inhibitor candidates, the success in establishing a purification system for human CSE also forms an important basis for the subsequent elucidation of the three-dimensional structure

of the enzyme, kinetics and site-directed mutagenesis studies for expounding upon the mechanism for the catalysis of H2S production, as well as in developing a polyclonal antibody which is specific towards this protein The human CSE gene was hence cloned into 3 different vectors as we would like to determine which would subsequently allow for the most efficient expression of protein Although the bacterial expression vector pGEX-4T-3 was likely to achieve this aim, the expressed enzyme may not be in the active form since post-transcriptional and post-translational mechanisms are absent in bacterial cells Mammalian expression vectors were hence included in our choice of vectors to which the human CSE gene would be cloned into The various vectors that were used would also aid in purification of our protein subsequently pGEX-4T-3, for instance, allows a GST-tag to be incorporated to the N-terminus of our protein This tag not only aids in easy purification of the protein via affinity chromatography, but is also believed to improve the solubility of the fusion protein (Donald et al., 1988) The presence of three FLAG epitopes in the mammalian expression vector, p3xFLAG-CMV-

10 is also likely to increase the efficiency of purification due to strong interactions with the anti-FLAG antibody that could be used in the purification process As the other mammalian expression vector pcDNA3.1(+) does not possess any tags for easy purification of the protein, a FLAG epitope would be designed in the forward primer used for PCR amplification of FLAG-CSE so that the expressed protein would subsequently possess the FLAG tag for easy purification Upon expression of CSE from the various recombinant plasmids, the H2S synthesizing activity of the crude mammalian and bacterial cell lysates would be determined and an expression system which allows for an economical production of large amounts of the protein would be selected for further expression and purification of the enzyme

3.2 Experimental

3.2.1 Preparation of recombinant human CSE plasmids

Polymerase chain reaction (PCR) amplification on human full length CSE cDNA (GenBank accession no BC015807) obtained from Open Biosystems (cat no MHS1010-73982) was performed using 30 PCR cycles (30 s at 94.0 °C, 30 s at 57.0 °C and 2 min at 72 °C) The primers used for the amplification process are listed in Appendix 1 25 μL of each of the respective purified PCR products were then cloned into pGEX-4T-3 and pcDNA3.1(+) with EcoRI/XhoI sites, and into p3XFLAG-CMV-10 with EcoRI/KpnI sites The constructs were fully sequenced (1st Base Pte Ltd) and found to contain the desired CSE inserts

3.2.2 Mammalian expression of FLAG-tagged CSE

293T cells were grown at 37 °C in Dulbecco's modified Eagle's medium (DMEM, Sigma) supplemented with 10 % fetal bovine serum (Hyclone), 2 mM L-glutamine (Gibco), 100 units/ml penicillin (Gibco) and 100 μg/ml streptomycin (Gibco) in a humidified atmosphere containing 5 % CO2 For transfection of plasmid DNA into 293T, cells were first seeded onto two 6cm plates to reach a confluency of about 50 % the next day The transfection mixture was prepared by adding 6 µL of FuGENE 6 transfection reagent and

2 µg of the plasmid DNA (pcDNA3.1(+)-FLAG-CSE, p3xFLAG-CMV-10-CSE and their corresponding mock vectors) to 94 µL of serum free DMEM medium, followed by incubation at room temperature for 15 min The transfection mixture was then added dropwise to the plated cells, and cells were harvested two days later by trypsinization For preparation of cell lysates, cells were first washed with phosphate buffered saline (PBS), pelleted by centrifugation, and subsequently lysed in appropriate volumes of mild lysis buffer (150 mM NaCl, 20 mM Tris-HCl pH 8.0, 1 % TritonX-100, 10 % w/v glycerol, 1

mM DTT) supplemented with protease and phosphatase inhibitors (2 mM PMSF, 4 µg/mL Leupeptin, 4 µg/mL Aprotinin, 2 µg/mL Pepstatin A, 2 mM Na3VO4, 10 mM NaF)

by passing the mixture through 21 G needle on ice After incubation on ice for an hour, the lysate was centrifuged at 13000 rpm for 15 min at 4 °C, and the supernatant aliquoted and kept at -80 °C until further use

3.2.3 Western blotting

Protein samples were first separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently transferred onto PVDF membranes using

transfer buffer without SDS at 60 mA for 2 h Blocking was performed by incubating the membrane in 3 % w/v non-fat skim milk prepared in Tris buffered saline (TBS) supplemented with 0.05 % Tween 20 (TBS/0.05 % Tween) at 4 °C overnight The membrane was then incubated at room temperature (25 °C) for 1 h with the specific primary antibody prepared in 1.5 % non-fat skim milk prepared in 1x TBS/0.05 % Tween Following that, the membrane was washed thrice with TBS/0.1 % Tween at room temperature for 10 min each on a bench top shaker Incubation with the respective secondary antibody (prepared in 1.5 % non-fat skim milk in TBS/0.05 % Tween) was performed at room temperature for 1h, after which the membrane was washed thrice with TBS/0.1 % Tween at room temperature for 10 min each on a bench top shaker Chemiluminescence analysis was then performed by incubating the membrane with enhanced chemiluminescence (ECL) substrates (PerkinElmer or Amersham) for 1 min at room temperature The image was then developed with X-ray film (Amersham) for exposure

3.2.4 Optimization of bacterial expression of human CSE

25 ng of the recombinant pGEX-4T-3-CSE plasmid was transformed into competent

Escherichia coli (E coli) BL21 cells 4 colonies of the pGEX-4T-3-CSE transformed

BL21 cells were inoculated into 20 mL of steam-autoclaved Luria-Bertani (LB) broth supplemented with 100 μg/mL of ampicillin (LB-Amp100) and incubated at 37 °C overnight with vigorous shaking 7 mL of the starter culture and 7 mL of 20 % w/v glucose solution was propagated in 126 mL of LB-Amp100 for cultures to be incubated

at 30 °C, while 5 mL of the starter culture and 5 mL of 20 % w/v glucose solution was

propagated in 90 mL of LB-Amp100 for cultures to be incubated at 18 °C Batches of bacteria were then induced with 0.1 mM isopropyl-β-D-thiogalactopyranoside (IPTG) at

OD600 0.5 or 1.0 Protein expression was performed at 30 °C for 3 h or 6 h, or at 18 °C for

4 h or 18 h The bacterial cells were then pelleted by centrifuging 6000 x g at 4 °C for 15 min, and subsequently lysed in lysis buffer (20 mM Tris pH 8.0, 150 mM NaCl, 1 mM EDTA, 2.5 mg/mL lysozyme and 1.5 % w/v sodium sarcosyl) supplemented with protease inhibitors (1 mM PMSF, 2 µg/mL Leupeptin, 2 µg/mL Aprotinin, 2 µg/mL Pepstatin A) for determination of protein solubility and enzyme activity

3.2.5 Miniaturized assay of H2S production

A miniaturized assay from that described in Section 2.2 was used to determine the amount of H2S produced from mammalian and bacterial cell lysates 100 µL of 10 % w/v TCA was added initially to the baseline control tubes The reaction mixture consisted of a smaller total volume of 300 µL and comprised saline (60 μL), PLP (30 mM, 20 μL), rat liver homogenate (2.5 % w/v) or either mammalian or bacterial cell lysates (200 μL) and

L-cysteine (150 mM, 20 μL) The tubes were parafilmed tightly, gently vortexed, and then incubated in a 37 °C water bath for 30 min After incubation, the tubes were cooled

on ice After addition of ZnAc (1 % w/v, 100 µL) and TCA (10 % w/v, 100 µL), the tubes were centrifuged at 10000 rpm for 2 min at 4 °C NNDPD (20 mM, 71.4 μL) and FeCl3 (30 mM, 71.4 μL) were then added for development of methylene blue The tubes were then treated as described above and the A670 readings of the supernatants were measured

3.2.6 Bacterial expression of human CSE enzyme

Four colonies of the pGEX-4T-3-CSE transformed BL21 cells were inoculated into 150

mL of steam-autoclaved Luria-Bertani (LB) broth supplemented with 100 μg/mL of ampicillin (LB-Amp100), and incubated at 37 °C overnight with vigorous shaking 50

mL of the starter culture and 50 mL of 20% w/v glucose solution were propagated separately into each of two 2.5 L Erlenmeyer flasks, each containing 900 mL of LB-Amp100 The mixture was incubated at 37 °C with vigorous shaking until an optical density at 600 nm (OD600) of 0.3 was attained, upon which bacterial growth was continued at 18 °C until an OD600 of 0.5 was attained IPTG was then added at a rate of 0.1 mM to induce the expression of human CSE at 18 °C for 18 h The cells were harvested by centrifugation in a Beckman JLA8.1000 rotor at 6000 rpm for 15 min at

4 °C The pellets were combined and resuspended in 200 mL of the supernatant, divided into five portions, and centrifuged again at 8000 rpm for 15 min at 4 °C using an Eppendorf centrifuge (5804R) equipped with a F-34-6-38 rotor The supernatants were removed, and the pellets (each containing 400 mL worth of the original cell culture) were kept at -80 °C until further use

3.2.7 Purification of human CSE enzyme

The bacterial pellet was thawed and resuspended in 40 mL of lysis buffer containing 50

mM Tris pH 8.0, 100 mM NaCl, 1 % Triton-X, 5 mM dithiothreitol (DTT), 1 mM PLP and 0.22 mg/mL of protease inhibitor mixture (Cat no.: P8465, Sigma) Cell lysis was performed by sonication on ice using the 30 % pulsed maximum output of a Sonics Vibracell sonifier equipped with a macrotip The lysate was then cleared by centrifuging

at 18000 rpm for 30 min at 4 °C using a Beckman JA25.50 rotor The supernatant was introduced to a 100 mL chromatography column containing 5 mL of glutathione sepharose beads previously equilibrated with 40 mL of the lysis buffer The column was sealed and incubated with slight shaking at room temperature (20 °C) for 30 min Following that, the flow-through was collected, and the beads were washed once with 40

mL of 20 mM Tris/HCl pH 8.0, 100 mM NaCl, 1 % Triton-X, 1 mM DTT and then twice with 40 mL of 20 mM Tris/HCl pH 8.0, 100 mM NaCl, 1 mM DTT On-column removal

of the gluthathione S-transferase (GST) tag was then performed by adding 10 mL of cleavage buffer (50 mM sodium phosphate buffer pH 8.2, 100 mM NaCl, 1 mM DTT) containing between 5 to 20 units of thrombin protease (Sigma), incubating at room temperature for 15 min, and eluting the cleaved protein solution This was repeated until minimal protein concentration was detected in the eluates The eluates were then pooled together and concentrated using an ultra-centrifugal filter (Millipore Amicon Ultra-4

30000 MWCO) at 4 °C The concentrated eluate was subsequently passed through a HiTrap Q HP anion exchange column (GE Healthcare) connected to a BioRad BioLogic Duo Flow FPLC equilibrated with buffer A (10 mM sodium phosphate buffer pH 8.2 and

1 mM DTT) at a flow rate of 1 mL/min CSE was subsequently eluted at about 0.15 M NaCl when a linear 50 ml-gradient from 0 to 1 M NaCl at 1 mL/min was applied Fractions containing CSE were identified by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), and then pooled together The pool was then passed through a Superdex-200 column connected to the BioRad BioLogic Duo Flow FPLC equilibrated with buffer A at a flow rate of 1 mL/min Peak fractions corresponding to the target CSE protein were pooled together and concentrated to about 5 mg/mL in 10 mM

sodium phosphate buffer pH 8.2 and 1 mM DTT using an ultra-centrifugal filter (Millipore Amicon Ultra-4 10000 MWCO) at 4 °C The protein was kept at 4 °C for short-term storage of up to two weeks, and at -80 °C for long-term storage

3.3 Results and discussion

3.3.1 Construction of recombinant human CSE plasmids

PCR amplification of the human CSE gene was rather efficient and gave high yield of the amplified PCR products DNA gel electrophoresis (Fig 2A) also showed that the PCR products were of the right size (~1.2 kb) Subsequent restriction enzymes digestion and ligation of the CSE inserts to their corresponding vectors were performed Successful clones were identified by DNA gel electrophoresis which displayed a ~1.2 kb DNA band after digestion with its corresponding restriction enzymes (Fig 2B) The accuracy of the CSE insert were fully sequenced and confirmed by DNA auto-sequencing

1 2 3 4 1 2 3 4 1 2 3 4

3.0

1.0 1.52.02.5

(kb)

4.0 5.06.08.0 10.0

pcDNA3.1(+)- CMV-10-CSE

3.3.2 Expression and determination of H2S synthesizing activity of recombinant human CSE

Transfection of recombinant pcDNA3.1(+)-FLAG-CSE and p3xFLAG-CMV-10-CSE into 293T cells was somewhat successful as can be observed from the presence of a band

at approximately 43 kDa when compared to the mock transfection sample (Fig 3) For pcDNA3.1(+)-FLAG-CSE, lesser amounts of the protein was expressed compared to that for p3xFLAG-CMV-10-CSE as the former transfection experiment had been performed

as a trial practice We hence believe that the transfection efficiency and protein yield would be improved in subsequent experiments Alternatively, some optimization of the transfection protocol may be performed to obtain larger amounts of the desired protein

Figure 3 Western blot analysis of the expression of FLAG-tagged human CSE in 293T cells transfected with recombinant pcDNA3.1(+)-FLAG-CSE and p3xFLAG-CMV-10-CSE plasmids Mock transfection of empty pcDNA3.1(+) and p3xFLAG-CMV-10 vectors are shown in Lanes 1 and 3 respectively Western blot was performed by blocking the membranes at 4 °C overnight followed by incubation with anti-FLAG antibody (1:1000 dilution, Sigma) Horse raddish peroxidase (HRP)-conjugated anti-mouse antibody (1:10000 dilution, Amersham) was used as secondary antibody Chemiluminescence was detected using enhanced luminol substrate from PerkinElmer

FLAG-CSE p3xFLAG-CMV-10-CSE

pcDNA3.1(+)-– + pcDNA3.1(+)-– +

FLAG-tagged CSE

Transfection

The bacterial expression of CSE was also successful under various conditions as shown

in Fig 4A Although a similar expression of CSE had been reported by Steegborn et al

(Steegborn, 1999), only about 10 % of the expressed protein was soluble in their study Slowing the protein expression rate by utilizing lower concentrations of IPTG and lower expression temperatures did not improve the solubility of their protein significantly In our study, we optimized the bacterial expression of CSE by varying the optical densities and temperatures to which induction was performed Optimization of the bacterial induction conditions not only allowed us to explore which would give the highest yield of soluble protein, but also enabled us to determine which condition would allow for the expression of the most active enzyme The results showed that all of the induction conditions experimented led to expression of large amounts of the fusion protein There was also no significant difference in the solubility of the expressed protein under different induction conditions (Fig 4B)

Figure 4 (A) 10 % SDS-PAGE gel showing expression of GST-CSE fusion protein (~66

kDa) under different induction conditions in the absence (-) and presence (+) of IPTG

over 6 h at 30 °C and 18 h at 18 °C (B) 10 % SDS-PAGE of total (T), soluble (S) and

insoluble (I) fractions of cell lysates from bacteria induced for 3 h at 30 °C or 18 h at

18 °C Broad molecular range protein ladder (Biorad) was loaded in Lane 1 for both gels (Adapted from Huang et al., 2007)

GST- CSE

T S I T S I T S I Induction at OD600= 0.5 OD600 = 1.0 OD600= 0.5 Temperature 30°C 18°C

200

116 97.4 66.2 45.0 31.0