INACTIVATION OF BENZOYLFORMATE DECARBOXYLASE

BY THMMIN THIAZOLONE DIPHOSPHATE

Daria Hi1 Ching Yu

A thesis submitted in conformity with the requirements For the degree of Master of Science Graduate Department of Chemistry

University of Toronto

Copyright by Daria Hil Ching Yu 2001

National Library

Bbliotheque nationale du Canada

Acquisitions and Bibliographic Services

Acquisitions et services bibliographiques

395 Wellington Street Ottawa ON K I A ON4

Ottawa ON K I A ON4

Canada

Canada

395. rue Wellington

The author has granted a nonexclusive licence dowing the National Library of Canada to reproduce, loan, distri'bute or sell copies of this thesis in microfonn, paper or electronic formats.

The author retains awnership of the copyright in this thesis. Neither the thesis nor substantial extracts fkom it may be printed or otherwise reproduced without the author's permission.

L'auteur a accordé une licence non exclusive permettant à la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/film, de reproduction sur papier ou sur format électronique. L'auteur conserve la propriété du droit d'auteur qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.

To Daddy, M o r n y and yenneroos

Abstract

INACTIVATION OF BENZOYLFORMATE DECARBOXYLASE BY TEUAMIN TElIAZOLONE DIPHOSPHATE Daria Hi1 Ching Yu Department of Chemistry, University of Toronto Master of Science, 200 1

In 1973, Gutowski and Lienhard designed and synthesized thiarnin thiazolone diphosphate ('ITDP) as a transition state analogue for E. coli pyruvate dehydrogenase, an enzyme whieh uses thiamin diphosphate as a cofactor in the conversion of pyruvate to

2-(1-hydroxyethy1)thiarnin diphosphate [Gutowski, J.A., and Lienhard, G.E. (1976)

J. Biol. Chem., 251,2863-28661.Other groups have reported different results for other ThDP-dependent enzymes. Here, we begin the investigation of the interaction between

TTDP and P. putida benzoylformate decarboxylase (BFD). First, a recombinant form of BFD was expressed and purified by nickel af£inity chromatography. Existing direct and coupled assays were modified to follow the activity of the enzyme with endogenously bound thiamin diphosphate and inhibition of enzyme activity with TTDP. Preliminary results suggest Ki = 5 ph4 for TTDP which is on the same order as the Kmof ThDP for the enzyme. Thus, TTDP is not a transition state analogue for BFD.

Acknosvledgements

1 would Iike to thank Dr. Ron Kluger for giving me an opportunity to do research in his lab. 1leamed how to read and think cntically about the physical chemistry of enzyme mechanisms. Without his constant emphasis on solving small problems, this thesis would still be on the cornputer. The lab group has also been supportive of m y idiosyncracies and crazy working hours. Past or present, they are Anna, Amer, Ian, Jesse, John, Lisa, Nik, N o m , Pete, and Steve. 1 would also like to thank Lena and Miriam of Purdue University, Indiana, Petra

and Martina of der Institut fur Enzyrntechnologie of Heninch-Heine Universitat Dusseldorf im Forschungszentrum Jülich, Germany for their invaluable correspondence and advice. Other people in the department have also made working here interesting. Also my fiiends totally unrelated to chernistry have made Iiving in Toronto enjoyable. Finally, 1would Iike to thank my family for their love and encouragement.

Table of Contents

Abstract ...............................................................................................................................

..

ri

...

............................................................................................

-111

Table of Contents ...............................................................................................................

iv

Acknowledgements..........,

. . . ..................................................................................... vi List of Tables ..................... . List of Figures ....................................................................................................................

vi

Structures and Abbreviations .................... . . ...................................................................

. vii

J

..................................................................................................

-1

1.1 Benzoylformate Decarboxylase .......................................................................

1

1-2 Catalytic Mechanism .......................................................................................

1

Chapter 1: Introduction

..

1-3 Transition State Theory................................................................................

................................................................................................

Chapter 2: Experimental

4 6

2.1 Instrumentation .................................................................................................6 2.2 Materials ........................................................................................................... 6 2.3 Syntheses.......................................................................................................... -6 2.3.1 Thiarnin Disulfide .............................................................................

-7

2.3 -2 Thiarnui Thiazolone .......................................................................... -7 2.3 -3 Thiamui Thiazolone Diphosphate ...................................................... 8

2.4 Expression and Purification of Benzoylformate Decarboxylase-His6..............9

2.5 Enzyme Assays ........................... . . .............................................................1 2.5.1 Direct Decarboxylase Assay ............................................................ 11 2.5.2 Coupled Decarboxylase Assay .....................

............................12

2.6 Inhibition with Thiamin Thiazolone Diphosphate At 5 min Incubation Period ............. .........................*...................................... 13 2.7 Time Dependence of Thiamui Thiazolone Diphosphate on BFD-His6 Activity ...................................................................................1 4

Chapter 3: Results

........................................................................................................... 15

3.1 Synthesis of Thiarnin Thiazolone Diphosphate ............................................. 1 5 3.2 Purification of Benzoylformate Decarboxylase-His6 .................. ..... .............. 15 3.3 Direct Decarboxylase Assay ..............*............................................................ 16 3.4 Coupled Decarboxylase Assay ................... .....

.......................................... 18

3.5 Inhibition with Thiamin Thiazolone Diphosphate .......................................... 20

Chapter 4: Discussion.....................................................................................................22 4.1 BFD-His6 Purification..................................................................................... 22 4.2 Cornparison of Direct and Coupled Decarboxylase Assays ...........................22

4.3 Inhibition with Thiarnin Thiazolone Diphosphate ......................................... 24 Chapter 5: Conclusions and Future Direction

............................*................................31

References ................... .................................................................

........=..........................32

List of Tables Components of direct decarboxylase assay Components of coupled decarboxylase assay Effect of TTDP on ThDP-BFD-His6 Incubation mixtures of BFD-His6 and TTDP Reaction mixture for tirne dependence of TTDP on BFD-His6 activity List of Figures Catalytic mechanism for benzoylformate decarboxylase Calibration cuve for Bradford assay Direct assay to follow activity of BFD-His6 Rate of reaction is proportional to amount of BFD-His6 Absorbance of NADH at 340 nrn Following oxidation of NADH in the coupled decarboxylase assay Checkhg for coupling of decarboxylation to NADH oxidation Following reduction in activity of 0.02 U BFD-His6 by TTDP BFD-His6 activity as a function of 'TTDP concentration Time dependence of TTDP on BFD activity Reactions in the direct and coupled decarboxylase assays A potential transition state analog for BFD

Possible compounds that could lead to inhibiton of BFD Free energy-reaction progress profiles for the non-enzyrnic decarboxylation of pymvate by ThDP and the reactions catalyzed by yeast pyruvate decarboxylase (SCPDC) and Zymmonas pyruvate decarboxylase (ZMPDC)

vii

Structures and Abbreviations

MoIecule

Abbreviation

Thiamin, ThDP

R = P2O4H2,Thiarnin diphosphate

BFD-His6

Benzoylformate decarboxylase- His6

(9-2-HPP

(5)-2-hydroxy- 1-phenyl-propanone

MThDP

2-(mande1yl)thiamin diphosphate, R =~

HBzThDP

~ 0 6 ' ~

2-(1-hydroxybenzyl)thiamin diphosphate, R = pz02

TT

R = H, Thiarnin thiazolone,

TTMP TTDP

R = POsH, TT monophosphate, R = PZO6H3,TT diphosphate,

TTTP

R = P3O9&, TT triphosphate

LB

Luria-Bertani

HLADH

Horse liver alcohol dehydrogenase

NADH

Dihydronicotinamide adenine dinucleotide

Structure

Structures and Abb reviations (continued)

Abbreviation

Molecule

Ni-NTA

Nickel-nitrilotriacetic acid

BSA

Bovine serum albumin

SDS-PAGE

Sodium dodecyl suljphate polyacrylamide gel electrophoresis

IPTG

Isopropyl-P-thiogalactopyranoside

PMSF

Phenylrnethanesulfonyl fluonde

LThDP

2-(a-1actyl)thiamin d i phosphate,

R = pzosJ

HEThDP

2-(1-hydroxyethyl)thiamin

diphosphate, R = P ~ Q ~ - ~

methyl acetylphosphonate

HBP

hydroxybenzylphosphonate

Structure

Chapter 1 - Introduction

1 .

Benzoylformate Decarboxylase Benzoylfonnate decarboxylase (Bm),an enzyme in the mandelate pathway of

Pseudomonas and Actinobacter species

', uses thiarnin diphosphate (ThDP) to catalyze the

Studies involving isotope effects 2, substrate and

decarboxylation of benzoylformate.

substrate anaiogs are consistent with Breslow's covalent catalytic mechanism for ThDPdependent decarboxylases 4. The crystal structure of BFD was solved to 1.6 A resolution and spectroscopie studies helped clarie the nature of the intermediates along the catalytic cycle. BFD also catalyzes a side reaction, carboligation, and has been used in the asyrnmetnc synthesis of chiral 2-hydroxy ketones '-Io.

For example, the "activated

benzaldehyde", 3, on the enzyme can add to acetaldehyde to yield (5')-2-hydroxy-1-phenylpropanone, ((S)-2-HPP, Scheme 1). The gene for the enzyme has been isolated, cloned

'

and rnodified to include a C-terminal hexahistidhe extension mis6) that permits easy purification of the enzyme using nickel affinity chromatography 'O. The activity of BFD-

His6 is the same as the native enzyme.

1.2

CataIytic Mechanism of Benzoylformate Decarboxylase The catalytic mechanism of ThDP-dependent decarboxylases was proposed by

Breslow and McNellis and is now widely accepted I l . A detailed review of the reaction intermediates for the decarboxylation of pymvate to acetaldehyde by pynivate decarboxylase has been published benzoylfonnate as the substrate.

12.13

.

BFD catalyzes an analogous reaction using

The enzyme environment favours the deprotonation of C-2 on the thiazolium ring. The resulting ylide, 1,attacks C-2 of benzoylformate, giving a covalent intermediate, 2-(rnande1yl)thiami.n diphosphate, 2. Loss of carbon dioxide and subsequent proton transfer gives 2-(1-hydroxybenzyl)thiamin diphosphate, 4. The proposed transition state for this step involves a species that has both carbanionic and enamine character, 3. The final step is the regeneration of the ylide and release of benzaldehyde. The decarboxylation cycle and carboligation side reaction for BFD are depicted in Scheme 1. It is also interesting to note that the non-enzyrnatic reaction between

In acidic and neutral solutions, 4

benzoylformate and thiarnin is pH dependent 14.

undergoes fragmentation to give 2,s-dimethyl-4-amino-pyrimidineand

'

2-benzoyl-5-(2-hydroxyethyl)-4-rnethylthile

In basic solution, the products are

thiarnin and benzaldehyde. BFD,having optimal activity at pH 6.0, circurnvents the

hgmentation reaction to give the desired elimination products.

SA* oz/

z

1.3

Transition State Theory In 1973, Lienhard presented the concept that transition state analogues could be

potent enzyme inhibitors

16.

The ideal transition state analogue could bind to the enzyme

with the same affinity as would the transition state. Gutowski and Lienhard applied this

theory to E. coli pynivate dehydrogenase, which initially catalyzes the conversion of pymvate to 1-hydroxyethylthiamin diphosphate and carbon dioxide. They designed thiarnin thiazolone diphosphate, 5, so that its carbonyl fünctionality at C-2 of the thiazokun ring would mimic the enamine character of 3. They found that TTDP binds at least 20 000 times more tightly than does ThDP 17, consistent with their expectations for a transition

state analogue. NH2

I

thiamin thiazolone diphosphate or TTDP, 5

However, the results have been subject to other interpretations since the affinity can be explained on other factors, including hydrophobicity of the active site 18*'9. TTDP has been tested with other TDP-dependent enzymes, where high affinity has not been observed. Gish (1984) showed TTDP to be an irreversible inhibitor for wheat germ pymvate Neither is TTDP an decarboxylase, binding only three times more tightly than ThDP lgv2*. effective transition state analogue for acetohydroxyacid synthase 21, nor is it for transketolase 22. TTDP is a weak competitive inhibitor for these two enzymes. TTDP is a

non-competitive inhibitor of rat brain pyruvate dehydrogenase 23. These results indicate that Lienhard's theory requires a more cornplex set of c i t e n a than TTDP c m

accommodate. Since we are interested in the mechanism of Pseudomona puriaa benzoylfcrmate decarboxylase with respect to ThDP, its interactions with TTDP need to be established.

Chapter 2 - Experimental 2.1

Instrumentation Proton NMR spectra were obtained at 400 MHz. Phosphorus NMR spectra were

obtained at 121.4 MHz. Enzyme assays and inhibition kinetics with TTDP were foI1owed by UV on a Perkin Elmer Lambda 2 or Lambda 19 spectrophotometer.

2.2

Materials The plasmid containing the BFD-His6 insert was a gift fiom the Iab of Dr. Martina

Pohl of der Institut für Enzymtechnologie of Heinrich-Heine Universit3t Dusseldorf irn Forschungszentnim Jülich, Germany. Thiamin hydrochloride was a gift fkom Novopharm. Competent E.coli BL2 1 cells were obtained fiom Novagen. The Ni-NTA agarose product was fiom Qiagen. HLADH was a crystallïne suspension, 33 U/ml, obtained fkom FIuka. Al1 other reagents and solvents were obtained in the highest purtty possible.

2.3

Syntheses Thiamin was converted to thiamin disulfide, 6, and then recyclized to forrn thiarnin

thiazolone, 7. Phosphorylation of 7 gave a mixture of the mono-, di-, and triphosphorylated species of thiamin thiazolone which were then purified by chromatography and characterized by NMR.

Thiamin thiazolone, 7, R = OH Thiarnin thiazolone diphosphate, 5, R =

Thiamin disulfide, 6

2.3.1

Thiamin DisuEde Thiamin disulfide was synthesized by the method described in Gish's thesis 18.

Thiamin hydrochlonde (20 g) was dissolved in 100 ml distilled water, which was adjusted to pH 11.75 with 6 N sodium hydroxide until the colour of the solution was very pale yellow (45 min). Then 100 ml of 23 % potassium femcyanide was added dropwise with continuous stirring. The solution became green and yellow crystals formed near the end of addition. Further cooling, filtration and recrystallization fiom water yielded pale yellow product (80%).

1

H NMR in DtO/DSSlDCI, p H 1:6 2.110 (6H, s, CH3-pyr), 2.50-2.80 (lOH, m, CH3C(4),

CHzC(S)), 3.620 (4H, t, 'J

= 6.6 Hz, CHzOD), 4.725

(4H, s, CH2N>, 7.962 (2H, s, H-pyr),

8.03 1 (2H, s, CHO).

2.3.2

Thiamin Thiazolone Thiamin disulfide was synthesized fiom the method of Todd and Sykes (19(10 g)

was suspended in 3-methyl-2-propanol(200 ml) and refluxed for 2 h. Leaving the solution

to cool at 4OC gave a pale yeilow solid. Recrystallization fkom ethanol and water (2:7 v/v) gave white needles (60%)

I

24.

H N M R in D20/DSS/DCl, pH 1: 6 2.12 (3H, s, CH3-pyr), 2.57 (3H, s, CH3C(4)), 2.8 1

(2H, t, J' = 4.2 Hz,CHzC(5)), 3.75 (ZH, t, 2~ = 4.2 Hz,CH20D), 7.83 (lH, s, H-pyr).

2.3.3

Thiamin Thiazolone Diphosphate The modified phosphorylation procedure of Viscontini et al., (1949) was used to

phosphorylate thiarnin thiazolone 25. Phosphoric acid (85 %, 3 ml) was heated over a flarne until the solution became white and syrupy. Upon cooling to room temperature, the symp hardened to a glassy solid. Thiarnin thiazolone (1.0 g) was added to this glassy solid and this mixture was heated again to 115OC in an oil bath for 20 min. with occasional stimng with a glass rod. Upon cooling, the mixture resolidified. In an ice bath, the mixture was dissolved in water (2 rnL) with stimng ovemight.

3 LNMR ~

(in DzO, DCI,

DSS) indicated that phosphorylation did take place (peaks are significantly srnaller than

inorganic phosphate and pyrophosphate), but 'H NMR clearly shows the solution to be a mixture of the monophosphorylated, diphosphorylated and triphosphorylated species.

Undissoived solids were removed by vacuum filtration. Upon addition of 200 rnL 1: 1 95 % ethanol: diethyl ether, the solution went cloudy and a light yellow residue formed. Once the solution was allowed to settie, the liquid was carefully decanted and the oily residue was taken up again in 5 rnL water and triturated

twice more. The "P NMR spectrum of this mixture better shows the three phosphorylated species,

3 1 N-MR ~ @20/DSS/DCI):

6 0.601 (s, inorganic phosphate and TTMP), -9.5 17 (m,

diphosphate and TTDP), -23 -958 (m, triphosphate and T?TP).

The mixture was then carefully titrated to pH 4.0 and appiied to a cellulose plate with fluorescent indicator (Eastman). The solvent system, 10:1:6 95% ethanol: n-butanol:

0.15 sodium citrate p H 4, was used to separate the components 24. Each band was then scraped off the pIate and extracted fiom the adsorbent with water, filtered and lyophilized.

NMR was perforrned to identiQ each band. TTDP had Rf= 0.50.

2.4

Expression and Purification of Benzoylformate Decarboxylase Competent BL2 l(DE3) cells (20 pl) were transformed with the plasmid carrying an

ampicillin resistance gene and the BFD-His6 insert (2 pl) by heat shock (5 min at 37OC).

The cells were grown in LB media (250 pL) at 30°C for L hr, plated onto LB plates containing O.lmg/rnL ampicillin, and grown for 20 h in a 37°C incubator. One colony was selected for a growth in a 30 rnL culture. When 0D600= 1.0,glycerol was added to the 30

mL culture to a final concentration of 20 %. This stock culture was then fiozen in liquid

nitrogen and stored at -78OC. A pipette tip was used to scratch the stock culture and transferred uito new LB culture (30 mL) and grown at 37OC at 250 rpm. The 30 mL culture

was then added to a 1 L culture and grown for 3 hrs and then transferred to a 3 L culture. When OD = 1.2-1-5, the culture was induced with IPTG (60mg& 238.3 g/mol, final conc.

0.2 mM). After 24 h of growth, the cells were spun at 15 000 rpm for 40 min., fiozen in liquid nitrogen, and stored at -78OC until fùrther purification. Al1 purification steps were performed on ice. Once the cells (usually about 13 g) were thawed in lysis buffer (2-5 mWg wet ce11 weight), lysozyme (1 mg/mL) and PMSF (to give 1 rnM, 0.17 mg/mL) were added and incubated for 30 min. The cells were cracked open by sonication with six 10 s bursts at 20 W with a IO s cooling period between each bunt. The lysate was centrifüged at 10 000 rpm for 30 min at 4OC to remove cellular debris. Ni-NTA agarose was added to the supernatant at 1 m u 4 mL and mixed gently by shaking on a rotary shaker at IO rpm ovemight. The lysate-Ni-NTA agarose mixture was then loaded into a column and the flow-through collected. The column was washed with eight 5 mL fiactions of wash buffer (lysis buffer containing 20 mM imidazole) and then eluted with ten 4 mL hctions of elution buffer (lysis buffer containing 250 m . imidazole). SDS-PAGE anaiysis on 5 uL aliquots of the flow-through, wash fiactions and the elution fractions was performed. The elution fractions that contained a significant band at

57 kDa were pooled, concentrated using a Centriprep centifûgal filter device with a YM-10 MW membrane and the elution buffer exchanged for 50 mM potassium phosphate buffer, pH 7.0.

The spectrophotometric assay was used to obtain an approxirnate protein concentration 26: Protein concentration ( m m ) = 1.55

- 0.76 A260

An aliquot of the BFD-His6 preparation was then diluted appropriately and protein

content was determined more accurately using the Bradford assay (Bio-Rad) using bovine senun albumin as standard. The BFD-His6 enzyme was stored in 50 rnM potassium phosphate buffer, pH 7.0 with 0.1 rng/rnL of sodium azide as a preservative. Unused enzyme can be stored at -20°C as a lyophilisate.

2.5

Enzyme Assays

2.5.1 Direct Decarboxylase Assay The activity of the purified BFD-His6 was determined by initiating the reaction with benzoylformate to the reaction mixture (Table 2.1) equilibrated at 30°C in a 1.7 mL cuvette. The consumption of benzoylformate was followed at 343

MI

for 40 min. for three

different arnounts (x) of BFD. The extinction coefficient at 343 nm for benzoylformate in 150 mM potassium phosphate, pH7.0, containing 2.5 mM MgS04 was determined to be 79

m ~ cm-' " (results not shown). 1 U is defined as the arnount of enzyme that decarboxylates 1 p o l benzoylformate per minute under the above conditions.

2.5.2

Coupled Decarbo-xylaseAssay The coupled decarboxylase assay consisted of the following components (Table

2.2) incubated for 5 min. at 30°C.Three minutes pnor to the addition of benzoylformate,

HLADH (30 fi)and 0.05X BFD-Hisa (x) were added to the reaction mixture. The reaction was followed by the decrease in absorbance of NADH at 340 nrn over 60 min.

Table 2.1. Components of Direct Decarboxylase Assay Volume to take, @

Final concentration

500 mM phosphate buffer, pH 7.0

300

150 rnM

40 mM magnesium sulfate

63

2.5 rnM

Stock solution

1 10 mM thiamin diphosphate

1

50

1

1 Water 1 1X BFD stock

I I

547 - x

r500 mM benzoylformate

l

I I

40

I

Total volume

X

1

1000

0.5 mM

I I

I

20 mM

I

Volumes, &, to take

Final concentration

250 mM phosphate buffer, pH 7.0

200

50 mM

40 mM magnesium sulfate

63

2.5 mM

10 mM ThDP

50

0.5 rnM

20 mM NADH

17

0.36 m M

--

I

Water

620-x

I I

0.05X BFD

x

-

10 rnM benzoylformate Total volume

I I

Table 2.2. Components of Coupled Decarboxylase Assay Stock solution

1

I

1

40

1

1

1000

1

0.01 - 0.08 U 0.40 mM

2.6

Inhibition with Thiamin Thiazolone Diphosphate at 5 min. Incubation Period BFD-His6 solution was added to the following reaction mixture containhg TTDP

(x) and/or ThDP equilibrated at 30oC for 5 min. The reaction was then started with the addition of benzoylformate and followed by W at 340 nm.

Table 2.3. Effect of TTDP on ThDP-bound BFD-His6

I

Stock solutions

I

1 250 mM phosphate buffer, pH 7.0 1

200 63

40 rnM rnagnesium sulfate

X (to give 0-5 pM)

12.5 pM TTDP

1 10 pM ThDP

Volumes to take, pl

I

1 NADH, 20 mM I 1 benzoylformate, 10 mM, pH 7.0 1 Total volume I I

Y (to give 0-10 FM)

17 40 1000

Tirne Dependence of TTDP on Bm-Bis6 Activity

2.7

Different incubation mixtures of O. lx BFD-His6, and TTDP were made to a total volume of 300 uL and incubated at SOCfor 13 min., 1 80 min., and 23 h. An aliquot (30 pl)

was then added to the reaction mixture (Table 2.5) and the change in absorbance at 340 nm was fol1owed.

Table 2.4 Incubation mixtures of BFD-His6 and TTDP

1

Solution

1

I

Volumes to take, pl

BFD-Hk6, O. 1X

100

1O0

100

L O0

1 O0

Water

100

90

80

60

O

TTDP, 10 j.iM

O

10

20

40

1 O0

[TTDPI, CtM

O

5

10

20

50

Table 2.5 Reaction mixture for time dependence of TTDP on BFD-His6 Activity

1 Phosphate buffer, 400 nM, pH 7.0

I

100

MgS04, 40 mM

63

HLADH, 33 U/ml

50

NADH, 20 mM

20

Incubation mixture (Table 2.4)

30

1

1 Benzoylformate, 10 rnM Water

37 700

Total volume

1000

I

1

-

Chapter 3 Results

3.1

Synthesis of Thiamin Thiazolone Diphosphate Synthesis of TTDP was simple, however its purification by ion-exchange

chromatography (amberlite, Dowex, etc) was extremely difficult. TLC using the Gutowski solvent system gave reproducible results although isolation of the T'IDP band gave a very small yield. The ratio of inorganic phosphate to TTDP in the h a I T'ï'DP sample was relatively high (10: 1) despite repeated triturations with 1:1 95 % ethanol: diethyl ether.

3.2 Protein Concentration of Benzoyiforrnate Decarboxylase-Efis6 Preparation From the spectrophotometric assay 26, the protein concentration of the 1 X BFD-His6 stock is 1.62 mg/rnL. The ratio, A28~A260 was found to be 1.3 1 indicating little contamination by nucleic acid. The calibration curve for the Bradford assay using BSA as protein standard is shown in Figure 4. Using this method, protein concentration in the 1 X BFD-His6 stock was found to be 4.162 mg/rnL.

mg/mL BSA

Figure 3- 1. Calibration curve for Bradford assay

3.3

Direct Decarboxylase Assay Although there was no ThDP added to the growth mixture or during the purification

mixture, the final BFD-His6 preparation contained endogenous enzyme-bound ThDP. Results of a Spica1 direct decarboxylase assay are shown in Figure 3.2. The activity is linearly proportional to amount of protein as shown in Figure 3.3 and specific activity is 0.4 U/mL.

4

6

8 IO 12 14 16 18 20 22 2 4 2 6 283032 time, min

Figure 3.2. Direct assay to follow activity of BFD-His6

20

40

uL ZX BFD Figure 3.3. Rate of reaction is proportional to amount of BFD-His6

60

3.4

CoupIed Decarboxylase Assay

Since benzaldehye is a substrate for HLADH, the coupled assay c m be used to folIow the decarboxylation reaction with the advantage that it is more sensitive to decreases in absorbance of NADH. Absorbance was found to be linear in the range of 0.05 to G.3

mM NADH (Figure 3 -4). The extinction coefficient for NADH at pH 7.0 in 50 mM phosphate buffer c o n t a k g 2.5 mM MgS04 was 3.192 rdK1cxK1. Literature '*'O cites the extinction coefficient to be 6.220 &'cm-'.

We cannot account for this discrepancy. H oxidation, multiples of 5

To ensure that decarboxylation was coupled to W

pL of a 1/20 dilution of BFD-His6 stock were used for the assays (Figures 3.5 and 3.6)-

Reaction rates were determined fiom the iinear section of the spectra and expressed as p o l of NADH consumed per minute.

O

0.05

0.1

0.15

02

025

Concentration of NADE& mi'vf

Figurs 3.4. Absorbance of NADH at 340 n m

0.3

035

0.4

19

volume of 0.05X BFD used

2

O

4

6

8

10

12

14

16

tim e, min

Figure 3.5. Following oxidation of NADH in the coupled decarboxylase assay

O

20

40

60 volume of 0.05X BFD, ul

80

Figure 3.6. Checking for coupling of decarboxylation to NADH oxidation

In Figure 3.6, it appears that the rates are coupled between 20 and 40 pL of O.OSX BFD, since the rate at 40 pL 0.05X BFD is double that at 20 pL 0.05X BFD. Specifïc activity for the O.05X BFD enzyme is thus 0.00 10 p o l NADH consumed per min per pL of protein or lU/mL of protein. As previously reported, the amount of coupling enzyme

HLADH had to be at least 20 times more than BFD in order for the reactions to be coupled

3.5

Inhibition with Thiamin Thiazolone Diphosphate

The effect of TTDP was studied on 0.02 U of BFD-His6 with endogenous ThDP (Figure 3.7). It appears that loss of activïty is linearly proportional to TTDP concentration (Figure 3 -8).

1O

20

30

timt. min

Figure 3.7. Following reduction in activity of 0.02 U BFD-His6 by TTDP at 5 min incubation penod

2

4

u M TTDP

Figure

BFD-His6 activity as a fùnction of TTDP concentration

However, when the time dependence of l"L'DP incubation with BFD-His6 with endogenously bound ThDP was exarnined, it appears that TTDP has no effect on long-term

BFD activity:

1-0-13min I

'i

j -0- 180 min j

j+23h

TTDP concentration, uM

Figure 3 -9 Time dependence of TTDP on BFD activity

j

Chapter 4 - Discussion

4.1

BFD-His6 purification

Purification of BFD-Es6 by nickel chromatographywas a very straightforward and efficient procedure. SDS-PAGE showed that there were some contaminant proteins in the final preparation but hardly significant compared to BFD-His6. The measured protein concentration by the spectrophotometric assay was 2.5 times smaller than the value measured by the Bradford assay. Since ThDP does not absorb at 595 nm, the protein concentration value obiained by the Bradford assay would be more accurate.

4.2

Direct and Coupled Decarboxylase Assays

Previous direct and coupled assays '*'O had to be modified so that the decrease in absorbance of benzoylformate or NADH was over at least a 20-minute interval. This allows for more subtle changes in rate to be measured. The direct assay is a good estimate

of enzyme activity when the protein preparation has high active BFD-His6 content. However, sornetimes the BFD-His6 preparation was not very pure or active (for unknown reasons) and so the direct assay was not sensitive enough to detect the rate of decarboxylation. Coupling the BFD-catalyzed reaction to HLADH-catalyzed oxidation of

NADH allows for larger differences in absorbance to be measured. It is important that there is enough HLADH in the reaction mixture so that NADH oxidation is not the ratedeterrnining reaction in the following reaction sequence:

direct assay

benzoylformate-

n

HLADH

BFD benzaldehyde

NADH NAD

benzyl alcoho1 +

coupled assay Figure 3.10. Reactions in the direct and coupled decarboxylase assays

The coupled assay reflects the rate of formation of benzaldehyde. Benzaldehyde is only consumed by HLADH as fast as it is being made by BFD. Matters are a little more complicated in the direct assay, wtiich follows the disappearance of benzoylformate. But exactly at which step does benzoyiformate "lose" its absorbance? Does this reflect the rate of formation of MThDP, the rate of decarboxylation of MThDP or the rate of protonation

of the enarnine/carbanion species (See Figure 1.1)? Do any of the intermediates andlor transient species between benzoylformate and benzaldehyde have significant absorbance at 340 nm? Sergienko et al., 2000 observed a transient species with hm

= 400 nm that was

attributed to an enamine intemediate with ap-nitro substituent 6. Perhaps the decrease in absorbance may not be due solely to benzoylformate but maybe to a composite of species that absorb at 340 nrn. The specific activities fiom the direct and coupled assays were 0.4 U/ml and 1 U / d respectively. This suggests that a step prior to elimination of benzaldehye is kinetically significant. The exact step is an exciting topic for fiirthur research and discussion.

Specific activity is an indication of the purity of the enzyme preparation. Specific activity for the enzyme is much lower cornpared to the values obtained by Iding et al., 2000 'O

and Weiss et al., 1989 '. During purification, SDS-PAGE analysis indicate that the

protein preparation contains the 57 kDa monomers but does not reveal whether or not the active tetrarneric form of the enzyme has formed properly. The enzyme could be fuahur purified by phenyl Sepahrose FPLC 6. Several problems were encountered while performing the assays making subsequent measurements for the inhibition assays difficult and unreliable. The stability of

NADH is pH-dependent, and thus the extinction coefficient for NADH had to be measured penodically. In addition, the enzyme slowly loses activity and the rernaining activity during the inhibition studies has to be normalized. A significant problem is that the coupled decarboxylase assay would sometimes become "uncoupled" giving sigmoidal curves.

4.3

Inhibition with Thiamin Thiazolone Diphosphate Removal of ThDP on the enzyme required dialysis in 12 x 1 L 50 mM phosphate

buffer, pH 7.0 over one week. Alteinatively, resuspension of cells in 12-15 ml lysis buffer/g wet ce11 weight preceding purification gave a BFD-His6 preparation without any activity. Although Iding et al.. 2000

'O

stated that reconstitution of the apo-enzyme in 1

pM ThDP gave half-maximal activity in 24 hrs, reconsitution in our lab did not occur at

any concentration of ThDP, even after one week. Attempts to quanti@ the concentration

of ThDP in the BFD-His6 preparation were not successfbl because the absorption spectra of proteins and ThDP overlap. The effect of TTDP on the ThDP-bound BFD-His6 enzyme was investigated, reflecting the ability of TTDP to cornpete for the two ThDP-sites in the enzyme. A plot of activity vs. TTDP concentration (Figure 3.8) suggests that Ki is 5 W. Initially, it appears that the added lTDP was able to reduce enzyme activity of the endogenous ThDP-bound BFD-His6 by 20%. However, at higher concentrations of TTDP and at longer incubation periods (Figure 3.9), it appears that TTDP does not greatly affect enzyme activity. TTDP in the presence of bound ThDP cannot get into the active site even after 23 h. So, TTDP is

not a transition state analogue as it binds less tightly than ThDP to BFD (KM= 1pM)

Io,

nor is TTDP a competitive inhibitor for BFD. TTDP is aiso not a transition state analog for other ThDP-dependent enzymes 2 1-23 with the exception of E. coli pyruvate dehydrogenase complex ". However, the type of inhibition pattern exhibited by TTDP on BFD may be more complex. An "alternating sites" reaction pathway has been proposed where decarboxylation at the first active site may be dependent on the presence of substrate at the other active site ? Also, decarboxylation at one single site can occur but at a much slower rate. So, the overall observed rate of decarboxylation may be an average of the individual rates and the "alternating site" rate. This provokes many questions: If only one molecule of TTDP occupies one active site, c m substrate still bind at this site and allow decarboxylation to occur at the other active site? Or could TTDP in one active site have no effect on the other active site? Can two molecules of TTDP displace both ThDP molecules

to prevent decarboxylation? On average, how many molecules of TTDP are bound to the enzyme? However, no one has yet measured the reaction order of inactivation by substrate

'

analogs such as Ip-(halomethyl)benzoyl]fomates nor by coenzyme anaiogs such as

TTDP, thiamin thiothiazolone diphosphate and thiochrome diphosphate. If TTDP is not a transition state analog for BFD, then what is the transition state and can we design one that will bind much more strongly than ThDP? TTDP was designed for pyruvate dehydrogenase, and does not reflect the methyl group derived fiom pymvate. This suggests that pyruvate dehydrogenase does not utilize the methyl group to achieve the transition state. For BFD, whose substrate contains a phenyl group, it may use the steric group for critical interactions in the transition state. A better transition state analog for BFD might be 10 which incorporates the phenyl group into the enamine

structure:

Figure 4.1. A potential transition state analog for BFD

Another possibility is that the enamine/carbanion, 9, does not reflect the transition state of a kinetically significant step. Whether or not the transition state looks more like MThDP or HBzThDP has to be further clarified. Jordan et al.,1999, synthesized HBzTHDP and used it as a probe for thc nature of the transition state of a mutant form of pyruvate decarboxylase 19.

This compound forms its conjugate base on the enzyme and

leads to recovery of enzyme activity. An analog of HBzThDP thus should have more affinity for BFD. No one has yet synthesized MThDP for BFD but Kluger et al., 1981 were able to

synthesize 2-(a-1actyl)thiamin and found that the rate of decarboxylation is 1'O times srnaller than the dissociation step ".

They suggest that the enzyme rnay have to

destabilize 2-(a-1actyl)thiamin diphosphate (LThDP) in order for decarboxylation to occur. Also, if decarboxylation were the rate-determining step for the overall rnechanism, then analogs of LThDP would better approxirnate the transition state. With this in mind, methyl acetyiphosphonate (MAP) was designed to form adducts with ThDP that are analogous to the labile LThDP. Pyruvate dehydrogenase, pyruvate oxidase and pynivate decarboxylase al1 responded differently to the pyruvate analog. MAP is a very potent inhibitor of pyruvate dehydrogefiase, binding to the enzyrne better than pyruvate by a factor of 104. Together, MAP and pymvate dehydrogenase generate a reactive intermediate analog However, MAP inhibits pyruvate oxidase only at high concentrations and appears not to inhibit yeast pyruvate decarboxylase ". The results were rationalized by differences in stereoelectronic alignment of the phosphonate relative to the carboxylate.

Despite the above results, phosphonate analogs of benzoylformate that could generate the desired reactive intermediate analog in situ or phosphonate analogs of MThDP

might be potential inhibitors of BR) (Figure 4.2). Benzoylphosphonate, benzoyhydroxamate, and the a-hydroxybenzylphosphonates (HBP) would bind to the

enzyme but they would not able to undergo decarboxylation by BFD.

0-

H benzoylformate

phenylacetate

\ /