METHIONINE SULFOXIDE REDUCTASES: STUDIES ON THE REDUCING
REQUIREMENTS AND ROLE IN THE METABOLISM OF SULINDAC
by
David J. Brunell
A Dissertation Submitted to the Faculty of
The Charles E. Schmidt College of Science
in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
Florida Atlantic University
Boca Raton, FL
August 2009
ACKNOWLEDGEMENTS
I would like to express my deepest appreciation to my committee chair and mentor, Dr. Herbert Weissbach, for his faith in me as a scientist and his patience, guidance and unselfish dedication through all the long hours it took to complete my studies. I would also like to thank my other committee members, Drs. Nathan Brot,
David Binninger, Howard Prentice and Keith Brew for their thoughtful criticism and time away from busy schedules and Dr. Daphna Sagher for passing along to me her considerable expertise gained through many years of experience in the lab. My students
Chan Le, Anna ToiGB, Diana Navarro and Shari Selesky not only made contributions to the research but also extended their sincere friendship at a time when it was needed and appreciated. Special thanks also go to the other lab members who contributed to the research including Dr. Ian Moench and Alex Kreymerman.
Most especially, I would like to thank my wife Kateri for her patience and support and my mother Ruth Brunell for instilling in me from a young age a deep curiosity about how the world works.
iii ABSTRACT
Author: David J. Brunell
Title: Methionine Sulfoxide Reductases: Studies on the Reducing Requirements and Role in the Metabolism of Sulindac
Institution: Florida Atlantic University
Dissertation Advisor: Dr. Herbert Weissbach
Degree: Doctor of Philosophy
Year: 2009
The methionine sulfoxide reductase (Msr) enzymes catalyze the reduction of methionine sulfoxide (Met(O)) to methionine. The Msr enzymes protect cells against oxidative stress and may have a role in aging. The MsrA family of enzymes reduces stereospecifically the S epimer of free and protein-bound Met(O) while the MsrB family reduces the R epimer of Met(O) in proteins. It has been generally accepted, primarily from studies on MsrA, that the biological reductant for the Msr enzymes is thioredoxin
(Trx), although high levels of dithiothreitol (DTT) can be used as the reductant in vitro.
In contrast, certain MsrB enzymes show less than 10% of the activity with Trx as compared to DTT. This raises the possibility that in animal cells Trx may not be the direct hydrogen donor for the MsrB enzymes. Studies with bovine liver extracts have shown that thionein, the apoprotein of metallothionein, can function as a reductant for
iv the Msr proteins. Certain selenium compounds such as selenocystamine and selenocystine can also serve as potent reducing agents for the Msr enzymes.
Since an increased activity of Msr enzymes can reduce the level of oxidative damage in tissues, compounds that could activate Msr may have therapeutic potential. A high-throughput screening assay has been developed to screen large chemical libraries to find activators of MsrA, as well as specific inhibitors that could be useful research tools.
This study will be done in collaboration with The Scripps Florida Research Institute.
Sulindac was originally developed as a non-steroidal anti-inflammatory drug but has also shown efficacy in the treatment of certain cancers. The S epimer of sulindac is known to be reduced by MsrA, but the enzymes responsible for reduction of the R epimer are not known. An activity has been purified from rat liver which is capable of reducing the R epimers of sulindac, free Met(O) and a dabsylated Met(O) substrate, the latter suggesting that this enzyme may have properties similar to the MsrB enzymes.
The oxidation of the epimers of sulindac to sulindac sulfone has also been characterized, and the members of the cytochrome P450 family involved in the oxidation have been identified.
v METHIONINE SULFOXIDE REDUCTASES: STUDIES ON THE REDUCING REQUIREMENTS AND ROLE IN THE METABOLISM OF SULINDAC
List of Tables...... x
List of Figures...... xii
1 INTRODUCTION...... 1
1.1 Reactive oxygen species, oxidative damage and aging...... 1
1.2 Role of the methionine sulfoxide reductases...... 3
1.3 Genetic studies on the role of the Msr enzymes...... 6
1.4 Mechanism of action of the Msr enzymes...... 9
1.5 Reducing system for the MsrB enzymes...... 12
1.6 Substrate specificity of the Msr Enzymes...... 13
1.7 Msr as a catalytic antioxidant...... 13
1.8 Assays for Msr activity...... 15
1.9 Msr and disease...... 17
1.10 Sulindac and its relationship to the Msr system...... 19
1.11 Objectives of the current study...... 22
2 MATERIALS AND METHODS...... 23
2.1 Materials...... 23
2.2 Expression and purification of enzymes...... 23
2.3 Colorimetric DABS assay for Msr activity...... 24
vi 2.4 Purification of metallothionein (MT) from bovine liver...... 25
2.5 Preparation of thionein (T) and T(O) and assay of T(O) reduction by Trx...... 26
2.6 Analysis of zinc content...... 27
2.7 Separation of the sulindac epimers...... 27
2.8 Preparation of plasma samples from sulindac-treated animals...... 27
2.9 Assay of sulindac reduction ...... 28
2.10 Purification of sulindac-R reductase from rat liver...... 28
2.11 Assay of sulindac oxidation via cytochrome P450 enzymes...... 29
2.12 Induction of sulindac oxidation in human hepatocytes ...... 30
3 RESULTS...... 31
3.1 Studies on the reducing requirements for the Msr enzymes...... 31
3.1.1 Reduced Trx is not an efficient reducing agent for hMsrB2 and hMsrB3....31
3.1.2 Zn-MT in the presence of EDTA can serve as a reducing agent for Msr...... 32
3.1.3 T can function in the Msr system in the absence of EDTA...... 38
3.1.4 Trx can reduce T(O)...... 39
3.1.5 Selenocystamine (SeCm) enhances activity with mammalian Msr
enzymes...... 42
3.1.6 SeCm is reduced by Trx reductase and can directly reduce the Msr
enzymes...... 44
3.1.7 Selenocystine enhances MsrB activity but is not directly reduced by
Trx reductase...... 45
3.1.8 Thionein can reduce SeCm...... 46
vii 3.2 A high-throughput screening assay for MsrA...... 50
3.2.1 Development of the assay...... 50
3.2.2 Miniaturization of the assay...... 55
3.2.3 Follow up of active compounds...... 59
3.2.4 Optimization of any lead compounds...... 61
3.2.5 Problems using a coupled system to measure NADPH oxidation...... 61
3.2.6 Absorbance at 340 nm of chemicals in the library...... 62
3.2.7 Elimination of non-specific activators or inhibitors...... 62
3.3 The metabolism of sulindac...... 64
3.3.1 Sulindac metabolites detected in rat tissues ...... 64
3.3.2 MsrA knockout mice can reduce both sulindac epimers in vivo...... 65
3.3.3 Liver homogenates from MsrA knockout mice do not reduce sulindac-S....66
3.3.4 Partial purification of sulindac-R reductase from rat liver...... 68
3.3.5 Reducing requirements and substrate specificity of the purified material....72
3.3.6 Evidence that the sulindac-R and DABS-Met(O)-R reductases are the
same enzyme...... 73
3.3.7 Evidence that MsrB1 may not be the active component that reduces
sulindac-R...... 74
3.3.8 Oxidation of the sulindac epimers by rat liver microsomes ...... 76
3.3.9 Sulindac oxidation is induced by the sulindac epimers in human
hepatocytes...... 77
3.3.10 Studies with purified cytochrome P450 enzymes ...... 79
viii 4 DISCUSSION...... 80
4.1 Thionein and selenocompounds as reducing agents for the Msr enzymes...... 80
4.2 The search for MsrA activators and inhibitors...... 89
4.3 Metabolism of the sulindac epimers and relationship to the Msr system...... 90
4.3.1 Reduction of sulindac epimers...... 90
4.3.2 Oxidation of sulindac epimers...... 92
4.4 Directions for further study...... 94
4.4.1 Reducing systems for the Msr enzymes...... 94
4.4.2 High throughput screening...... 94
4.4.3 Further purification and identification of the sulindac-R reductase...... 95
4.4.4 Biological activity of the sulindac epimers...... 95
REFERENCES...... 97
ix TABLES
Table 1: Cellular molecules sensitive to oxidative damage...... 2
Table 2: Amino acids sensitive to oxidation...... 2
Table 3: Mammalian Msrs...... 6
Table 4: Genetic studies on the role of MsrA...... 9
Table 5: Thioredoxin and DTT as reducing agents with various Msr proteins...... 32
Table 6: Comparison of the activity of Msr proteins in the presence of Zn-MT or
DTT...... 37
Table 7: Trx stimulates the activity of MsrB3 in the presence of T(O)...... 41
Table 8: Effect of SeCm on the activity of Msr proteins using the Trx reducing
system...... 44
Table 9: Effect of addition of selenite and selenocystine to hMsrB reactions
containing the Trx system...... 46
Table 10: Effect of SeCm on the activity of Msr proteins using thionein (T) as the
reducing agent...... 47
Table 11: Metabolites of sulindac found in plasma of MsrA knockout mice
following IP injection...... 66
Table 12: Activity of mouse liver S-100 with sulindac-R and sulindac-S as substrates...68
Table 13: Purification of sulindac-R reductase...... 71
x Table 14: Relative activity purified protein: reducing system requirement and activity
with DABS-Met-(O)-R and sulindac-R...... 72
Table 15: Purification summary for the reduction of DABS-Met(O)-R and
sulindac-R...... 73
Table 16: Reductase activity of MsrB1 cysteine analog...... 76
Table 17: Sulindac oxidation by cytochrome P450 enzymes...... 79
xi FIGURES
Figure 1: Reactive oxygen species...... 1
Figure 2: Oxidation of methionine forms two different epimers...... 3
Figure 3: MsrA and MsrB reduce different epimers of Met(O)...... 5
Figure 4: Over expression of MsrA in flies extends their life span...... 7
Figure 5: Effect of over expression of MsrA in cardiac myocytes exposed to
hypoxia/reoxygenation...... 8
Figure 6: MsrA homology across organisms...... 10
Figure 7: MsrA reaction mechanism...... 11
Figure 8: Coupled reaction for the reduction of Met(O) catalyzed by MsrA...... 11
Figure 9: MsrB reaction mechanism...... 12
Figure 10: Alternative substrates for MsrA...... 13
Figure 11: Role of the Msr system in permitting Met to act as a catalytic antioxidant....14
Figure 12: DABS-Met(O) substrate...... 16
Figure 13: HPLC separation of DABS-Met and DABS-Met(O)...... 17
Figure 14: Interconversion of sulindac metabolites...... 19
Figure 15: Sulindac has a protective effect on normal cells and potentiates the
killing of cancer cells...... 21
Figure 16: Effect of heated liver S-100 concentration and EDTA on MsrB3 activity...... 33
xii Figure 17: Purification and properties of the active factor...... 35
Figure 18: T can supply the reducing system for hMsrB3 activity in the absence of
EDTA...... 39
Figure 19: Reduction of T(O) by Trx...... 41
Figure 20: Structure of selenium compounds...... 43
Figure 21: Effect of SeCm concentration on the activity of hMsrB3 using either the
Trx reducing system or T as the reducing agent...... 48
Figure 22: Time curve of hMsrB3 activity using the Trx reducing system either
with or without 50 µM SeCm...... 49
Figure 23: The Effect of MsrA concentration on the oxidation of NADPH...... 52
Figure 24: The stimulatory effect of selenocystamine (SeCm) on the activity of
MsrA...... 53
Figure 25: The inhibitory effect of N-ethylmaleimide (NEM) on the activity of
MsrA...... 55
Figure 26: Effect of SeCm and N-ethylmaleimide (NEM) on the reduction of
DMSO by MsrA in an 80 μl assay format...... 57
Figure 27: Effect of SeCm and NEM on the reduction of DMSO using a
fluorescence assay...... 58
Figure 28: Separation of the sulindac epimers using a chiral column...... 64
Figure 29: Chromatogram showing metabolites of R and S epimers of sulindac
in plasma of rats given IP injections of sulindac...... 65
Figure 30: DEAE separation of sulindac-R reductase activity...... 69
xiii Figure 31: G50 separation of sulindac-R reductase activity...... 70
Figure 32: Quaternary ammonium separation of sulindac-R reductase activity...... 71
Figure 33: Competition of sulindac-R activity with DABS-Met(O)-R and Met(O)-R....74
Figure 34: Sulindac sulfone formation by rat liver microsomes...... 77
Figure 35: Induction of sulindac oxidation in HepG2 human hepatocytes...... 78
Figure 36: Postulated role of Trx and MT in supplying the reducing requirement
for the Msr enzymes...... 85
Figure 37: Putative reactions involved in the reduction of the various Msr enzymes
by Trx, T, and SeCm...... 87
Figure 38: Metabolism of sulidac-R and sulindac-S to sulindac sulfone and
sulindac sulfide...... 93
xiv 1 INTRODUCTION
1.1 Reactive oxygen species, oxidative damage and aging
In 1956, Denham Harman proposed the free radical theory of aging1. According to this theory, the aging and ultimate death of an organism is due, in part, to damage caused to cellular components by free radicals (reactive oxygen species, ROS) of both exogenous and endogenous origin. As shown in Figure 1, these ROS have unpaired electrons or are otherwise highly reactive.
Figure 1: Reactive oxygen species.
Endogenous ROS are created mainly as a byproduct of metabolism, particularly in reactions within the mitochondrial electron transport chain. Regardless of the source of these ROS, they can cause damage to the cell’s biomolecules by virtue of their high reactivity. Since most of the ROS are created in the mitochondria, the majority of the damage also occurs there2. Despite a bewildering array of antioxidants and antioxidant enzyme systems including manganese superoxide dismutase (MnSOD), copper zinc
1 superoxide dismutase (Cu/ZnSOD), glutathione (GSH), glutathione reductase (GR) and catalase (CAT), damage to lipids, nucleic acids and proteins still occurs. Table 1 summarizes some of this damage.
Table 1: Cellular molecules sensitive to oxidative damage.
Molecule Damage DNA Thymine oxidation Lipids Peroxidation Proteins Amino acid oxidation
DNA damage and damage repair mechanisms have been extensively studied elsewhere3, and we shall restrict our discussion to the oxidation of amino acids in proteins and methionine in particular. Some of the amino acid oxidation products which can be formed are listed in Table 2.
Table 2: Amino acids sensitive to oxidation. Amino acid Oxidation product Cysteine Cystine Histidine Imidazole oxidation Lysine Carbonyl derivative Methionine Methionine sulfoxide Tyrosine Ring oxidation Tryptophan Oxyindole
2 1.2 Role of the methionine sulfoxide reductases
Methionine is particularly susceptible to oxidation. When methionine is chemically oxidized, it can form two different epimers of methionine sulfoxide
(Met(O)), R and S (Figure 2), Met(O) being chiral about the sulfur center.
Figure 2: Oxidation of methionine forms two different epimers. The primary
oxidation products are the R and S epimers of Met(O). Further oxidation yields
methionine sulfone.
Several enzymes, collectively referred to as methionine sulfoxide reductases
(Msr), can reverse this oxidation (Figure 2). The existence of an enzyme in yeast that reduced free Met(O) to Met was demonstrated as far back as 19604. Ejiri et al. first characterized the enzyme in E. coli which reduces free Met(O)5 by showing that methionine auxotrophs grew when supplemented with Met(O). Met(O) cannot be used for protein synthesis; therefore, a reductase for the free Met(O) must exist.
3 Around 1980, Brot et al. first isolated and characterized the enzyme which reduces protein-bound Met(O)6. In the particular system under study at the time, it was demonstrated that oxidation of the three methionine residues of E. coli ribosomal protein
L12 was reversible in reactions that contained three protein factors: thioredoxin (Trx), thioredoxin reductase (TrxB) and a protein initially referred to as peptide methionine sulfoxide reductase. This Msr would later be identified as MsrA and its location on the
E. coli chromosome determined through chromosome walking7. In E. coli, there are at least six members of this family that differ in their substrate specificity and location within the bacterial cell8,9. The protein that was first isolated and studied in greatest detail was E. coli MsrA (eMsrA)6,10-13, which specifically reduces Met-S-(O), whether in peptide linkage or as a free amino acid13,14.
In 2001, a second form of Msr, MsrB, was discovered in prokaryotes. A protein called YeaA from E. coli, in combination with MsrA, was demonstrated to completely reduce the oxidized methionine residues of calmodulin, whereas MsrA or YeaA alone could not15. The explanation for this was that MsrA and YeaA (now called MsrB) reduce two different epimers of Met(O), MsrA the S epimer and MsrB the R epimer (Figure 3).
About the same time, it was discovered that the pilB protein from Neisseria gonorrhoeae contained three domains: an N-terminal region which has a Trx-like motif,
MsrA and MsrB16. Concurrently, many investigators were identifying homologous genes in other organisms. MsrA and MsrB are now thought to exist in nearly all organisms17.
It was also soon discovered that multiple forms of MsrB exist, some containing zinc and others not18. Generally speaking, MsrB proteins reduce the Met-R-(O) in proteins but
4 have very weak activity with free Met-R-(O)15,16. However, in E. coli, there are specific soluble Msr reductases that reduce free Met-S-(O) and free Met-R-(O)8,19 as well as membrane-associated activities that can reduce both epimers of Met(O), whether free or peptide-linked9.
Figure 3: MsrA and MsrB reduce different epimers of Met(O).
The situation in mammalian cells is somewhat different. It appears that there is one gene that codes for MsrA20,21 but three separate genes that code for MsrB proteins22-
24. MsrA localizes primarily in the mitochondria or cytoplasm20,21, whereas the three
MsrB proteins have different subcellular localizations22. MsrB1 (originally called SelX) is a selenoprotein that is primarily found in the cytoplasm and nucleus, MsrB2
5 (originally called Cbs1) is present mainly in the mitochondria, and MsrB3, having two splice variants, is localized primarily in the endoplasmic reticulum and mitochondria.
All mammalian MsrBs contain zinc in their holo form22. MsrB1 contains selenocysteine in place of cysteine at the active site and requires a special stem-loop structure in the mRNA to code for insertion of selenocysteine in the translated protein via a special tRNA22. Table 3 summarizes the various forms of mammalian Msr enzymes.
Table 3: Mammalian Msrs. Name Subcellular location MsrA Mitochondria, cytoplasm MsrB1 (SelX) Nucleus, cytoplasm MsrB2 (Cbs1) Mitochondria MsrB3A Endoplasmic reticulum MsrB3B Mitochondria
1.3 Genetic studies on the role of the Msr enzymes
Most of the genetic studies have been done with MsrA, where it is known that when the gene is knocked out in E. coli25,26 yeast27 and animal cells28, these cells become more sensitive to oxidative stress. Over expression of MsrA in animal cells make them more resistant to oxidative stress29-32. In one study on mice lacking MsrA, the animals demonstrated an extreme sensitivity to high oxygen levels, a shortened life span and a neurological lesion33. A second, more recent, study also showed an increased sensitivity to oxidative stress but failed to show a shortened life span or neurological lesion34.
The over expression of MsrA in Drosophila makes them more resistant to paraquat treatment, but even more impressive is that these flies exhibit a 70% increase in their life span, as shown in Figure 435. Although aging in most species is generally
6 associated with a decline in physical activity, in Drosophila over expressing MsrA there was a marked delay in the onset of the senescence-induced decrease in physical activity.
There was also a delay in the overall loss of reproductive vigor of these animals. These animals produced more pupa than control animals and retained fertility for a longer period of time. Thus, their extended life span is also accompanied by a higher quality of life.
Figure 4: Over expression of MsrA in flies extends their life span. For this
experiment, a GAL4-UAS promoter system was used, with the GAL4 activator line (elav)
expressing predominantly in the nervous system. The parents were homozygous for
UAS-MsrA (○) and elav-GAL4 (□). The offspring (●) over expressing MsrA
demonstrate a 70% increase in life span. (from Ruan et al.35)
Another example is a study performed with cardiac myocytes32, where hypoxia and reperfusion were used to cause cell death due to oxidative damage. Figure 5 shows that over expression of MsrA, obtained by transfection of an adenovirus containing the msrA gene, significantly protected these cells, as seen by the reduced release of lactate dehydrogenase, which is a marker for cardiac cell damage.
7 In the case of the MsrB family, the deletion of a free Met(O)-R reductase in yeast resulted in an increased sensitivity to oxidative stress and a decreased lifespan, while over expression in the same organism resulted in greater resistance to oxidative stress36.
However, in flies, a recent study found no increase in life span or resistance to oxidative stress37 from over expression of MsrB.
Figure 5: Effect of over expression of MsrA in cardiac myocytes exposed to
hypoxia/reoxygenation. An adenovirus vector was used to transfect the cells and the
death of the cells was assayed by measuring the release of lactate dehydrogenase.
Where indicated, hypoxia was for 24 hours followed by 20 hours of reoxygenation.
(Taken from Prentice et al.32)
Table 4 summarizes the genetic studies relating to the knockout or over expression of MsrA.
8 Table 4: Genetic studies on the role of MsrA. Knockout of MsrA Over expression of MsrA Bacteria, yeast and animal Animal cells in culture cells are more sensitive to have increased resistance to ROS ROS Mice may have shortened life Flies are resistant to span, are oxygen sensitive and oxidative stress and have may have a neurological extended life span lesion
1.4 Mechanism of action of the Msr enzymes
MsrA is a protein of about 25 kDa., and the three-dimensional structures have been determined for bovine, E. coli and Mycobacterium tuberculosis MsrA, and the catalytic mechanism has been elucidated10,11,38,39. There is a highly-conserved amino acid motif, GCFWG, among MsrA enzymes across different species (Figure 6). Most organisms also have at least one resolving cysteine in the C-terminal region and many have two. The basic mechanism for the reduction of Met(O) and the regeneration of
MsrA is illustrated in Figure 7 using the bovine enzyme40. The three cysteine residues highlighted in Figure 6 are labeled A, B and C in Figure 7. Cysteine A, located near the
N terminus, is the main catalytic site that interacts with the substrate Met(O). After a nucleophilic attack on the methionine sulfoxide residue and the formation of a sulfenic acid intermediate, a disulfide bond is formed between cysteines A and B. After a nucleophilic attack of cysteine C upon cysteine B, the MsrA enzyme is left with a disulfide bond between cysteine residues B and C. For in vitro experiments, this bond can be reduced by either the Trx system (Trx, TrxB and NADPH) or dithiothreitol
(DTT), and reduction of free Met(O) was shown to utilize Trx in E. coli 41,42. Trx
9 contains two cysteine thiol groups which are involved in the reduction of Msr enzymes.
TrxB transfers hydrogens from NADPH to Trx to keep it in the reduced state (Figure 8).
MLSASRRT--LQLLSSSIPVRMMGDSSSKVISAEEALPGRTESIPVAAKHHVSGNRTVEP 58 RAT MLSASRRA--LQLLSSANPVRRMGDSASKVISAEEALPGRTEPIPVTAKHHVSGNRTVEP 58 MOUSE MLSATRRA--LQLFHSLFPIPRMGDSAAKIVSPQEALPGRKEPLVVAAKHHVNGNRTVEP 58 BOVIN MLSATRRACQLLLLHSLFPVPRMGNSASNIVSPQEALPGRKEQTPVAAKHHVNGNRTVEP 60 HUMAN ------MSLFD------KKHLVSPADALPGRNTPMPVATLHAVNG-HSMTN 38 ECOLI
FPEGTQMAVFGMGCFWGAERKFWLLKGVYSTQVGFAGGYTRNPTYKEVCSEKTGHAEVVR 118 RAT FPEGTQMAVFGMGCFWGAERKFWVLKGVYSTQVGFAGGHTRNPTYKEVCSEKTGHAEVVR 118 MOUSE FPEGTQMAVFGMGCFWGAERKFWTLKGVYSTQVGFAGGYTPNPTYKEVCSGKTGHAEVVR 118 BOVIN FPEGTQMAVFGMGCFWGAERKFWVLKGVYSTQVGFAGGYTSNPTYKEVCSEKTGHAEVVR 120 HUMAN VPDGMEIAIFAMGCFWGVERLFWQLPGVYSTAAGYTGGYTPNPTYREVCSGDTGHAEAVR 98 ECOLI ^^^^^ VVYRPEHVSFEELLKVFWENHDPTQGMRQGNDCGTQYRSAVYPTSAVQMEAALKSKEEYQ 178 RAT VVYRPEHISFEELLKVFWENHDPTQGMRQGNDFGTQYRSAVYPTSAVQMEAALRSKEEYQ 178 MOUSE VVFQPEHISFEELLKVFWENHDPTQGMRQGNDHGSQYRSAIYPTSAEHVGAALKSKEDYQ 178 BOVIN VVYQPEHMSFEELLKVFWENHDPTQGMRQGNDHGTQYRSAIYPTSAKQMEAALSSKENYQ 180 HUMAN IVYDPSVISYEQLLQVFWENHDPAQGMRQGNDHGTQYRSAIYPLTPEQDAAARASLERFQ 158 ECOLI
K-VLSKHGFGPITTDIREGQVFYYAEDYHQQYLSKNPDGYCGLGGTGVSCPTAIKK 233 RAT K-VLSKHNFGPITTDIREGQVFYYAEDYHQQYLSKNPDGYCGLGGTGVSCPMAIKK 233 MOUSE K-VLSEHGFGLITTDIREGQTFYYAEDYHQQYLSKDPDGYCGLGGTGVSCPLGIKK 233 BOVIN K-VLSEHGFGPITTDIREGQTFYYAEDYHQQYLSKNPNGYCGLGGTGVSCPVGIKK 235 HUMAN AAMLAADDDRHITTEIANATPFYYAEDDHQQYLHKNPYGYCGIGGIGV-CLPPEA- 213 ECOLI ^ ^ Figure 6: MsrA homology across organisms. The conserved GCFWG motif and the
C-terminal resolving cysteines are shown in bold type.
10
Figure 7: MsrA reaction mechanism. A Cys 72, B Cys 218, C Cys 227. Details of
the mechanism are given in the text. (From Weissbach et al40.)
Figure 8: Coupled reaction for the reduction of Met(O) catalyzed by MsrA.
NADPH is the hydrogen donor for the reduction of oxidized Trx (Trx(O)) by thioredoxin
reductase (TrxB). The reduced Trx reduces Met(O) catalyzed by MsrA with the
formation of Met and regeneration of Trx(O).
The mechanism for the MsrBs is somewhat more complicated (Figure 9)43. For
MsrB1, the active-site cysteine is replaced by selenocysteine, and a selenenic acid intermediate is formed upon reduction of the substrate. A selenenylsulfide bond with a resolving cysteine is then formed, and Trx is again assumed to reduce MsrB1 to its fully-
11 reduced form. In the case of MsrB2 or MsrB3, where there is no resolving cysteine, a single cysteine participates in the reaction and a sulfenic acid intermediate is formed, which was assumed to be directly reduced by Trx.
Figure 9: MsrB reaction mechanism. (From Kim and Gladyshev43)
1.5 Reducing system for the MsrB enzymes
No comprehensive study comparing the reducing requirements of MsrA and
MsrB has been done to date. Based on the studies with MsrA, it was assumed that Trx was the reducing agent for all of the Msr enzymes. Recently, the human MsrB genes have been of interest to us as part of our studies on the role of the Msr system in protecting lens and retinal cells against oxidative damage31,44,45. Using a colorimetric assay for Msr activity (see Materials and Methods), it was surprising to find that Trx serves very poorly as the reducing system for human recombinant MsrB2 (hMsrB2) and
MsrB3 (hMsrB3). This prompted a search for a factor in animal tissues which could provide greater reducing power for MsrB, as described below in Results.
12 1.6 Substrate specificity of the Msr Enzymes
Unlike most members of the MsrB family, MsrA has activity towards a broad range of substrates46. This enzyme will reduce the S-epimer of Met(O) in protein linkage, free Met(O) and the S epimer of other compounds that have a methyl sulfoxide moiety, including the drug sulindac47 and a compound as simple as DMSO46 (see Figure
10). In all cases the sulfoxide moiety is reduced to a sulfide.
Human MsrB2 and MsrB3 were tested, and it was found that they will reduce protein-bound but not free Met(O)-R. There is an enzyme that reduces free Met(O)-R in
E. coli8,19 and yeast36, but no homologous enzyme has been found in mammalian tissue.
Figure 10: Alternative substrates for MsrA.
1.7 Msr as a catalytic antioxidant
The ability of Msr to repair oxidatively-damaged proteins makes this enzyme an important contributor to the cell’s ability to manage oxidative stress, as discussed above.
It has been suggested48 that the repeated oxidation of methionine residues, followed by
13 reduction by Msr, is a way to catalytically scavenge ROS, which minimizes potentially- irreversible damage to other amino acid residues as well as lipids and nucleic acids. In this mechanism, Met residues in proteins function as catalytic antioxidants. Each round of Met oxidation and reduction by the Msr system will destroy one equivalent of ROS
(see Figure 11).
Figure 11: Role of the Msr system in permitting Met to act as a catalytic
antioxidant. Met or another suitable methyl sulfide reacts with ROS, forming a
sulfoxide and preventing damage to other biological molecules. With repeated recycling
from sulfoxide to sulfide, the target molecule acts as a catalytic antioxidant in the
presence of Msr enzymes.
Support for this catalytic antioxidant role of Met in proteins has come from cell culture experiments. Over expression of MsrA in PC12 cells causes the level of ROS to decrease30, whereas knocking out of MsrA in lens cells results in an increase in the ROS levels in the cells28. Thus, the unique feature of the Msr system is that it can protect cells against oxidative damage in two ways, by repairing damage to proteins in which critical
Met residues have been oxidized and as an ROS scavenging mechanism in which Met
14 residues in proteins function as catalytic antioxidants due to the action of MsrA and
MsrB.
1.8 Assays for Msr activity
Effective study of any enzyme system requires a fast, reliable, accurate and cost- effective assay. Enzyme assays can be divided into two distinct groups: direct and coupled. In the direct Msr assay, the formation of the sulfide product or loss of sulfoxide substrate is measured. In the coupled assay, the metabolism of another compound is related stoichiometrically to the amount of substrate utilized or product formed.
An example of a direct assay for the Msr system would be measuring the formation of Met from Met(O). Since the concentration of Met is not easily measured spectrophotometrically (its low-UV absorbance overlaps that of many other molecules), its detection can be facilitated by reacting it with nitroprusside, which forms a colored compound that can be easily measured in the visible region of the spectrum. A sensitive, radioactive, extraction-based method was also developed using tritium-labeled N-acetyl-
Met(O), which is converted to N-acetyl-Met49. Another method for the direct measurement of Msr activity uses the colorimetric substrate 4-N,N- dimethylaminoazobenzene-4-sulfonyl-Met(O) (DABS-Met(O), see Figure 12). The reduced form is less hydrophilic than the sulfoxide and can be extracted from an aqueous solution using aromatic solvents. After incubation of the substrate with Msr, the reduced substrate, DABS-Met, can be extracted with benzene or xylene and its absorbance read
15 at 436 nm. The absorbance is proportional to the amount of substrate reduced and, thus, to the activity of the enzyme.
Figure 12: DABS-Met(O) substrate.
An alternative to extraction of the product is to separate product and substrate using thin-layer chromatography or HPLC. Figure 13 shows that the use of HPLC permits the measurement of very small quantities of substrate and/or product.
16 Figure 13: HPLC separation of DABS-Met and DABS-Met(O). For the separation, a
4.6 X 75 mm C-18 column and a mobile phase of 50 mM sodium acetate buffer, pH 4.73
and methanol, 50:50, were used. 10 pmol of each chemical species was injected.
As an example of a coupled assay for Msr activity, first consider the coupled reactions in Figure 8. As Met(O) is reduced to Met, there is a stoichiometric conversion of NADPH to NADP. NADPH has an absorbance at 340 nm (ε = 6220 M-1cm-1) while
NADP does not. It is therefore possible to determine the progress of the reaction by measuring the decrease in absorbance at 340 nm.
1.9 Msr and disease
It is now widely accepted that oxidative damage is an important factor in many age-related diseases and may be the most important factor in the etiology of aging. In addition to the involvement of MsrA in aging, Met(O) and MsrA have also been implicated in some diseases such as emphysema, cardiovascular disease and lens and retinal diseases. Alpha-1-proteinase inhibitor (α1PI) is a major regulator of neutrophil elastase activity in the respiratory tract of humans. This protein binds elastase and
17 thereby inhibits its function. An imbalance between α1PI and elastase has been hypothesized to be a cause of pulmonary emphysema caused by smoking as well as adult respiratory distress syndrome due to the inactivation of α1PI. This inactivation is thought to be the result of the oxidation of a methionine residue essential for binding and inhibition of elastase activity50. Further proof that methionine oxidation was involved in
α1PI inactivation came from studies that showed that the chemical oxidation of α1PI resulted in the inability of the protein to bind and thus inactivate elastase51. This inactivation was partially reversed when the oxidized protein was incubated with
MsrA51. A similar result was obtained when the source of the α1PI was isolated from the bronchoalveolar lavage fluid of patients suffering from adult respiratory distress syndrome52. In this case the α1PI was already oxidized in vivo due to the disease.
Cardiovascular disease is another example of the involvement of Met(O) residues in proteins. ApoA1, which is the major protein of high density lipoprotein (HDL), is involved in reverse cholesterol transport, which protects against atherosclerosis by removing cholesterol from cells of the artery wall. It is known that oxidation of ApoA1 by HOCl results in impaired cholesterol efflux. This is due, in part, to the oxidation of methionine residues to Met(O) in ApoA1. The presence of Met(O) residues in circulating
HDL has been observed. Recent in vitro studies have shown that when oxidized ApoA1, which is defective in transporting cholesterol out of cells, is incubated with MsrA/B, reverse cholesterol transport is restored along with the reduction of the Met(O)53 residues. Thus the Msr system could play an important role in atherosclerosis. MsrA has also been shown to protect lens and retinal cells against oxidative damage28,46. It is
18 clear from the above studies that an activator of MsrA could provide a novel therapeutic approach to diseases caused by oxidative damage resulting from increased ROS production.
1.10 Sulindac and its relationship to the Msr system
Our interest in sulindac arose from the finding that the S epimer is a substrate for
MsrA47. Sulindac, first described in 197254, was originally developed as a non-steroidal, anti-inflammatory drug (NSAID)55. Sulindac is a prodrug with a sulfoxide moiety which requires in vivo reduction to sulindac sulfide, the active metabolite56. Sulindac sulfide exhibits anti-inflammatory activity by inhibiting the cyclooxygenase (COX1 and COX2) enzymes and hence prostaglandin synthesis57.
Figure 14: Interconversion of sulindac metabolites.
Sulindac, like Met(O), has a chiral sulfur center and therefore exists as both R and S epimers (Figure 14). In animals, its in vivo metabolism results in the
19 interconversion between the sulfoxide, sulfide and sulfone forms. As explained earlier, the S epimer of sulindac is reduced to sulindac sulfide by MsrA47. The enzyme responsible for reduction of the R epimer is not known. Sulindac sulfide is oxidized to sulindac-R by flavin-containing monooxygenase in a stereospecific manner58. The oxidation to the sulfone is considered irreversible in vivo. Like most xenobiotics, the oxidation of sulindac is thought to be via the cytochrome P45059 system of enzymes.
Previous studies have shown induction of the P450 system by sulindac via the aryl hydrocarbon receptor60,61, but they did not characterize the specific enzymes responsible for oxidation of the individual epimers.
In addition to its anti-inflammatory effects, sulindac has also shown activity against a number of cancer cell lines. In 1983, Waddell et al. demonstrated that sulindac was superior to other NSAIDS in the treatment of colorectal polyposis62. This led to a series of investigations into the effectiveness of sulindac against a number of different cancers including colon63, lung64, prostate65 and multiple myeloma66. Recently, our lab has shown that sulindac has efficacy in treating skin cancer when combined with a source of oxidative stress such as hydrogen peroxide67. Other investigators have reported on the treatment of internal tumors, using sulindac combined with arsenic trioxide as the source of oxidative stress64. Figure 15 illustrates a recent study showing that, in the presence of oxidative stress, sulindac can protect normal lung cells and potentiate the killing of lung cancer cells68.
20 Figure 15: Sulindac has a protective effect on normal cells and potentiates the killing of cancer cells. Lung cancer cells (A) or normal lung cells (B) were incubated in the presence (■) or absence (□) of 500 μM sulindac for 48 hr. Cells were then washed to remove the free sulindac prior to incubation for 2 hr with the indicated concentration of
TBHP and cell viability was measured using the MTS assay. From Marchetti et al.68
21 1.11 Objectives of the current study
Based on the background presented in this section, there are a number of unanswered questions that this study will address:
• Given the poor in vitro activity of the MsrB2 and MsrB3 enzymes using Trx as a
reducing system, are there other reducing systems or compounds that can
function in place of Trx?
• Given the therapeutic potential for activators of the Msr enzymes, can a high-
throughput screening (HTS) assay be developed to search compound libraries for
activators of the Msr enzymes? Specific inhibitors would have experimental
application as well.
• What is the role of the Msr enzymes in the metabolism of the sulindac epimers;
specifically, which enzymes reduce the R epimer?
22 2 MATERIALS AND METHODS
2.1 Materials
Methionine sulfoxide, dabsyl chloride (4-N,N-dimethylaminoazobenzene-4- sulfonyl chloride, DABS-Cl), 4-(2-Pyridylazo)resorcinol (PAR), selenocystamine
(SeCm), selenocysteamine (SeCem), selenocystine, sodium selenite (Na2SeO3), sodium selenate (Na2SeO4), selenomethionine, ebselen, sulindac [(Z)-5-Fluoro-2-methyl-1-[p-
(methylsulfinyl)benzylidene]indene-3-acetic acid], NADPH, glucose-6-phosphate, magnesium chloride, DTT and certain proteins, including human Trx and rabbit liver Zn-
MT, were purchased from Sigma-Aldrich, unless specified otherwise. DABS-Met-S-(O) and DABS-Met-R-(O) were prepared by derivatizing the amino group of the Met-R-(O) or Met-S-(O) epimers69 with DABS-Cl70. Rat Trx reductase (TR3) was a generous gift from Vadim Gladyshev (University of Nebraska, Lincoln). Recombinant cytochrome
P450 enzymes (Supersomes™) were obtained from BD Gentest (Woburn, MA, USA).
Sprague-Dawley rats of both sexes were supplied by Charles River Laboratories. Mice homozygous for knockout of the MsrA gene were generously supplied by Dr. R. Levine at the National Institutes of Health.
2.2 Expression and purification of enzymes
Clones for Trx and TrxB (E. coli), bovine MsrA (bMsrA), E. coli MsrA and MsrB
(eMsrA and eMsrB) and human MsrB2 (hMsrB2), the latter generously provided by
23 Todd Lowther (Wake Forest University School of Medicine, Winston–Salem, NC), were over expressed in E. coli, and the respective proteins were purified as described in the respective references16,46,71. The human MsrB3 (hMsrB3) cDNA from the human lens, graciously provided by Dr. J. Fielding Hejtmancik (NIH, Bethesda, MD), was amplified by PCR, cloned into a pET vector, and over expressed in BL21 E. coli cells. The harvested cells were suspended in 1/100 volume of original culture by using 50 mM Tris
(pH 7.4). After sonication and centrifugation at 10,000 × g, the supernatant was fractionated on a Sephadex G-75 column. Active fractions were combined, and protein purity (> 80%) was confirmed by SDS-PAGE.
2.3 Colorimetric DABS assay for Msr activity
This assay is based on the reduction of DABS-Met(O) to DABS-Met72 using a modification of a previously described procedure47. Reaction mixtures contained 100 mM Tris, pH 7.4, 100 nmol of the indicated DABS-Met(O) epimer and either 1–5 µg of purified Msr enzyme, as indicated, or suitable amounts of tissue extract. The reducing system consisted of either 15 mM DTT, 5.6 nmol of thionein or the Trx system (1 mM
NADPH, 2.4 µg of Trx reductase, and 10 µg of Trx). The total reaction volume was 100 or 200 µl, as specified, and the incubations were done at 37°C for the time specified.
Under these incubation conditions, the enzymatic reactions were proportional to enzyme concentration until 75% (75 nmol) of the substrate was reduced. The amount of the reduced product DABS-Met was determined either spectrophotometrically or via HPLC.
For the spectrophotometric assay, the 200 µl reactions were terminated by adding 200 µl
24 of 1 M sodium acetate, pH 6.0, followed by 100 µl of acetonitrile and extracted with 1 ml of benzene. 100 nmol of the product, DABS-L-Met, extracted into the benzene layer, gave an optical density of 1.7 at 436 nm. For the HPLC assay, the 100 µl reactions were terminated by adding 300 µl acetonitrile followed by vortexing, centrifugation and injection of 20 µl of the supernatant onto a C18 column (4.6 × 75 mm). With a mobile phase of 55/45 50 mM sodium acetate buffer (pH 4.73)/acetonitrile, DABS-Met(O) appears at 0.98 minutes and DABS-Met appears at 1.7 minutes. The results are presented as nmol or pmol of product formed in the incubations in the specified time.
Where appropriate, assays were repeated a minimum of three times, and the results in the tables represent typical experiments. When the bovine liver fractions were tested for thionein reducing capability, DTT was omitted, and the Trx reducing system and 5 mM
EDTA were added where indicated. In experiments using oxidized thionein (T(O)), the incubations did not contain EDTA. This assay could be used with purified preparations of MsrA and MsrB as well as crude extracts of mammalian tissues. Bacterial extracts, in the presence of a reducing system, destroyed the substrate, and further studies are needed to adapt the assay to bacterial extracts.
2.4 Purification of metallothionein (MT) from bovine liver
Fresh bovine liver was homogenized in three volumes of 50 mM Tris (pH 7.4) and centrifuged at 10,000 × g for 30 min and then at 100,000 × g for 16 h (S-100). The
S-100 fraction was heated at 80°C for 5 min and centrifuged to remove precipitated proteins (heated S-100). Once the active material was suspected to be a metallothionein
25 (MT), further purification followed an established method for MT73. Using a Bio-Rad
DuoFlow HPLC system, the heated S-100 was placed on a sizing column (Superdex 75
HR 10/30) followed by DE-52 anion-exchange chromatography. The fractions were routinely monitored at 240 and 280 nm. In accord with prior studies73, two distinct peaks of activity were eluted from the DE-52 column that corresponded to Zn-MT-1 and
Zn-MT-2, as described in Results.
2.5 Preparation of thionein (T) and T(O) and assay of T(O) reduction by Trx
T and T(O) were prepared from Zn-MT by modification of the procedure described by Klein et al74. Briefly, purified Zn-MT was dialyzed against 10 mM HCl
(pH 2.0) containing 150 mM NaCl for 12 h at 4°C. The protein after dialysis is reduced, metal-free T and appears stable when left at pH 2.0. To study the activity of T in the Msr system, T was neutralized and added to the reaction mixtures immediately before the incubations were initiated. To oxidize T, 0.75 volumes of 50 mM Tris base were added to the T sample to bring the pH to 8.5. Under these conditions, ≈50% of the sulfhydryls became oxidized after 4 h at room temperature or 2 h at 37°C. With longer incubations, the T(O) started to precipitate. The assay for free sulfhydryl groups used 5,5′- dithiobis(2-nitrobenzoic acid) (DTNB) as described by Li et al75.
To study the reduction of T(O) by Trx, the reaction mixtures contained (in a total volume of 1 ml) 100 μM NADPH, 26 μg of Trx, 6 μg of TrxB and 28 μg of partially- oxidized T (see above). The oxidation of NADPH was followed at 340 nm at room temperature.
26 2.6 Analysis of zinc content
Zinc was quantitatively determined in the MT preparations by using the PAR reagent76,77. The samples (100 μl) were incubated with 10 mM N-ethylmaleimide for 1 h at room temperature. PAR (100 nmol) was then added, and the sample was diluted with water to 1 ml. The Zn–PAR complex was measured at 500 nm. Ten nanomoles of zinc gave a reading of 0.720 at 500 nm. A complete metals analysis of the purified MT preparation was performed by Joseph Caruso (University of Cincinnati, Cincinnati) by using an Agilent 7500 inductively-coupled plasma mass spectrometer (Agilent
Technologies, Palo Alto, CA). Molecular weight determinations of the purified protein were performed by Peter Yau (University of Illinois, Urbana–Champaign) by using electrospray MS.
2.7 Separation of the sulindac epimers
The R and S epimers of sulindac were separated as described previously58 with the exception that a mobile phase of 65/35/0.1 hexane/ethanol/acetic acid was used. The eluent was monitored at 330 nm, and the R and S epimers appear at 12.2 and 14.5 minutes post-injection, respectively.
2.8 Preparation of plasma samples from sulindac-treated animals
Heparinized blood samples from either rats or mice were obtained immediately after euthanizing the animals, and the plasma was separated by centrifugation at 18,000
× g for 20 minutes.
27 2.9 Assay of sulindac reduction
Reaction mixtures (100 μl) consisted of 50 nmol sulindac epimer, 15 mM DTT and the appropriate amount of partially-purified rat liver extract in 100 mM Tris buffer, pH 7.4. Where indicated, 5 μg thioredoxin, 1.2 μg thioredoxin reductase and 1 mM
NADPH were substituted for the DTT. The reactions mixtures were incubated at 37°C for one hour and the reactions stopped by the addition of three volumes of acetonitrile.
Quantitation of reduced product was by C18 HPLC using a mobile phase of 25/75 50 mM sodium acetate buffer, pH 4.73/acetonitrile. Detection was by UV absorbance at a wavelength of 330 nm.
2.10 Purification of sulindac-R reductase from rat liver
Rat livers from freshly-euthanized animals were homogenized in a Waring blender at high speed for 60 seconds in three volumes of 50 mM Tris buffer, pH 7.4.
The homogenate was centrifuged for 20 minutes at 10,000 × g, and the supernatant thereof was centrifuged for 2 hours at 100,000 × g. The supernatant after ultracentrifugation (S-100) contained the soluble fraction of cytosolic proteins. An ammonium sulfate precipitation was performed at 4°C, with the proteins remaining soluble between a 30-70% range of salt saturation retained. The material was then applied to a 50 ml diethylaminoethyl (DEAE) fast flow column, after dialysis against the equilibration buffer of 20 mM Tris, pH 8.0. Selective elution from the DEAE column was performed by introducing a salt gradient from 0-200 mM KCl, with the active material eluting at a salt concentration of approximately 100 mM (a flow rate of 5
28 ml/min. was maintained throughout). The active fractions were pooled and concentrated and applied to a G50 superfine gel filtration column, 1 × 30 cm, equilibrated with 50 mM Tris, pH 7.4, plus 0.15 M NaCl. Half-milliliter fractions were collected from the
G50 column at a flow rate of 0.25 ml/min., and the active material was eluted after approximately 19 ml of flow volume post injection. After pooling and buffer exchange, the peak fractions from the G50 column were applied to a quaternary ammonium (Q) ion exchange column, 7 × 35 mm, equilibrated with 20 mM Tris, pH 8.0. Selective elution was performed over a KCl gradient of 0-150 mM, with the active material appearing at a salt concentration of approximately 25 mM.
2.11 Assay of sulindac oxidation via cytochrome P450 enzymes
A total of 50 nmol of sulindac epimer was incubated with rat microsomes or recombinant human and/or rat CYP1A1, 1A2, 1B1, 2B1, 2C8, 2C9, 2C19, 2D6, 3A2, or
3A4 (BD Supersomes™) and an NADPH-regenerating system (1.5 mM NADPH, 5 mM glucose 6-phosphate, 50 ng glucose 6-phosphate dehydrogenase, and 5 mM MgCl2) in potassium phosphate buffer (pH 7.4, 50 mM) for 60 minutes at 37°C. The reaction was then stopped with 3 volumes of acetonitrile and centrifuged. The supernatants were directly assayed by HPLC with UV detection. Positive controls for enzyme activity were performed with various substrates according to the manufacturer's (BD Gentest) recommendations (data not shown).
29 2.12 Induction of sulindac oxidation in human hepatocytes
HepG2 human hepatocarcinoma cells were obtained from ATCC and cultured in modified Eagle's medium containing 10% fetal bovine serum. Cells were plated in 24- well plates and grown to 80% confluence. The medium was replaced with 1 ml per well of fresh medium containing 125 μM of the R or S epimer of sulindac, and the cells were incubated in 5% C02 at 37°C for the duration specified. The inducing medium was then removed, the cells washed with PBS and 100 μl per well of fresh medium containing 200
μM of R or S sulindac was added for the assay. Incubation was carried out for 1 hour at
37°C after which the medium was then transferred to microcentrifuge tubes, three volumes of acetonitrile added and the samples analyzed by HPLC as described.
Previous experiments had shown that there was little sulindac sulfone retained by the cells, so it was not deemed necessary to analyze cellular contents.
30 3 RESULTS
3.1 Studies on the reducing requirements for the Msr enzymes
3.1.1 Reduced Trx is not an efficient reducing agent for hMsrB2 and hMsrB3
In the course of developing the DABS colorimetric assay for Msr activity (see
Materials and Methods), it was confirmed that eMsrA and eMsrB could use either DTT or Trx to supply the reducing power, with similar or more activity observed with Trx in vitro. However, although both hMsrB2 and hMsrB3 could use DTT as the reducing agent, these proteins showed very little activity with Trx. Table 5 compares the activity of several recombinant Msr proteins using either DTT or Trx. It can be seen that eMsrA, bovine MsrA (bMsrA) and eMsrB are active with either DTT or Trx as the reducing system. In fact, eMsrB was much more active with Trx than with DTT. In contrast, hMsrB2 and hMsrB3 work very poorly with Trx, having <10% of the activity seen with
DTT. One possibility is that the hMsrB proteins specifically required mammalian Trx and not the bacterial Trx that was used in these experiments. Therefore, hMsrB3, as well as eMsrA and eMsrB, were tested with mammalian Trx and mammalian Trx reductase.
Reduced mammalian Trx, like the bacterial Trx, gave very poor activity with hMsrB3, but both eMsrA and eMsrB efficiently used Trx from either source (data not shown). It should be noted that it was not possible to test mammalian MsrB1, which differs from
31 MsrB2 and MsrB3 in being a selenoprotein, because all attempts to obtain sufficient amounts of this recombinant protein were unsuccessful.
Table 5: Thioredoxin and DTT as reducing agents with various Msr proteins.
DABS-Met, nmol Msr Proteins Ratio Trx/DTT Trx DTT eMsrA 46.8 46.2 1.0 eMsrB 60.3 11.1 5.4 bMsrA 15.8 30.0 0.53 hMsrB2 0.8 31.3 0.03 hMsrB3 1.7 53.8 0.03 DABS-Met-S-(O) was used as substrate with MsrA proteins, and DABS-Met-R-(O) was
used as a substrate with MsrB proteins. The incubation conditions and assay are
described in Materials and Methods. The amounts of Msr protein used were as
follows: eMsrA, 1.6 μg; eMsrB, 2.7 μg; bMsrA, 2 μg; hMsrB2, 2.7 μg; hMsrB3, 2.3 μg.
The weak activity of hMsrB2 and hMsrB3 with Trx suggested that there may be another reducing system for these proteins in mammalian cells that either functions in place of Trx or is an intermediate hydrogen carrier between Trx and the human MsrB proteins.
3.1.2 Zn-MT in the presence of EDTA can serve as a reducing agent for Msr
In an attempt to search for a biological factor that was more efficient than Trx in supplying the reducing system for hMsrB2 and hMsrB3, an S-100 from bovine liver was initially tested. Using hMsrB3, significant reducing activity was detected in the liver S-
100 fraction, but only in the presence of EDTA. The active material was stable to heating at 80°C for 10 min. Figure 16 shows the effect of protein concentration of the
32 heated S-100 extract and the almost complete dependency on EDTA for hMsrB3 activity.
Optimal activity was seen with levels of EDTA >2.5 mM (data not shown). Routinely, 5 mM EDTA has been used in the experiments. EDTA by itself had no significant effect on hMsrB3 activity, although hMsrB3 contains zinc. Other chelating agents were tested in place of EDTA with Zn-MT. 1,10-Phenanthroline (5 mM) gave ≈40% of the activity of EDTA, whereas EGTA (5 or 20 mM), deferoxamine (5 mM), and zincon (500 μM) were inactive. Thus, EDTA was used throughout the present studies. A series of metal salts could not replace EDTA or the heated S-100 in the reaction (data not shown).
Figure 16: Effect of heated liver S-100 concentration and EDTA on MsrB3 activity.
Activity was measured without EDTA (○) or with 5 mM EDTA (●) in the reaction mix.
MsrB3 (2 μg) was incubated with the indicated amount of heated S-100 in the absence
of a reducing system (DTT or Trx) as described in Materials and Methods.
33 The heat stability and EDTA requirement suggested that the active factor might be a metallothionein (MT), and the heat-stable factor was further purified as described in
Materials and Methods. Figure 17A shows the elution profile from a DE-52 cellulose column, the last step in the purification. Two distinct peaks of reducing activity were observed, and the fractions in both peaks were active in the Msr assay using hMsrB3, but only in the presence of EDTA. The purification profile suggested that the two peaks correspond to MT-1 and MT-2, based on their elution from the DE-52 column.
34 Figure 17: Purification and properties of the active factor. A. Elution profile from a
DE-52 column of the factor showing Msr activity and zinc content. Two peaks of
reducing activity with hMsrB3 could be separated, and they are labeled MT-1 and MT-2.
Activity (●) is expressed as total nanomoles of DABS-Met formed per 1-ml fraction in
the Msr reaction using hMsrB3. Zinc concentration (μM) also is shown (○). B. Spectra
of purified factor at pH 7.4 (solid line) and pH 2.0 (dashed line). An extinction
−1 −1 coefficient of ε220 = 48,600 M ·cm at pH 2.0 was used to calculate the amount of MT
in the fractions.
Because of the requirement for EDTA for the fraction to be active with hMsrB3, metal analyses were initially performed on purified preparations by using inductively coupled plasma MS. Zinc was found in significant amounts [60,795 parts per billion
35 (ppb)], with trace levels of copper and silver (688 and 739 ppb, respectively). Besides using nanopure water, no special precautions were taken to remove trace metals, so the source of these trace metals in the protein sample is unknown. As shown in Figure 17A, the active, highly-purified fractions from the DE-52 column contained high levels of zinc that coeluted with the fractions active in the Msr assay. The amount of MT could be
−1 −1 determined spectrophotometrically (ε220 = 48,600 M ·cm at pH 2.0), and zinc analyses using the 4-(2-pyridylazo)resorcinol (PAR) reagent (see Materials and Methods) showed that there were close to seven zinc atoms per mole of MT in each fraction.
Although zinc appears to be the major metal associated with the active factor, the presence of lower levels of other metals in the sample cannot be eliminated. Figure 17B shows the UV spectrum of a fraction from peak 2 from the DE-52 column (both peaks displayed similar spectral characteristics). It can be seen that the active factor has a high absorbance in the 215- to 250-nm range but essentially no absorbance at 280 nm, indicating the absence of aromatic amino acids. Upon acidification, the high UV absorbance is markedly decreased. On SDS-PAGE, the purified protein, as well as a commercial rabbit liver MT preparation, migrated as a diffuse band in the 13- to 16-kDa range (data not shown), double the size of Zn-MT, which is ≈6 kDa. This gel migration pattern could be due to the unique shape of the protein or the presence of dimers through intermolecular bond formation. The liver MT obtained from a commercial source also supported hMsrB3 activity in the presence of EDTA (data not shown). The presence of zinc as well as the spectral and other characteristics of the active fractions indicated that the two peaks off the DE-52 column were Zn-MT-1 and Zn-MT-2. These peak fractions
36 were further analyzed by electrospray MS, and the molecular weights matched those of bovine MT-1 and MT-2 (5,987 and 6,013, respectively). EDTA removed >90% of the zinc from Zn-MT in <10 min., as measured by the appearance of free SH groups (data not shown). From this evidence, it was concluded that the purified factor is a Zn-MT, which, in the presence of EDTA, is converted to the metal-free reduced T, and that T, because of the high content of cysteine residues, is able to supply the reducing system for the Msr reaction. The results shown below used Zn-MT-2, although similar results were obtained with Zn-MT-1.
As seen in Table 6, the purified Zn-MT is not a specific reducing agent for hMsrB3 because it also supports eMsrA, eMsrB, and bMsrA, dependent on EDTA.
However, the liver factor showed very little activity with hMsrB2 under the conditions used in Table 6.
Table 6: Comparison of the activity of Msr proteins in the presence of Zn-MT or
DTT.
DABS-Met, nmol Msr proteins Zn-MT DTT eMsrA 33.9 45.7 eMsrB 8.3 27.8 bMsrA 14.1 38.9 hMsrB2 0.9 27.1 hMsrB3 18.0 53.7 Msr proteins were incubated as described in Materials and Methods with either 20
nmol of purified Zn-MT or 15 mM DTT. Incubations with Zn-MT routinely contained 5
mM EDTA, and no significant activity was detected in the absence of EDTA. The
amounts of proteins used were as follows: eMsrA, 1.6 μg; eMsrB, 5.4 μg; bMsrA, 2.0
μg; hMsrB2, 2.7 μg; hMsrB3, 2.3 μg.
37 3.1.3 T can function in the Msr system in the absence of EDTA
Although it appeared likely that the requirement for EDTA was to release zinc from Zn-MT to form T, this system was obviously artificial, and it was important to demonstrate directly that T could serve as the reducing agent for the Msr system. T was prepared as described in Materials and Methods and tested with hMsrB3 as shown in
Figure 18. It can be seen that hMsrB3 activity was supported by both T and Zn-MT, although T was active in the absence of EDTA, whereas Zn-MT required EDTA for activity. Shorter incubations were used for these experiments to minimize the oxidation of T that occurred at neutral pH. T also was active with MsrA in the absence of EDTA
(data not shown). These results support the view that the requirement for EDTA with
Zn-MT is to release the zinc from Zn-MT and form T and that T is able to provide the reducing system for the Msr enzymes.
38 Figure 18: T can supply the reducing system for hMsrB3 activity in the absence of
EDTA. The incubations contained 4.5 μg of MsrB3, 20 nmol of T, or 20 nmol of Zn-
MT. The incubations with T did not contain EDTA, but 5 mM EDTA was added to the
incubations with Zn-MT. At 20 min., Zn-MT in the absence of EDTA formed 1.3 nmol,
whereas T in the presence of 5 mM EDTA formed 23.5 nmol. ●, T; ○, Zn-MT plus
EDTA.
3.1.4 Trx can reduce T(O)
The reaction mechanism for both MsrA and MsrB involves the formation of an oxidized enzyme intermediate that must be reduced for the Msr protein to act catalytically10,11,16,38,43. If T is capable of reducing oxidized Msr, the T would become partly or fully oxidized to T(O), and, ideally, there should be an enzymatic system that could regenerate T and permit it to recycle. T(O) was prepared as described in
Materials and Methods. This material had generally lost ≈50–60% of its free SH groups but still remained mostly soluble (see Materials and Methods). Any insoluble material that was formed was removed by centrifugation. Trx was considered a possible
39 candidate to reduce T(O), which could be shown directly by measuring NADPH oxidation in the presence of the Trx reducing system and T(O). As shown in Figure 19, the oxidation of NADPH depended on Trx, Trx reductase and T(O). In addition, as shown in Table 7, T(O) could support hMsrB3 activity in the presence of the complete
Trx reducing system (row 1) but not in the absence of Trx, Trx reductase or NADPH
(rows 3–5). As discussed previously, the Trx system alone showed very low activity
(row 2). It is also apparent from the results in Table 7 that the free SH groups remaining in the T(O) cannot support the Msr reaction, indicating that the SH groups in T are not all equivalent with respect to their ability to function as a reducing agent for the Msr system. The results in Figure 19 and Table 7 indicate that disulfide bonds in T(O) can be reduced by the Trx system. Thus, Trx may be one of the cellular agents that can enable
T(O) to recycle and function as a metabolic reducing system.
In contrast to the results with hMsrB3, hMsrB2, which had low activity with either Trx or Zn-MT (see Table 6), was also not stimulated when both T(O) and the Trx reducing system were present (data not shown).
40 Figure 19: Reduction of T(O) by Trx. The preparation of T(O) and the incubation conditions are described in Materials and Methods. The oxidation of NADPH was followed at 340 nm. ■, complete system; □, minus Trx; ○, minus Trx reductase; ●, minus
T(O).
Table 7: Trx stimulates the activity of MsrB3 in the presence of T(O).
Trx No. T(O) Trx NADPH DABS-Met, nmol reductase 1 + + + + 23.4 2 - + + + 2.4 3 + - + + 0.5 4 + + - + 3.4 5 + + + - 3.9 The incubations contained hMsrB3 (2.3 μg) as described in Materials and Methods.
Where indicated by plus sign, 8.3 nmol of T(O), 10 μg of Trx, 2.4 μg of Trx reductase and 100 nmol of NADPH were added. In this experiment, with the same amount of hMsrB3 used, 52.7 nmol of DABS-Met was formed in the presence of 15 mM DTT.
41 3.1.5 Selenocystamine (SeCm) enhances activity with mammalian Msr enzymes
As described in the previous section, it was discovered that thionein (T) can function as a reducing system for hMsrB3 and Trx can reduce oxidized thionein (T(O)), permitting T to recycle. In previous studies on the oxidation and reduction of Zn-MT, it was shown that selenium compounds, such as selenocystamine (SeCm, Figure 20A), can markedly increase the release of zinc from Zn-MT or the uptake of zinc by T, depending on the oxidation state of the protein78. In the presence of a reducing agent such as GSH, the SeCm is reduced to selenocysteamine (SeCem, Figure 20B)79, which greatly accelerates the reduction of T(O) and the uptake of zinc to form Zn-MT78. Previous studies have also shown that reduced selenium compounds can function as reducing agents for lipid hydroperoxides80. One of the members of the Msr family, MsrB1, is a selenoprotein, although MsrB2 and MsrB3 do not contain selenium but are zinc proteins22. In the results below, the effect of SeCm and other selenium compounds on the activity of several Msr enzymes in the presence of either the Trx system or T is examined.
42 Figure 20: Structure of selenium compounds. A. SeCm. B. SeCem. C. selenocystine.
D. selenocysteine.
Table 8 summarizes the results using five members of the Msr family: eMsrA, eMsrB, bMsrA, hMsrB2, and hMsrB3. All of the reactions in Table 8 contained
NADPH and Trx reductase. As seen in Table 8, column 1, in the absence of E. coli Trx or SeCm, there was no Msr activity. In the presence of Trx (Table 8, column 2), good activity was seen with eMsrA, eMsrB and bMsrA, with low activity observed with hMsrB2 and hMsrB3, as discussed earlier (Table 5). In these experiments, E. coli Trx reductase was used; however, it was shown previously that mammalian Trx reductase is also not active with hMsrB2 and hMsrB372. Table 8, column 3 shows that the addition of
SeCm to the Trx reducing system has only a small effect on the activity of the E. coli enzymes, eMsrA and eMsrB (10–20% stimulation). These enzymes are known to use
43 the Trx system most efficiently72. In contrast, the activities of bMsrA and especially the two human MsrBs, hMsrB2 and hMsrB3, are markedly stimulated by the presence of
SeCm. The bMsrA activity increases more than 3-fold, and the activity of both hMsrB2 and hMsrB3 increase 30-50-fold in the presence of SeCm.
Table 8: Effect of SeCm on the activity of Msr proteins using the Trx reducing
system.
nmoles DABS-Met formed Enzyme No addition Trx Trx + SeCm SeCm eMsrA 1.9 15.3 18.6 12.2 eMsrB <1 22.5 24.2 5.6 bMsrA <1 13.1 45.9 41.2 hMsrB2 <1 2.3 65.3 43.3 hMsrB3 <1 1.0 55.7 62.3
Incubation conditions and assay are described under Materials and Methods. All of the
incubations contained Trx reductase (2.4 µg) and NADPH (1 mM). Where indicated,
Trx (10 µg) and/or SeCm (50 µM) were added. Enzyme amounts used were as follows:
1.6 µg of eMsrA (70 pmol); 2.7 µg of eMsrB (175 pmol); 3 µg of bMsrA (120 pmol);
2.0 µg of hMsrB2 (95 pmol); 2.2 µg of hMsrB3 (110 pmol). Incubations were for 20
minutes. SeCm, in the absence of either Trx reductase plus NADPH or an Msr enzyme,
was not active (data not shown).
3.1.6 SeCm is reduced by Trx reductase and can directly reduce the Msr enzymes
Until now, the only reducing agent that has given good activity with hMsrB2 has been DTT. However, it is evident from the results in Table 8, column 3, that the Trx reducing system can function with hMsrB2 in the presence of a selenol compound. To
44 determine whether hydrogen transfer to SeCm catalyzed by Trx reductase requires Trx,
Trx was omitted in the experiments shown in Table 8 column 4. As seen in column 4, all of the enzymes have significant activity, demonstrating that Trx reductase can transfer hydrogen directly from NADPH to SeCm to form SeCem. These results indicate that
SeCem can supply the reducing power for the Msr enzymes. It should be noted that
SeCm by itself had no activity in the absence of an Msr enzyme or a reducing system.
3.1.7 Selenocystine enhances MsrB activity but is not directly reduced by Trx
reductase
Other selenium compounds were tested for their ability to serve as reducing agents with MsrB2 and MsrB3. As shown in Table 9, selenocystine (Figure 20C), which can be reduced to selenocysteine (Figure 20D), can efficiently support the Msr activity in the presence of the complete Trx system. Some activity, although much lower, was also detected with sodium selenite in the presence of the Trx system. It should be noted that the concentration of selenite used was 25 µM, which was optimal. Higher concentrations of selenite inhibited the reaction. However, unlike SeCm, these selenocompounds could not accept hydrogens from E. coli Trx reductase in the absence of Trx (data not shown). This was previously reported for selenite81. Three other selenium compounds tested, sodium selenate, ebselen, and selenomethionine, were inactive under the same conditions.
45 Table 9: Effect of addition of selenite and selenocystine to hMsrB reactions
containing the Trx system.
DABS-Met formed, nmol Enzyme No addition Selenite Selenocystine hMsrB2 1.6 4 55.7 hMsrB3 1.7 10.3 36.8
Incubation conditions and assay using the Trx reducing system (NADPH, Trx reductase
and Trx) are as described under Materials and Methods and Table 8. Selenocystine (50
µM) and sodium selenite (25 µM) were used where indicated. Enzyme amounts used
were as follows: 2.2 µg of hMsrB3 and 1.6 µg of hMsrB2.
3.1.8 Thionein can reduce SeCm
As reported above, T can serve as a reducing agent for all of the Msr enzymes tested, with hMsrB2 showing the least activity. Table 10 shows the results of experiments using T in place of the Trx reducing system, with and without SeCm. In the presence of SeCm, there is also a marked stimulation of T activity with all of the Msr enzymes, including hMsrB2. Once again, the most striking effects are seen with bMsrA, hMsrB2, and hMsrB3. These results indicate that T, similar to Trx and Trx reductase, can reduce SeCm to SeCem, which can supply the reducing system for the Msr enzymes.
The activity with DTT, a commonly used in vitro reducing agent, is also shown in Table
10. The activity with all of the Msr enzymes in the presence of T and SeCm is similar to or better than that obtained with DTT.
46 Table 10: Effect of SeCm on the activity of Msr proteins using thionein (T) as the
reducing agent.
Dabs-Met formed, nmol Enzyme T T + SeCm DTT eMsrA 12.4 24.9 27.5 eMsrB 3.3 12.8 10.5 bMsrA 8.7 39.4 37.1 hMsrB2 1.4 35.6 17.7 hMsrB3 3.9 50.3 38.2
Incubation conditions and assay are described under Materials and Methods. In these
experiments, Trx, Trx reductase, and NADPH were omitted, and 5.6 nmol of T or 15
mM DTT were added where indicated. The amounts of bMsrA, hMsrB2, hMsrB3, and
SeCm used were the same as in Table 8 However, 3.2 µg of eMsrA and 5.4 µg of eMsrB
were used. Incubations were for 20 minutes.
Figure 21 shows the effect of SeCm concentration on the activity of hMsrB3 using either the Trx reducing system or T as the reducing agent. Under the assay conditions, 50 µM SeCm appears to be the saturating concentration for T, whereas for the Trx-dependent reaction, the maximal activity requires more than 150 µM SeCm.
47 Figure 21: Effect of SeCm concentration on the activity of hMsrB3 using either the
Trx reducing system (●–●) or T (■–■) as the reducing agent. hMsrB3 was incubated
with the Trx reducing system (Trx, Trx reductase, and NADPH, see Materials and
Methods and Table 8) or with 5.6 nmol of T as the reducing agent (see Materials and
Methods and Table 10). Amounts of enzyme used were 2.2 µg of hMsrB3 with T and
1.1 µg with the Trx system. SeCm was added at the concentrations specified. A 20-min
incubation was used.
Figure 22 shows a time curve for hMsrB3 activity using the Trx reducing system in the presence or absence of 50 µM SeCm. The reaction is close to linear for up to 20 min and is dependent on SeCm.
48 Figure 22: Time curve of hMsrB3 activity using the Trx reducing system either with
(●–●) or without (■–■) 50 µM SeCm. Details of the incubation are described under
Materials and Methods.
From the preceding sections, it can be seen that thioredoxin, although an excellent reducing system for E. coli MsrA and MsrB and bovine MsrA, is not an efficient reducing agent for either hMsrB2 or hMsrB3. However, thionein and certain selenium compounds, such as selenocystamine and selenocystine, can function as reducing agents for most of the Msr enzymes, including the MsrB enzymes. The in vivo relevance of this is not known but will be addressed further in the Discussion section.
49 3.2 A high-throughput screening assay for MsrA
As mentioned in the Introduction, since oxidative damage is thought to be involved in many diseases, especially age-related diseases, any compound that would increase the activity of MsrA in tissues could have important therapeutic value. Genetic experiments have indicated that increasing MsrA activity in vivo may prolong life span.
Thus, it would be advantageous to develop an assay that would permit us to use high throughput screening (HTS) of chemical libraries to find activators of this enzyme, as well as specific inhibitors that could be useful research tools as described below. Any compound found in the HTS that would increase MsrA activity could have important implications for intervention in various diseases.
3.2.1 Development of the assay
As discussed earlier, MsrA has activity towards a broad range of substrates including the S-epimer of Met(O) in protein linkage, free Met(O) and the S epimer of other compounds that have a methyl sulfoxide moiety, including the drug sulindac47 and a compound as simple as DMSO46. In all cases the sulfoxide moiety is reduced to a sulfide. Assays that involve extraction procedures or HPLC separation are not practical for use in HTS. However, for all substrates it is possible to follow the oxidation of
NADPH at 340 nm when Trx is used as the reducing system46. Thus, NADPH oxidation appears to be the most logical path for developing an HTS assay for MsrA activity (see
Figure 8).
50 An initial problem in developing an HTS assay was the realization that the chemicals (at 1-10 mM concentration) in the libraries were all dissolved in DMSO, a known substrate for MsrA46. However, the ability of MsrA to reduce DMSO has made it possible to develop a spectrophotometric assay that is amenable to HTS. It is anticipated that an 80 μl incubation will be used containing between 100-300 nl of each library compound. At these concentrations the incubations would contain between 16-50 mM
DMSO which would be an overwhelming competitive substrate for all of the substrates that have been routinely used to assay MsrA activity. However, in view of this significant problem, it was realized that there is no reason the DMSO in which the chemicals are dissolved could not be used as a substrate for the enzyme and the activity assayed by measuring the oxidation of NADPH using Trx as the reducing system, as summarized in Figure 8. Because of the high concentration of DMSO used, one would not expect to see competitive inhibitors showing up in the HTS screen.
Our preliminary experiments to develop an assay for MsrA activity used a 500 μl incubation volume and DMSO as substrate since, as mentioned, the compounds to be tested are dissolved in DMSO. Figure 23 shows a time course of the reaction using varying amounts of MsrA that result in up to 60% oxidation of the NADPH in 30 minutes. The assay was highly reproducible under these conditions. To demonstrate that the assay could be used to identify both activators and inhibitors of the enzyme, a known activator, selenocystamine (SeCm), and a known inhibitor, N-ethylmaleimide (NEM), were tested.
51
Figure 23: The Effect of MsrA concentration on the oxidation of NADPH. Each
incubation (500 μl) contained 100 mM Tris pH 7.4, 2 μl DMSO (56 mM), 0.3 mM
NADPH, 10 μg E. coli Trx, and 2.3 μg E. coli Trx reductase. Various amounts of
purified bovine MsrA were added, and the reaction was carried out at room temperature
for 30 minutes, and the course of the reaction was assayed by the change of OD at
340nm. Although not shown, the amounts used of each of the components had been
optimized.
As shown above, selenium compounds, such as SeCm (Figure 20A), can stimulate the activity of MsrA indirectly by increasing the rate at which the enzyme is reduced during the catalytic reaction82. Figure 24 shows the effect of varying concentrations of SeCm on the activity of MsrA, using a low level of MsrA in the standard assay. It can be seen that the enzyme activity can be increased close to 3 fold in the presence of 20 μM SeCm in a concentration-dependent manner. SeCm is known to be reduced by the Trx system82. As described in the legend of Figure 24, the SeCm was preincubated with all of the components in the reaction mixture, except MsrA, for 5
52 minutes to adjust for the rapid initial reduction of SeCm by the Trx system in the absence of MsrA. The oxidation of NADPH was then measured after the addition of
MsrA, and a correction was also made for the slow reduction of SeCm observed during the 30 minute incubation in the absence of MsrA.
Figure 24: The stimulatory effect of selenocystamine (SeCm) on the activity of
MsrA. The assay conditions are as described in the legend for Figure 23. 1.5 μg of
MsrA were added to each incubation, and, where indicated, various concentrations of
SeCm. Because SeCm causes an initial rapid oxidation of NADPH, followed by
oxidation at a slower rate, readings were started after 5 minutes of incubation and
continued for another 30 minutes. The zero time values reflect the drop in absorbance at
340 nm due to the initial oxidation of NADPH during the 5 minute preincubation. The
values in the figure are also corrected for the low level of NADPH oxidation due to the
reduction of SeCm in the absence of MsrA during the 30 minute incubation.
Previous studies in eukaryotic systems have shown that a knockout of msrA results in increased ROS levels28,46 and shortened life span33, and in the case of some bacteria, decreased adherence and virulence83,84. Thus, instead of relying on genetic
53 mutations to eliminate the enzyme activity, it would also be of great value to obtain a chemical inhibitor of the enzyme. Clearly a specific inhibitor would eliminate the need for generating knockouts and permit one to modulate the level of enzyme activity. This would be especially useful in cell culture systems. Although the primary goal of this proposal is developing an HTS method to look for activators of MsrA, the assay as described could be modified to detect inhibitors of MsrA. As an example, in Figure 25
NEM was used since this compound inhibits the enzyme by reacting with the SH groups on the enzyme. A higher concentration of MsrA was used in these experiments so that the inhibition could be easily detected. In addition, as described in the legend to Figure
25, the NEM was preincubated with MsrA alone for 5 minutes, and the NEM was then destroyed by the addition of 1 mM mercaptoethanol, prior to the addition of the other incubation components. As shown in Figure 25, the inhibition observed was concentration dependent.
54 Figure 25: The inhibitory effect of N-ethylmaleimide (NEM) on the activity of
MsrA. Various concentrations of NEM were incubated at 37°C with 22.5 μg of MsrA in
a total volume of 30 μl. After 5 minutes, 1 mM mercaptoethanol was added to destroy
the NEM and 15 μl of the NEM-treated MsrA incubations were added to to a standard
500 μl of the reaction mixture as described in the legend to Figure 23. The course of the
reaction was followed at 340 nm.
3.2.2 Miniaturization of the assay
Having determined the optimum assay conditions at the level of a 500 μl incubation, the next step was to miniaturize the assay for HTS and determine whether it could be adapted for use in an 80 μl volume using a 384-well plate. The experiments shown in Figures 23-25 were essentially repeated, in a modified fashion, using a 384- well plate and an 80 μl incubation. Figure 26 shows the results of these studies. Each time point on the curves represents the mean of 5 replicate experiments. The decrease in
OD at 340 nm in the presence of DMSO and MsrA in these 80 μl incubations and the
55 stimulation by SeCm as well as the inhibition by NEM closely parallels the results obtained with the larger incubations and indicates that it is a valid assay at these smaller volumes. The Z factor is a useful measurement to assess the statistical validity of the assay85, and the values have been calculated for the SeCm and NEM experiments (n = 5), and they are 0.95 (S.D. 0.003) and 0.92 (S.D. 0.007), respectively. These values were calculated at the 20 minute time point, the time that would likely be used for the HTS, although there was little variation over the course of the experiment. These Z factors indicate that the assay is robust and reproducible.
56 Figure 26: Effect of SeCm and N-ethylmaleimide (NEM) on the reduction of DMSO
by MsrA in an 80 μl assay format. Each reaction contained 50 mM Tris pH 7.4,
25mM DMSO, 4 μg Trx, 0.5 μg TrxB, 40 nmol NADPH, 4 μg MsrA and 20 μM
selenocystamine or 30 μM NEM, where indicated. For these incubations, MsrA was
pretreated with NEM for 5 minutes and then 1 mM mercaptoethanol was added to
destroy the NEM. An aliquot was removed which contained 4 μg MsrA and added to the
reaction mixture lacking MsrA. The reactions were carried out in 384-well black, clear-
bottom plates and read in a Molecular Devices Spectramax spectrophotometer at 340nm.
In addition to the absorbance assay described above, there is also a fluorescence assay for NADPH. The fluorescence assay has been successfully used in an HTS format to screen for inhibitors of the Schistosoma mansoni redox cascade86. Because NADPH is naturally fluorescent, emitting at 450 nm, while NADP is not, it would be relatively easy to switch to this type of assay. Although, at present, there are no problems anticipated with the absorbance assay that cannot be controlled for (see section below on potential pitfalls), but in the event that there are problems, optimized conditions for a
57 fluorescence-based NADPH assay have also been developed. Figure 27 shows the results of experiments using fluorescence to assay for the change in NADPH concentration as a function of the stimulation or inhibition by SeCm or NEM, respectively. As can be seen, there is a significant stimulation by SeCm which closely parallels the absorbance assay. The calculated Z factor for this assay at 20 minutes of incubation is 0.90.
Figure 27: Effect of SeCm and NEM on the reduction of DMSO using a
fluorescence assay. The conditions are identical to those used in Figure 26 except that
fluorescence was assayed instead of absorbance. The fluorescence emitted was followed
at 450 nm after excitation at 365 nm using the Spectramax.
The MLSCN compound library at the Scripps Florida μHTS facility will be screened in collaboration with Dr. Peter Hodder. Since the equipment at the μHTS
58 laboratory utilizes 1536-well plates, the validity of the assay will need to be ascertained in that format as well.
With regard to the amount of proteins needed, a screen of 100,000 compounds in an 80 μl reaction volume would require 30-60 mg of bovine MsrA, 150-300 mg of E. coli Trx and 30-60 mg of E. coli Trx reductase. These are recombinant proteins which have been routinely expressed in E. coli and purified in our laboratory in smaller quantities. However, because of the need for such large quantities, arrangements have been made with Dr. W. Todd Lowther of Wake Forest University, who has access to fermentors, to supply us with the required amounts of the purified proteins.
3.2.3 Follow up of active compounds
It cannot be estimated a priori what percentage of library compounds will affect the enzyme activity, but any active compounds that appear to be specific for MsrA (more than twofold stimulation or greater than 30% inhibition) will be verified in the following ways. Initially, the compound will have to be obtained as a powder or in a solution that does not contain DMSO. The compound will then be checked for activity using any of the assays that have routinely been used to measure MsrA activity. The most convenient assay uses DABS-Met(O) and measures the formation of DABS-Met either colorimetrically72 or by HPLC27. Once the activity is confirmed with purified enzymes using a standard substrate, cells in culture will be tested to see whether the compound can increase MsrA activity in whole cells, assuming that the compound can enter cells.
MsrA activity in cell culture is based on DABS-Met(O) reduction using an HPLC assay.
59 Based on what has been previously reported, increased activity of MsrA should also protect cells against oxidative damage. These are experiments that are routinely performed in our laboratory. In general, the compound to be tested is incubated with the cells in culture for 1-2 hours and then the viability of the cells is determined after a 2- hour exposure to either hydrogen peroxide or tert-butyl hydroperoxide. Viability would be measured using a tetrazolium salt (MTS) that measures mitochondrial NADH oxidase activity. Additional controls would be used to insure that the compound being tested does not interfere spectrophotometrically with MTS and also that it does not directly affect reduction of MTS. For the cell culture experiments, a particular interest is the protection of cardiac myocytes against oxidative damage due to hypoxia and reperfusion. As described in the Introduction (Figure 5), over expression of MsrA in cardiac myocytes protects them against hypoxia/reperfusion oxidative damage32.
Any compounds active in the cell culture systems will then be administered in vivo to laboratory animals, and the level of MsrA activity will be determined in tissues using our standard assay. The in vivo systems that are of most interest to us at this time would be testing Drosophila for increased resistance to paraquat and extended life span, as described in our earlier studies35. Another in vivo system involves the whole heart system (Langendorff model) in which the heart is exposed, for a short period of time, to ischemia followed by reperfusion. Under these conditions the heart undergoes apoptotic death due to oxidative damage that can be assayed by measuring the release of lactic dehydrogenase. In the in vivo models, the animals would either be injected intraperitoneally or fed the active compound 2-4 days before the heart is removed for the
60 Langendorff procedure. Dr. Howard Prentice, a faculty member at FAU, is available to collaborate with us on this type of experiment.
3.2.4 Optimization of any lead compounds
Any compounds that appear to be specific activators of MsrA will almost certainly have to be chemically modified to obtain derivatives with higher activity, better permeability, longer in vivo half life and less toxicity. Analyses of the structures of the active compounds arising from the HTS might provide platform structures that could be the starting point for chemical modification. For these studies, the expertise of chemists with a background in organic syntheses will be required. At FAU, Dr. Salvatore Lepore, who has extensive industrial and academic experience as a synthetic organic chemist, has agreed to provide assistance at this stage of the project. In addition, there is a strong chemical group at Scripps Florida who will also be available for custom synthesis.
3.2.5 Problems using a coupled system to measure NADPH oxidation
Although saturating amounts of both Trx and Trx reductase are being used, the assay involves a coupled system in which reduced Trx is generated using NADPH and
Trx reductase. Thus, any active compounds that appeared in the screen would have to be tested to make sure they were not affecting the activity of the Trx system. This is easily done by setting up a Trx reductase assay without MsrA, in which substrate levels of oxidized Trx are used so that the oxidation of NADPH is dependent only on the
61 reduction of the Trx. Any increase in NADPH oxidation would indicate that the compound was stimulating the Trx reaction and not the MsrA activity. Alternatively, any active compound would be obtained as a powder or DMSO-free solution and assayed in a different system (see section below) which uses a different substrate and dithiothreitol
(DTT) as the reductant.
3.2.6 Absorbance at 340 nm of chemicals in the library
An additional potential assay problem would occur if the chemical in the library had significant absorbance at 340 nm. This would appear as an increase in the absorbance of the reaction mixture at zero time, for which a correction could easily be made87. Alternatively the fluorescence assay, as shown above, could be used in cases where the absorbance of the compound poses a significant problem.
3.2.7 Elimination of non-specific activators or inhibitors
There are no known direct activators of MsrA reported in the literature. The only activators of the system that have been reported have involved the reducing requirement for the enzyme, including certain selenium compounds and thionein (see Figure 24)72,82.
It would therefore be surprising to see a large number of non-specific activators of
MsrA. The first test would be to see whether the compounds were activators of other members of the Msr family such as MsrB2 and MsrB3. These enzymes have different three-dimensional structures and substrate specificities than MsrA and do not use Trx
62 efficiently as the reducing system. There are highly-sensitive assays to measure these activities that do not involve NADPH oxidation but instead measure the reduction of
DABS-Met(O) containing the R epimer of methionine sulfoxide using a colorimetric procedure72.
However, one would expect that there may be many non-specific inhibitors seen in the HTS assay for MsrA since the reaction mechanism involves cysteine residues on the protein. Any sulfhydryl-reacting compound, like NEM, would inhibit MsrA as well as other enzymes having a cysteine residue at the active site. Thus, any inhibitor would be tested for its ability to react with the free SH groups on the enzyme. There is a simple colorimetric assay to measure SH groups on proteins88 that has been used in previous studies72.
In summary, a coupled system has been developed to measure the activity of
MsrA based on the oxidation of NADPH using DMSO as the substrate for the enzyme.
The assay as described can be used to measure activators of the enzyme as well as inhibitors.
63 3.3 The metabolism of sulindac
As discussed in the introduction, sulindac may have an important therapeutic use in the treatment of cancer and possibly in the protection of normal tissues from oxidative stress. As noted, sulindac is a mixture of the R and S epimers, and there is little information on the biological activity or metabolism of the individual epimers.
Preliminary studies had shown that MsrA can reduce only the S epimer of sulindac, and no pathway is known for reduction of the R epimer. To examine this directly, the sulindac epimers were separated in order to study their metabolism individually (see
Figure 28).
Figure 28: Separation of the sulindac epimers using a chiral column. See Materials
and Methods for specifics of the separation.
3.3.1 Sulindac metabolites detected in rat tissues
It is known that sulindac can be reduced to sulindac sulfide and oxidized to sulindac sulfone in vivo56. As a first step in establishing the in vivo metabolic pathways
64 for sulindac, rats were given the individual sulindac epimers by intraperitoneal injection.
After four hours, the sulindac metabolites found in plasma were analyzed by HPLC, and a typical result is shown in Figure 29. Regardless of which epimer was given, sulindac, sulindac sulfone and sulindac sulfide were all seen at detectable levels. In several experiments it was generally observed that the sulindac S epimer yielded more sulfide and less sulfone than the R epimer.
Figure 29: Chromatogram showing metabolites of R and S epimers of sulindac in
plasma of rats given IP injections of sulindac. Sulindac appears at 1.1 minutes,
sulindac sulfone at 1.5 minutes and sulindac sulfide at 7.1 minutes.
3.3.2 MsrA knockout mice can reduce both sulindac epimers in vivo
As noted above, the only enzyme known to reduce the S epimer of sulindac to sulindac sulfide is MsrA, and no enzyme has been described that reduces the R epimer.
To determine the role of MsrA in catalyzing the reduction of the S epimer of sulindac,
65 the in vivo metabolism of the sulindac epimers was investigated in MsrA knockout mice.
MsrA knockout mice were given 0.2 mg of either sulindac-R or sulindac-S by IP injection four hours prior to sacrifice, and plasma samples were analyzed (see Materials and Methods). As shown in Table 11, sulindac, sulindac sulfone and sulindac sulfide were found in the plasma of all animals, similar to the results with rats. Plasma levels of sulindac sulfide in rats injected with the S epimer are significantly lower (about one fourth), but not zero, in knockout mice compared to wild type. It was concluded from these results that there is another enzyme besides MsrA capable of reducing the S epimer in tissues, or the reduction was carried out by intestinal flora.
Table 11: Metabolites of sulindac found in plasma of MsrA knockout mice following
IP injection.
Knockout Wild type sulindac-R sulindac-S sulindac-R sulindac-S (nmol/ml) Sulfoxide 3.7 2.7 6.4 5.1 Sulfone 13.5 5.6 11.3 6.2 Sulfide 4.4 2.9 11.1 10.7
Mice were given 0.2 mg of either sulindac-R or sulindac-S by IP injection four hours
prior to sacrifice, and plasma samples were analyzed as described in Materials and
Methods.
3.3.3 Liver homogenates from MsrA knockout mice do not reduce sulindac-S
A previous study on the metabolism of sulindac indicated that most of the metabolism occurred in the liver, with reduction to the sulfide primarily in the cytosol and oxidation to the sulfone in the microsomes89. To further characterize reduction of
66 the sulindac epimers in the MsrA knockout mice, the soluble cytosolic fraction (S-100, see Materials and Methods) was isolated from livers of both MsrA knockout and wild type mice and tested for its ability to reduce each of the sulindac epimers. The S-100 was incubated with the epimers, with or without a reducing system (either DTT or the thioredoxin system). As seen in Table 12, there was no reduction of either epimer in the absence of a reducing system. In contrast to the results from the in vivo experiments
(see Table 11), the MsrA knockout mice showed almost no reduction of the S epimer with either reducing system. However, both MsrA knockout and wild type extracts showed reduction of the R epimer with either DTT or thioredoxin, and the wild type mice showed reduction of the S epimer with either reducing system. This suggests that the major sulindac-S reductase in liver is MsrA.
67 Table 12: Activity of mouse liver S-100 with sulindac-R and sulindac-S as
substrates.
MsrA Knockout Wild type sulindac-R sulindac-S sulindac-R sulindac-S nmol sulindac sulfide per reaction No reducing system 0.1 0.1 0.1 0.1 DTT 14.7 0.6 21.4 22.7 Thioredoxin 2.9 0.15 6.9 18.2
450 μg of S-100 was incubated with 50 nmol of the indicated sulindac epimer, 0.1 M
Tris pH 7.4 and either 15 mM DTT or a thioredoxin reducing system in a total reaction
volume of 100 μl. The thioredoxin reducing system consisted of 5 μg thioredoxin, 1.1
μg thioredoxin reductase and 1 mM NADPH.
3.3.4 Partial purification of sulindac-R reductase from rat liver
The reductase for sulindac-R was of particular interest since there is presently no known enzyme which catalyzes this reaction. An earlier study suggested that aldehyde oxidase may reduce the mixed sulindac epimers in vitro90. There is no evidence that this occurs in vivo and did not stimulate sulindac-R reduction under our conditions (data not shown). Several different MsrB enzymes, including human MsrB2 and MsrB3, which are known to catalyze the reduction of protein-bound methionine-R-sulfoxide, were tested and found to have no activity with the R epimer (data not shown). The sulindac-R reductase activity has been partially purified from rat liver S-100 as described in
Materials and Methods.
The first separation, by diethylaminoethyl ion-exchange chromatography (Figure
30), shows one major peak of R-reductase activity. A second, minor peak can be seen in
68 the salt wash at the end of the separation, but this material was not pursued further. The fractions corresponding to the main peak eluting at approximately 150 mM KCl were pooled and concentrated for application to a G50 gel-filtration column.
Figure 30: DEAE separation of sulindac-R reductase activity. See Materials and
Methods for details of the separation.
There are two peaks of sulindac-R reductase activity eluting from the G50 column, one at a high molecular weight in the exclusion range (> 50 kDa) and one at a low molecular weight (see Figure 31). This column has been well characterized with other proteins of known molecular weight, and the elution volume of the low molecular weight active material corresponds to a molecular weight in the range of 10 to 15 kDa.
Since the vast majority of protein elutes in earlier fractions at higher molecular weights,
69 this separation provides a high degree of purification (about 25-fold). The activity appearing at higher molecular weights was not pursued further at this time.
3 1.80 2.5 1.50 e d i f 2 1.20 l u s
c 0 a 8 1.5 0.90 d 2 Activity n A i l
u A280
1 0.60 s
l o m
0.5 0.30 n 0 0.00 16 18 20 22 24 26 28 30 32 Fraction
Figure 31: G50 separation of sulindac-R reductase activity. See Materials and
Methods for specifics of the separation.
The third purification step is a small (1 ml) quaternary ammonium, strong ion- exchange column. This step provides an additional twofold purification, and the active material elutes in a single peak (Figure 32).
70 0.035 3.0
0.030 2.5
0.025 e d i f
2.0 l u s 0.020 c a 0 1.5 d 8 n 2 i Activity 0.015 l A u
s A280
1.0 l 0.010 o m n 0.005 0.5
0.000 0.0 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Fraction
Figure 32: Quaternary ammonium separation of sulindac-R reductase activity. See
Materials and Methods for specifics of the separation.
A summary of a typical purification is shown in Table 13. An overall purification of 200-fold with a recovery of 5% has been obtained.
Table 13: Purification of sulindac-R reductase.
Step Total Total Protein Specific Purification Recovery
Activity (mg) Activity Factor S-100 2240 248 9 1 100% Ammonium sulfate 30-70% 1989 147 13.5 1.5 89% DEAE 730 23 31.5 3.5 33% G50 293 0.37 787 87 13% Quaternary ammonium 110 0.063 1735 200 5%
71 3.3.5 Reducing requirements and substrate specificity of the purified material
Since the active material reduces the R epimer of sulindac, this suggested an
MsrB-like enzyme, perhaps MsrB1. However, as discussed in the Introduction, there are no known enzymes in mammals which efficiently reduce free Met(O)-R, and, given its structure, one could consider sulindac to be more similar to free Met(O) than protein- bound Met(O). Also, DTT, which is not a physiological reductant, has been used as the reducing system throughout the purification. In order to further characterize the purified protein, it was tested using both sulindac-R and DABS-Met(O)-R as substrates and with both DTT and Trx as the reducing system. The results are shown in Table 14. Activity with DABS-Met(O)-R is about 2.6-fold higher than with sulindac-R. Also, Trx cannot act as a reducing system for the enzyme, regardless of which substrate is used. The in vivo reducing system for this activity remains unknown at this point.
Table 14: Relative activity purified protein: reducing system requirement and
activity with DABS-Met-(O)-R and sulindac-R.
nmol sulfide Substrate DTT Trx Dabs-Met(O)-R 2.33 0.08 sulindac-R 0.89 0
If the same enzyme is reducing both of these substrates, the ability to reduce
DABS-Met(O)-R, a substrate that has been used as an analog of peptide-bound Met(O) to assay the MsrB enzymes, is a further indication that this enzyme may be a member of the Msr family.
72 3.3.6 Evidence that the sulindac-R and DABS-Met(O)-R reductases are the same
enzyme
Initially, to determine if there was one or multiple sulfoxide reductase enzymes in the purified material, the active fractions throughout the purification were assayed using both DABS-Met(O)-R and sulindac-R as substrates. If there is only one enzyme, then the level of purification at each step in the procedure should be similar for both substrates. As shown in Table 15, the degree of purification at each step, from ammonium sulfate precipitation through to the final quaternary ammonium step, is fairly close for the two substrates.
Table 15: Purification summary for the reduction of DABS-Met(O)-R and sulindac-
R.
Purification factor DABS-Met(O)-R sulindac-R S-100 1 1 Ammonium sulfate 3.6 3.9 DEAE 2.5 2.2 G50 20.8 19.4 Quaternary Ammonium 1.6 1.7 Total 289 293
DTT was used as the reducing system throughout this assay.
To further examine whether a single enzyme may be responsible for the reduction of both DABS-Met(O)-R and sulindac-R, and to see if free Met(O)-R is a substrate, a competition experiment was performed. If the same enzyme is responsible for the reduction of both substrates, then it should be possible to compete the activity with one
73 substrate by adding an excess of another substrate. This study was performed with sulindac-R as the primary substrate, using 10-fold excess of DABS-Met(O)-R and
Met(O)-R as competing substrates. Figure 33 shows the results of this study. Under these conditions, DABS-Met(O)-R and Met(O)-R inhibit the sulindac-R activity by 89% and 77%, respectively. These data suggest that Met(O)-R is also a substrate and that there is only one enzyme responsible for the reduction of sulindac-R, DABS-Met(O)-R and Met(O)-R.
400 350 300 e d i f
l 250 u s
c 200 a d n i
l 150 u s
l
o 100 m p 50 0 10X DABS-R Alone 10X Met(O)-R
Figure 33: Competition of sulindac-R activity with DABS-Met(O)-R and Met(O)-R.
DTT was used as the reducing agent.
3.3.7 Evidence that MsrB1 may not be the active component that reduces
sulindac-R
Given that the sulindac-R reductase activity elutes in the 10-15 kDa. range from the sizing column, it was initially suspected that it might be MsrB1, which is 12.8 kDa.
However, the results discussed below indicate that MsrB1 is not the active factor.
74 Analysis of purified protein by mass spectrometry
To try to identify the active component, the peak fraction from the quaternary ammonium column was placed on an SDS-PAGE gel, and a prominent band was noted at approximately 15 kDa. This band was excised and then analyzed by LC/MS/MS on an LTQ ion-trap mass spectrometer (University of Massachusetts). The result of this analysis showed that the most abundant proteins were fatty-acid binding proteins, hemoglobin and cystatin-B, none of which are Msr-like enzymes or even oxioreductases.
The MS analysis indicates that the major band seen on the gel does not correspond to the sulindac-R reductase activity and is not MsrB1
Studies with a cysteine analog of MsrB1
As mentioned earlier, because of the selenocysteine at the active site of mammalian MsrB1, it has not been possible to over express significant amounts of this enzyme in E. coli. However, a mutant which replaces selenocysteine with cysteine was previously prepared by Kryukov et al.23 and kindly supplied by Dr. V. Gladyshev. This recombinantly-expressed enzyme was used to test for activity with sulindac-R and
DABS-Met(O)-R. As shown in Table 16, although this enzyme showed activity with
DABS-Met(O)-R, using either DTT or Trx as the reducing system, no activity was seen with sulindac-R.
75 Table 16: Reductase activity of MsrB1 cysteine analog.
DABS-Met(O)-R Sulindac-R Reducing system nmol sulfide DTT 3.7 0.1 Trx 3.4 0
24 μg of enzyme were used per reaction.
The results suggest that the active reductase has MsrB-like activity but is not
MsrB1, MsrB2 or MsrB3.
3.3.8 Oxidation of the sulindac epimers by rat liver microsomes
In mammals, most xenobiotics are oxidized by the cytochrome P450 enzymes, which are found predominantly in liver microsomes. The ability of rat liver microsomes to oxidize both the R and S epimers of sulindac was therefore examined. As seen in
Figure 34, rat liver microsomes, under uninduced conditions, catalyzed the oxidation of both R and S epimers, dependent on NADPH. The reaction was dependent on microsome concentration and no activity was seen in the absence of NADPH. Under these conditions the S epimer is oxidized to the sulfone at a rate roughly twice that of the
R epimer in this system.
76 Figure 34: Sulindac sulfone formation by rat liver microsomes. Each reaction (60
minutes at 37°C) included 50 nmol sulindac, 1.6 mM NADPH, 3.3 mM glucose-6-
phosphate, 15 ng glucose-6-phosphate dehydrogenase, 3.3 mM magnesium chloride, 100
mM potassium phosphate buffer and the indicated amount of microsomes in a total
volume of 100 μl.
3.3.9 Sulindac oxidation is induced by the sulindac epimers in human hepatocytes
Previous studies60 showed that sulindac could induce several of the P450 enzymes that were regulated by the aryl hydrocarbon receptor. These authors used ethoxyresorufin-O-deethylase (EROD) as substrate but did not investigate the metabolism of sulindac under their conditions. In order to see whether sulindac induced
P450 enzymes involved in its metabolism, human HepG2 cells were treated for periods ranging from 2 to 24 hours with either sulindac-R or sulindac-S. After this pretreatment
77 period, the medium containing sulindac was washed out and replaced with fresh R or S epimer for a one-hour incubation period (see Materials and Methods). After this second incubation, metabolites were analyzed by HPLC. As shown in Figure 35A, pretreatment with the R epimer induces P450 enzymes that can oxidize the R epimer to the sulfone, but not the S epimer. However, pretreatment of the cells with the S epimer results in a much larger induction of P450 enzymes that induce oxidation of both the R and S epimers (Figure 35B). It should be noted that the R epimer is oxidized at a faster rate than the S epimer, regardless of which epimer is used in the pretreatment to induce the P450 enzymes.
Figure 35: Induction of sulindac oxidation in HepG2 human hepatocytes. A.
Induction by pretreatment with sulindac-R. B. Induction by pretreatment with sulindac-
S. Cells were pretreated for the indicated time with medium containing 125 μM of the
inducing epimer. Medium was then removed, cells were washed with PBS, and 100 μl
of fresh medium containing 200 μM of R or S sulindac was added per well. Incubation
continued for one hour at 37°C. Analysis of metabolites was as described in Materials
and Methods.
78 3.3.10 Studies with purified cytochrome P450 enzymes
In order to determine which specific P450 enzymes were responsible for the metabolism of the sulindac epimers, 11 rat or human purified recombinant P450 enzymes were incubated with each epimer and an NADPH regenerating system (see
Materials and Methods), and the metabolites were analyzed by HPLC. As seen in
Table 17, most enzymes tested showed some activity with both sulindac epimers. The primary P450 enzymes responsible for R epimer oxidation were CYP1A1, 1A2, 1B1 and
3A4, which are under control of the aryl hydrocarbon receptor. For oxidation of the S epimer, the major enzymes are 2C8 and 3A4. The only enzyme which showed a preference for the S substrate over the R substrate was 2C8.
Table 17: Sulindac oxidation by cytochrome P450 enzymes.
sulindac-R sulindac-S Rat 1A1 470 39 Rat 1A2 513 41 Human 1B1 128 9 Rat 2B1 15 17 Rat 3A2 42 43 Human 3A4 121 93 Human 1A2 170 76 Human 2D6 64 42 Human 2C19 41 0 Human 2C8 38 60 Human 2C9 19 1.1
Values are given in pmol sulindac sulfone per pmol enzyme per hour.
79 4 DISCUSSION
4.1 Thionein and selenocompounds as reducing agents for the Msr enzymes
Until the present studies, it had been assumed that Trx was the biological reducing system in cells for all of the Msr proteins. The initial experiments using eMsrA indicated that Trx was the biological reducing agent6,91, in agreement with earlier experiments42. In those experiments, it was shown that Met(O) could support growth of a Met-requiring strain but not if the organism was also Trx deficient, indicating that Trx is necessary for the conversion of free Met(O) to Met in E. coli. The present experiments are in agreement with these earlier results. It appears that MsrA from both bacterial and mammalian sources utilizes Trx very efficiently, as does MsrB from E. coli.
However, the studies reported here show that hMsrB2 and hMsrB3 (and, presumably,
MsrB proteins from other mammalian sources) use Trx very poorly. As noted above, hMsrB1 was not tested in the present studies because of an inability to over express and purify the protein from E. coli, but Trx may work well with this protein.
Kim and Gladyshev43 postulated that, in MsrB1, a cysteine was required in addition to a selenocysteine for Trx to function. In contrast, with MsrB2 and MsrB3, only the active-site cysteine was required, and Trx was thought to directly reduce the sulfenic acid intermediate on the enzyme43. It seems clear from the low activity using
Trx with both MsrB2 and MsrB3 that this reaction is not efficient, which raises the possibility that Trx may not be the direct biological reducing system for MsrB2 and
80 MsrB3. It should be noted that a recent investigation92 has demonstrated good activity with MsrB2 and MsrB3 using a mitochondrial form of Trx (Trx2). However, that study required the use of human thioredoxin reductase at a concentration seven times greater than that used in our study (see Materials and Methods), which may not be physiologically relevant. Indeed, when they reduced the thioredoxin reductase amount to the levels we used, there was very little activity. We performed a set of experiments using Trx2 and a much higher level of thioredoxin reductase and were unable to replicate their results (data not shown). However, one must consider the possibility that a unique species of Trx may be the reducing agent or that another protein is required to transfer electrons from Trx to MsrB2 and MsrB3.
The ability of a heated bovine liver extract to support Msr activity with hMsrB3 in the absence of an exogenous reducing system provides evidence that animal cells contain a factor that, in the presence of EDTA, can substitute for Trx in this reaction.
The identification of Zn-MT as the active factor was based on the heat stability, purification characteristics, absorbance spectra at neutral and acidic pH values, gel analysis, metal determination and molecular weight analysis. The role of EDTA appears to be to release the zinc from the Zn-MT to form T, the apoform of MT, which can function as a reducing agent because of its high content of cysteines. In support of this conclusion, it was shown that T, prepared by acid treatment (see Materials and
Methods), could function as a reducing agent in the Msr system without EDTA. It is known that T is a small protein having ≈60 amino acids and a molecular mass in the range of 6–7 kDa. Of the total amino acids, approximately one-third are cysteines,
81 which could make this protein an important cellular source of sulfhydryl groups. For many years, it was felt that MT’s primary function was to scavenge free radicals and/or detoxify metals. However, in 1998, Maret and Vallee, in a seminal study93, postulated that the zinc-sulfur clusters in MT also acted as a sensor for the redox state of the cell.
Oxidation of Zn-MT resulted in release of zinc which then could be mobilized within the cell, whereas under reducing conditions, T would efficiently bind zinc. Thus, the major role of MT may be to control cellular zinc mobilization as a function of the redox state of the cell. However, there does not appear to be much information on other possible functions of T, the reduced apoform of MT, in addition to its critical role in binding zinc.
In one report similar to the current studies, it was reported that Zn-MT in the presence of
EDTA can reactivate the S-sulfonated (inactive) form of ribonuclease94. These researchers also suggested that the thiol groups in T are part of the pool of cellular thiols that can regulate redox reactions in a mechanism that is modulated by zinc chelation.
The results showing the ability of T to supply the reducing system for some of the Msr proteins support this conclusion and link the MT proteins to another cellular antioxidant system.
Although the results indicate that T can supply the reducing system for all of the
Msr enzymes tested, with the exception of hMsrB2, it is clear that the Trx system is the preferred reducing system for MsrA and eMsrB. If there is an important reducing role of
T, it is with hMsrB3. One of the unexplained findings in this study was the failure of T
(or Zn-MT and EDTA) to stimulate hMsrB2. As mentioned, all of the other Msr proteins that were tested showed significant activity with Zn-MT in the presence of EDTA.
82 Because MsrB2 and MsrB3 are both zinc proteins, they are thought to have similar reaction mechanisms22,43, which makes the lack of activity of T (the active agent) with hMsrB2 puzzling. One possibility is that different sulfhydryls on T react with hMsrB3 and hMsrB2. Thus, the active sulfhydryls in T that can interact with hMsrB3 cannot reduce the oxidized hMsrB2 intermediate. Since hMsrB2 shows only weak activity with the Trx system, it is not clear what may be the normal reducing system for this enzyme.
Because both MT-1 and MT-2 gave similar results in supporting Msr activity in the presence of EDTA, it was assumed that T derived from other MTs, such as MT-3
(found in the brain and reported to have growth inhibitory activity) and MT-495,96, would behave in a similar fashion. The electrospray MS analysis did not show the presence of
MT-3 in the samples. However, it is possible that slight structural differences in the MTs might be important, and it will be necessary to test the individual MT species for their ability to provide a reducing system for the Msr enzymes.
The in vivo significance of the present results is clearly not known. Certainly, there is no chelating agent, like EDTA, that is present in the cell to convert the MT to T.
However, Yang et al.97 have reported that as much as 50% of the total MT in mammalian tissues is present as T. Thus, the high concentration of T in tissues is consistent with a possible role of T as a cellular reducing agent, especially if there are mechanisms to regenerate T from T(O), as shown here with Trx. The heat step should have destroyed any T in the liver preparations, although, as shown in Figure 16, there was a slight activity in the heated S-100 in the absence of EDTA that could have been due to T that was not destroyed by the heat step.
83 Because oxidative stress is believed to release zinc and other metals from MT, one can postulate a reaction sequence, summarized in Figure 36, in which cells, under oxidative stress, mobilize zinc from Zn-MT for use in the many zinc-containing proteins.
The loss of the zinc from MT as a result of oxidation would yield T(O). As postulated in
Figure 36, T(O) can be reduced to T by the Trx system, and evidence for this reaction is shown in Table 7 and Figure 19. T can serve as a cellular reducing agent and reduce the oxidized Msr intermediates, either an enzyme-bound disulfide or sulfenic acid10,11,38,43. At present, it is not known how many of the cysteines in T can function to reduce the oxidized Msr proteins, but there is evidence that more than one cysteine on T is functional. Trx may be only one of the possible cellular reducing systems that can reduce T(O). It is known that oxidized glutathione can oxidize MT and cause the release of Zn from Zn-MT and that reduced glutathione can reduce T(O), which can bind zinc.
Of interest were the findings that certain selenium compounds, such as selenocystamine, can accelerate these reactions98,78. Preliminary experiments in the system under study have indicated that selenocystamine can be reduced to selenocysteamine by T and that selenocysteamine can supply the reducing system for both hMsrB2 and hMsrB3.
Further studies are required to determine whether the interaction between the Msr system and T has physiological relevance, especially because both may play an important role in protecting cells against oxidative damage. In addition, the possibility should be considered that T may be playing an important role as a cellular reductant for other systems.
84 Figure 36: Postulated role of Trx and MT in supplying the reducing requirement
for the Msr enzymes.
Since neither Trx nor T function well with hMsrB2, it seemed clear that there may be other factors that play a role in supplying the reducing system for hMsrB2 and possibly for hMsrB3. It has been known that selenium compounds can accelerate the binding of zinc to T and the release of zinc from Zn-MT in the presence of GSH and
GSSG, respectively78. In addition, previous work has demonstrated that Trx reductase can catalyze the reduction of oxidized selenium compounds such as selenocystine and ebselen and that reduced selenium compounds can reduce small molecules such as
99 80 81 H2O2 or lipid hydroperoxides as well as ferricytochrome c . It is also known that selenium-containing enzymes can receive hydrogen directly from mammalian Trx reductase, also a selenoprotein100. The present studies confirm that SeCm and other
85 selenium compounds, such as selenite and selenocystine, can be reduced by the Trx reducing system, i.e. NADPH, Trx reductase, and Trx. In the case of SeCm, there is efficient reduction to SeCem even in the absence of Trx, demonstrating that E. coli Trx reductase is able to directly reduce SeCm to SeCem in the presence of NADPH. Once reduced, the SeCem formed is a potent reducing agent for hMsrB2 and hMsrB3. As mentioned in the previous section, Trx is a poor reducing agent for both hMsrB2 and hMsrB3. However, in the presence of SeCm, the Trx reducing system is very effective with the MsrB enzymes. The SeCem, formed by the reduction of SeCm, appears to be an intermediate hydrogen carrier between the Trx reducing system and the oxidized
MsrB enzyme intermediate. A similar situation appears to be the case for T, which is also a much more efficient reducing agent for the MsrB enzymes in the presence of
SeCm. In this case, T reduces SeCm to SeCem, which serves as the direct reducing agent. In the case of the MsrA enzymes and eMsrB, which have at least one additional free cysteine to form a disulfide, it is known that the oxidized enzymes contain a disulfide bond that must be reduced in order for the enzyme to recycle38,10,11,101. These enzymes showed much less stimulation by SeCm in the presence of Trx or T. In contrast, neither hMsrB2 nor hMsrB3 have a free cysteine to form a disulfide bond with the cysteine at the catalytic site, and it has been postulated that the sulfenic acid intermediates on both MsrB2 and MsrB3 are directly reduced by Trx43. Our results suggest that Trx is a poor reducing agent for these enzymes but that in vitro, Trx and T can be potent reducing agents in the presence of an appropriate selenium compound, such as SeCm or selenocystine. These reactions are summarized in Figure 37. It should
86 be noted again that MsrB1, a known selenoprotein that also contains a second cysteine capable of forming a Se–S bond on the enzyme43, has not been studied because of the difficulty in expressing the recombinant protein.
Figure 37: Putative reactions involved in the reduction of the various Msr enzymes
by Trx, T, and SeCm. A. MsrA or other Msr enzymes that contain a second cysteine,
in addition to the catalytic cysteine, that is capable of forming a disulfide bond on the
enzyme. B. MsrB2 and MsrB3, examples of enzymes that do not contain a second
cysteine.
The in vivo significance of the role of selenium in the Msr reactions is not clear.
It would be of great interest if one could identify a naturally-occurring selenium compound in mammalian cells, similar to SeCem, that can act as an intermediate hydrogen carrier between Trx (or T) and the MsrB enzymes, similar to what has been observed in the in vitro studies described here. It should be noted that SeCm and
87 selenite have been identified as selenium metabolites in human urine102, but there is no evidence that they function as oxidoreductants. Selenocysteine is normally found in selenoproteins, where it can function as an oxidoreductant, and one should consider the possibility that there is a selenoprotein in cells that can transfer hydrogen directly to the
Msr enzymes. Although the possible role of selenium compounds as cellular reducing agents for the Msr enzymes remains unclear, administering compounds such as SeCm or selenocystine might be a novel way to increase the intracellular activity of the Msr enzymes, especially MsrB2 and MsrB3.
There are some data on the effect of selenium-deficient diets on Msr activity.
Mice on a selenium-deficient diet have lower in vitro levels of MsrB activity, which are presumed to be due to reduced MsrB1 activity since MsrB1 is a selenoprotein103.
However, based on the studies here, one might expect lower activity of both MsrB2 and
MsrB3 in vivo if a selenium compound is able to function as an oxidoreductant for
MsrB2 and MsrB3 in cells. Thus, a selenium-deficient diet could result in a severe reduction of total MsrB activity in cells.
In summary, the poor activity of the tested MsrB enzymes with the Trx system led us to investigate what other enzymes or factors might contribute to the efficient reduction of MsrB2 and MsrB3. Thionein and certain selenium compounds have been shown to provide strong reducing potential in vitro, but it remains to be demonstrated if these factors have any in vivo relevance. Thionein is an abundant protein, but whether or not it is available in its reduced state in proximity to the Msr enzymes is unknown, and the availability of selenocompounds in the cell is also unknown. Further experiments
88 are also warranted to identify any selenium-containing proteins which may interact with the Msr system.
4.2 The search for MsrA activators and inhibitors
As explained in the Introduction, activators of the Msr enzymes have therapeutic potential for the treatment of age-related diseases. Studies in yeast and flies suggest that improved MsrA activity may prolong life. Since it is not yet possible to introduce extra copies of MsrA in humans via gene therapy, the search for an activator is the most promising avenue of investigation. Also, inhibitors of MsrA could be used in in vitro studies to characterize mechanisms related to the management of oxidative stress.
The high-throughput assay described herein has the advantage of allowing the automated screening of thousands of compounds. The HTS will be done at The Scripps
Florida Research Institute, Jupiter, FL (TSRI). Florida Atlantic University has a special arrangement with TSRI to encourage collaboration on scientific problems of common interest. Dr. Peter Hodder at TSRI has been extremely helpful through the advice that he has given.
This high-throughput assay is unique in that it uses the solvent (DMSO) that the library compounds are dissolved in as a substrate for MsrA. The kinetics of the reaction are such that the Km value for DMSO (500 μM) is low enough that the substrate will not be rate limiting at any reasonable concentration of test compound. There are likely to be some challenges when translating the assay to 1536-well format, but the high Z factors at
89 the 384-well level provide a good degree of confidence that it will be possible. Since the assay has been worked out with both absorbance and fluorescence measurement, there will be flexibility in the method should problems be encountered in the translation to
μHTS.
4.3 Metabolism of the sulindac epimers and relationship to the Msr system
In prior studies on sulindac, the metabolism and biological effects were investigated only with the mixed epimers, although there was one study indicating a stereospecific oxidation of sulindac sulfide to sulindac-R by flavin-containing monooxygenases58. A complete understanding of sulindac's metabolism requires study of the individual epimers, and the lack of investigation of the in vivo effects of the individual epimers, despite the finding of the efficacy of sulindac in the treatment of certain cancers, seemed to us an area that should be investigated. In the rat feeding experiments, both the sulfone and sulfide metabolites are seen in plasma, regardless of which epimer is given. This provided the basis for studying both the reduction and oxidation pathways.
4.3.1 Reduction of sulindac epimers
Regarding the reduction of sulindac to sulindac sulfide, previous studies indicated that MsrA demonstrated stereospecific preference for the S epimer of sulindac, as with other methyl sulfoxides. The evidence that there was sulindac-S reductase activity detected in vivo in MsrA knockout mice (Table 11) suggests that either 1) MsrA
90 may not be the only enzyme responsible for the reduction of sulindac-S, or 2) the reduction is catalyzed by intestinal flora. In contrast to the results from the in vivo experiments, liver extracts from MsrA knockout mice showed very little reduction of the
S epimer.
With regard to sulindac-R reduction, it was found that MsrB2 and MsrB3 are capable of reducing the R epimer of peptide-bound methionine sulfoxide but could not reduce sulindac-R (data not shown). This finding, combined with the discovery of sulindac sulfide in the plasma of rats fed the R epimer, prompted a search for the unknown sulindac-R reducing enzyme(s). Despite a purification from rat liver of 200- fold or greater, it has not been possible to completely isolate and identify the enzyme.
As discussed in the Results section, there are couple of reasons to suggest that the enzyme is not MsrB1, namely that a cysteine variant of MsrB1 did not reduce sulindac-R and that an MS analysis did not identify MsrB1 in the band excised from the SDS-PAGE gel of the purified protein. However, the protein does clearly elute from the G50 sizing column in the 10-15 kDa. range, and a bioinformatic search of rat proteins shows only three oxioreductases in that size range: MsrB1, Trx and glutaredoxin-1 (Glrx1). Glrx1 can catalyze the reduction of disulfide bonds but not sulfoxides104, and Trx has no direct activity against sulindac-R (used as a control in Trx-dependent assays, data not shown).
Regardless of whether or not the unidentified protein is MsrB1, it has the signature of a MsrB-like enzyme, based on the facts that DABS-Met(O)-R can be reduced by it and that the competition experiments suggest that it is the same enzyme
91 that is reducing sulindac-R. It may be a new, as-yet-uncharacterized member of the
MsrB family.
4.3.2 Oxidation of sulindac epimers
Rat liver microsomes were initially used to study the oxidation of the epimers.
Due to differing activities with each epimer, it was demonstrated that there are independent pathways for the oxidation and reduction of the two epimers. After finding oxidative activity with both sulindac epimers in microsome preparations, it was suspected that cytochrome P450 enzymes were responsible. Due to the limited number of major P450 drug-metabolizing enzymes and their commercial availability, it was decided to characterize the oxidation of the epimers using these commercial preparations.
Studies with the purified P450 enzymes showed that the major enzymes responsible for oxidation of the R epimer include CYP1A1, CYP1A2, CYP1B1 and
CYP3A4, while CYP1A2, CYP2C8 and CYP3A4 are responsible for oxidation of the S epimer. This is consistent with the previous finding that, in rat feeding experiments, the mixed epimers of sulindac induce hepatic mRNA expression of CYP1A1, CYP1A2 and
CYP1B160. CYP1A1, which showed the highest level of mRNA induction from the mixed epimers, demonstrated a strong preference for oxidation of the R epimer in these studies.
In the studies with human hepatocytes, it was shown that the R epimer is oxidized by the relevant enzymes more effectively than the S epimer, even though the S epimer had a greater capacity to induce the enzyme system. From this observation, it
92 could be suggested that the S epimer has a higher affinity for the aryl hydrocarbon receptor or other P450-inducing receptors but that the enzymes themselves, collectively, reduce the R epimer at a greater rate. At first glance this might be considered odd behavior, but one must remember that sulindac is a xenobiotic and that other, naturally- occurring compounds were responsible for driving the evolution of the mammalian P450 system.
Figure 38 summarizes the results of the sulindac metabolism studies for both oxidation and reduction.
Figure 38: Metabolism of sulidac-R and sulindac-S to sulindac sulfone and sulindac
sulfide.
93 4.4 Directions for further study
4.4.1 Reducing systems for the Msr enzymes
At the time of the original discovery of an enzyme that reduces methionine sulfoxide, little was known about the full role of the family of Msr enzymes. Today it is known that there is an entire family of Msr enzymes, conserved across most organisms, which have varying degrees of specificity for a range of substrates: the R and S epimers of free and protein-bound Met(O) and a variety of other methyl sulfoxides. The catalytic activity of the Msr enzymes requires an efficient reducing system, and not all Msr enzymes can be efficiently reduced by the Trx system, as pointed out in this study. The physiological reducing system for the MsrB enzymes remains to be elucidated and represents a viable subject for further investigation. It would be useful to clarify which specific types of thioredoxin and thioredoxin reductase interact with which Msr enzymes in which cellular compartments and at what relative concentrations. Also, there may be other intermediate hydrogen carriers such as the selenocompounds discussed herein which play a role in the efficient catalysis of the full Msr cycle.
4.4.2 High throughput screening
As discussed above, the DMSO-based assay that has been worked out in 384- well plates should translate nicely to the 1536-well format. We have an R21 grant application that will be submitted for additional funding necessary to finish the assay development. The screening will be run first against a 1280-compound library (Library
94 of Pharmacologically Active Compounds, LOPAC) in 384-well format. This initial screening should illuminate any problems that remain before switching to the 1536-well format. The actual screening will be done at Scripps Florida, and after the screening run is done there will be an extensive process of follow up on the “hits,” as described in the
Results section.
4.4.3 Further purification and identification of the sulindac-R reductase
Normally, achieving a protein purification of greater than 200-fold should allow identification of the enzyme under study using modern protein identification methods.
Unfortunately, identification of the sulindac-R reductase has proven elusive. It is possible that it is in such low abundance that further purification will be necessary before it will appear as a band on a SDS-PAGE gel. The two major proteins which were found by MS analysis were fatty-acid binding protein and hemoglobin. There are published methods to remove these proteins by affinity purification105,106, and this might be a way to achieve another tenfold or more purification.
4.4.4 Biological activity of the sulindac epimers
The finding that sulindac may mediate its anticancer effects via reactive oxygen species ties this important drug to the Msr system, in addition to the fact that it is metabolized in part by Msr enzymes. As mentioned in the introduction, sulindac must be reduced to sulindac sulfide in order to serve as an NSAID. In preliminary studies in
95 this laboratory, cancer cells reduced sulindac-R at a greater rate than sulindac-S while normal cells showed a preference for sulindac-S (data not shown). This suggests that sulindac-R may be the preferred epimer for cancer therapy as there is the potential for lower NSAID-related toxicity in normal cells. Also, this differential metabolism of the epimers may offer clues to the pathways involved in the selective killing of cancer cells in the presence of oxidative stress. Further studies on the use of sulindac or sulindac derivatives in the treatment of cancer may some day provide effective therapies for a wide range of tumors.
96 REFERENCES
1. Harman, D. Aging: a theory based on free radical and radiation chemistry. J
Gerontol 11, 298-300 (1956).
2. Liu, Y., Fiskum, G. & Schubert, D. Generation of reactive oxygen species by the
mitochondrial electron transport chain. J Neurochem 80, 780-7 (2002).
3. (Entire issue devoted to DNA repair). Trends in Biochemical Sciences 20, 381
(1995).
4. Black, S., Harte, E.M., Hudson, B. & Wartofsky, L. A Specific Enzymatic
Reduction of l(-) Methionine Sulfoxide and a Related Nonspecific Reduction of
Disulfides. J. Biol. Chem. 235, 2910-2916 (1960).
5. Ejiri, S.I., Weissbach, H. & Brot, N. The purification of methionine sulfoxide
reductase from Escherichia coli. Anal Biochem 102, 393-8 (1980).
6. Brot, N., Weissbach, L., Werth, J. & Weissbach, H. Enzymatic reduction of protein-
bound methionine sulfoxide. Proc Natl Acad Sci U S A 78, 2155-8 (1981).
7. Rahman, M.A., Moskovitz, J., Strassman, J., Weissbach, H. & Brot, N. Physical
map location of the peptide methionine sulfoxide reductase gene on the Escherichia
coli chromosome. J Bacteriol 176, 1548-9 (1994).
8. Etienne, F., Spector, D., Brot, N. & Weissbach, H. A methionine sulfoxide
reductase in Escherichia coli that reduces the R enantiomer of methionine
sulfoxide. Biochem. Biophys. Res. Commun 300, 378-382 (2003).
97 9. Spector, D., Etienne, F., Brot, N. & Weissbach, H. New membrane-associated and
soluble peptide methionine sulfoxide reductases in Escherichia coli. Biochem.
Biophys. Res. Commun 302, 284-289 (2003).
10. Lowther, W.T., Brot, N., Weissbach, H. & Matthews, B.W. Structure and
mechanism of peptide methionine sulfoxide reductase, an "anti-oxidation" enzyme.
Biochemistry 39, 13307-13312 (2000).
11. Boschi-Muller, S. et al. A sulfenic acid enzyme intermediate is involved in the
catalytic mechanism of peptide methionine sulfoxide reductase from Escherichia
coli. J. Biol. Chem 275, 35908-35913 (2000).
12. Weissbach, H. et al. Peptide methionine sulfoxide reductase: structure, mechanism
of action, and biological function. Arch. Biochem. Biophys 397, 172-178 (2002).
13. Moskovitz, J. et al. Identification and characterization of a putative active site for
peptide methionine sulfoxide reductase (MsrA) and its substrate stereospecificity.
J. Biol. Chem 275, 14167-14172 (2000).
14. Sharov, V.S., Ferrington, D.A., Squier, T.C. & Schöneich, C. Diastereoselective
reduction of protein-bound methionine sulfoxide by methionine sulfoxide
reductase. FEBS Lett 455, 247-250 (1999).
15. Grimaud, R. et al. Repair of oxidized proteins. Identification of a new methionine
sulfoxide reductase. J. Biol. Chem 276, 48915-48920 (2001).
16. Lowther, W.T., Weissbach, H., Etienne, F., Brot, N. & Matthews, B.W. The
mirrored methionine sulfoxide reductases of Neisseria gonorrhoeae pilB. Nat Struct
Biol 9, 348-52 (2002).
98 17. Zhang, X. & Weissbach, H. Origin and evolution of the protein-repairing enzymes
methionine sulphoxide reductases. Biol Rev Camb Philos Soc 83, 249-57 (2008).
18. Kumar, R.A., Koc, A., Cerny, R.L. & Gladyshev, V.N. Reaction mechanism,
evolutionary analysis, and role of zinc in Drosophila methionine-R-sulfoxide
reductase. J Biol Chem 277, 37527-35 (2002).
19. Lin, Z. et al. Free methionine-(R)-sulfoxide reductase from Escherichia coli reveals
a new GAF domain function. Proc. Natl. Acad. Sci. U.S.A 104, 9597-9602 (2007).
20. Hansel, A. et al. Mitochondrial targeting of the human peptide methionine sulfoxide
reductase (MSRA), an enzyme involved in the repair of oxidized proteins. FASEB J
16, 911-913 (2002).
21. Vougier, S., Mary, J. & Friguet, B. Subcellular localization of methionine
sulphoxide reductase A (MsrA): evidence for mitochondrial and cytosolic isoforms
in rat liver cells. Biochem. J 373, 531-537 (2003).
22. Kim, H. & Gladyshev, V.N. Methionine sulfoxide reduction in mammals:
characterization of methionine-R-sulfoxide reductases. Mol Biol Cell 15, 1055-64
(2004).
23. Kryukov, G.V., Kumar, R.A., Koc, A., Sun, Z. & Gladyshev, V.N. Selenoprotein R
is a zinc-containing stereo-specific methionine sulfoxide reductase. Proc. Natl.
Acad. Sci. U.S.A 99, 4245-4250 (2002).
24. Huang, W., Escribano, J., Sarfarazi, M. & Coca-Prados, M. Identification,
expression and chromosome localization of a human gene encoding a novel protein
99 with similarity to the pilB family of transcriptional factors (pilin) and to bacterial
peptide methionine sulfoxide reductases. Gene 233, 233-240 (1999).
25. Moskovitz, J. et al. Escherichia coli peptide methionine sulfoxide reductase gene:
regulation of expression and role in protecting against oxidative damage. J
Bacteriol 177, 502-7 (1995).
26. St John, G. et al. Peptide methionine sulfoxide reductase from Escherichia coli and
Mycobacterium tuberculosis protects bacteria against oxidative damage from
reactive nitrogen intermediates. Proc. Natl. Acad. Sci. U.S.A 98, 9901-9906 (2001).
27. Moskovitz, J., Berlett, B.S., Poston, J.M. & Stadtman, E.R. The yeast peptide-
methionine sulfoxide reductase functions as an antioxidant in vivo. Proc. Natl.
Acad. Sci. U.S.A 94, 9585-9589 (1997).
28. Marchetti, M.A. et al. Silencing of the methionine sulfoxide reductase A gene
results in loss of mitochondrial membrane potential and increased ROS production
in human lens cells. Exp. Eye Res 83, 1281-1286 (2006).
29. Moskovitz, J. et al. Overexpression of peptide-methionine sulfoxide reductase in
Saccharomyces cerevisiae and human T cells provides them with high resistance to
oxidative stress. Proc. Natl. Acad. Sci. U.S.A 95, 14071-14075 (1998).
30. Yermolaieva, O. et al. Methionine sulfoxide reductase A protects neuronal cells
against brief hypoxia/reoxygenation. Proc. Natl. Acad. Sci. U.S.A 101, 1159-1164
(2004).
100 31. Kantorow, M. et al. Methionine sulfoxide reductase A is important for lens cell
viability and resistance to oxidative stress. Proc. Natl. Acad. Sci. U.S.A 101, 9654-
9659 (2004).
32. Prentice, H.M. et al. MsrA protects cardiac myocytes against
hypoxia/reoxygenation induced cell death. Biochem Biophys Res Commun 366,
775-8 (2008).
33. Moskovitz, J. et al. Methionine sulfoxide reductase (MsrA) is a regulator of
antioxidant defense and lifespan in mammals. Proc Natl Acad Sci U S A 98, 12920-
5 (2001).
34. Salmon, A.B. et al. Lack of methionine sulfoxide reductase A in mice increases
sensitivity to oxidative stress but does not diminish life span. The FASEB Journal
(2009). doi:10.1096/fj.08-127415 (in press).
35. Ruan, H. et al. High-quality life extension by the enzyme peptide methionine
sulfoxide reductase. Proc Natl Acad Sci U S A 99, 2748-53 (2002).
36. Le, D.T. et al. Functional Analysis of Free Methionine-R-sulfoxide Reductase from
Saccharomyces cerevisiae. J. Biol. Chem. 284, 4354-4364 (2009).
37. Shchedrina, V.A. et al. Overexpression of methionine-R-sulfoxide reductases has
no influence on fruit fly aging. Mech. Ageing Dev (2009).
doi:10.1016/j.mad.2009.04.003 (in press).
38. Lowther, W.T., Brot, N., Weissbach, H., Honek, J.F. & Matthews, B.W. Thiol-
disulfide exchange is involved in the catalytic mechanism of peptide methionine
sulfoxide reductase. Proc. Natl. Acad. Sci. U.S.A 97, 6463-6468 (2000).
101 39. Taylor, A.B., Benglis, D.M., Dhandayuthapani, S. & Hart, P.J. Structure of
Mycobacterium tuberculosis methionine sulfoxide reductase A in complex with
protein-bound methionine. J. Bacteriol 185, 4119-4126 (2003).
40. Weissbach, H., Resnick, L. & Brot, N. Methionine sulfoxide reductases: history and
cellular role in protecting against oxidative damage. Biochim Biophys Acta 1703,
203-12 (2005).
41. Porque, P.G., Baldesten, A. & Reichard, P. The Involvement of the Thioredoxin
System in the Reduction of Methionine Sulfoxide and Sulfate. J. Biol. Chem. 245,
2371-2374 (1970).
42. Russel, M. & Model, P. The role of thioredoxin in filamentous phage assembly.
Construction, isolation, and characterization of mutant thioredoxins. J Biol Chem
261, 14997-5005 (1986).
43. Kim, H. & Gladyshev, V.N. Different catalytic mechanisms in mammalian
selenocysteine- and cysteine-containing methionine-R-sulfoxide reductases. PLoS
Biol 3, e375 (2005).
44. Lee, J.W. et al. Gene structure, localization and role in oxidative stress of
methionine sulfoxide reductase A (MSRA) in the monkey retina. Exp. Eye Res 82,
816-827 (2006).
45. Marchetti, M.A. et al. Methionine sulfoxide reductases B1, B2, and B3 are present
in the human lens and confer oxidative stress resistance to lens cells. Invest.
Ophthalmol. Vis. Sci 46, 2107-2112 (2005).
102 46. Moskovitz, J., Weissbach, H. & Brot, N. Cloning and expression of a mammalian
gene involved in the reduction of methionine sulfoxide residues in proteins. Proc.
Natl. Acad. Sci. U.S.A 93, 2095-2099 (1996).
47. Etienne, F., Resnick, L., Sagher, D., Brot, N. & Weissbach, H. Reduction of
Sulindac to its active metabolite, sulindac sulfide: assay and role of the methionine
sulfoxide reductase system. Biochem Biophys Res Commun 312, 1005-10 (2003).
48. Stadtman, E.R., Moskovitz, J., Berlett, B.S. & Levine, R.L. Cyclic oxidation and
reduction of protein methionine residues is an important antioxidant mechanism.
Mol Cell Biochem 234-235, 3-9
49. Brot, N., Werth, J., Koster, D. & Weissbach, H. Reduction of N-acetyl methionine
sulfoxide: a simple assay for peptide methionine sulfoxide reductase. Anal.
Biochem 122, 291-294 (1982).
50. Johnson, D. & Travis, J. The oxidative inactivation of human alpha-1-proteinase
inhibitor. Further evidence for methionine at the reactive center. J. Biol. Chem 254,
4022-4026 (1979).
51. Abrams, W.R., Weinbaum, G., Weissbach, L., Weissbach, H. & Brot, N. Enzymatic
reduction of oxidized alpha-1-proteinase inhibitor restores biological activity. Proc.
Natl. Acad. Sci. U.S.A 78, 7483-7486 (1981).
52. Cochrane, C.G., Spragg, R. & Revak, S.D. Pathogenesis of the adult respiratory
distress syndrome. Evidence of oxidant activity in bronchoalveolar lavage fluid. J.
Clin. Invest 71, 754-761 (1983).
103 53. Shao, B. et al. Myeloperoxidase impairs ABCA1-dependent cholesterol efflux
through methionine oxidation and site-specific tyrosine chlorination of
apolipoprotein A-I. J. Biol. Chem 281, 9001-9004 (2006).
54. Shen, T.Y. et al. Synthesis of a new anti-inflammatory agent, cis-5-fluoro-2-methyl-
1-[p-(methylsulfinyl)benzylidenyl]-indene-3-acetic acid. Fed. Proc. 31, 577 (1972).
55. Huskisson, E.C. & Scott, J. Sulindac. Trials of a new anti-inflammatory drug. Ann.
Rheum. Dis 37, 89-92 (1978).
56. Duggan, D.E., Hooke, K.F., Risley, E.A., Shen, T.Y. & Arman, C.G. Identification
of the biologically active form of sulindac. J Pharmacol Exp Ther 201, 8-13
(1977).
57. Meade, E.A., Smith, W.L. & DeWitt, D.L. Differential inhibition of prostaglandin
endoperoxide synthase (cyclooxygenase) isozymes by aspirin and other non-
steroidal anti-inflammatory drugs. J Biol Chem 268, 6610-4 (1993).
58. Hamman, M.A., Haehner-Daniels, B.D., Wrighton, S.A., Rettie, A.E. & Hall, S.D.
Stereoselective sulfoxidation of sulindac sulfide by flavin-containing
monooxygenases: Comparison of human liver and kidney microsomes and
mammalian enzymes. Biochemical Pharmacology 60, 7-17 (2000).
59. Plant, N. The human cytochrome P450 sub-family: Transcriptional regulation,
inter-individual variation and interaction networks. Biochimica et Biophysica Acta
(BBA) - General Subjects 1770, 478-488 (2007).
104 60. Ciolino, H.P., MacDonald, C.J., Memon, O.S., Bass, S.E. & Yeh, G.C. Sulindac
regulates the aryl hydrocarbon receptor-mediated expression of Phase 1 metabolic
enzymes in vivo and in vitro. Carcinogenesis 27, 1586-1592 (2006).
61. Ciolino, H.P., Bass, S.E., MacDonald, C.J., Cheng, R.Y.S. & Yeh, G.C. Sulindac
and its metabolites induce carcinogen metabolizing enzymes in human colon cancer
cells. Int J Cancer 122, 990-8 (2008).
62. Waddell, W.R. & Loughry, R.W. Sulindac for polyposis of the colon. J Surg Oncol
24, 83-87 (1983).
63. Hixson, L.J. et al. Antiproliferative effect of nonsteroidal antiinflammatory drugs
against human colon cancer cells. Cancer Epidemiol. Biomarkers Prev 3, 433-438
(1994).
64. Park, J. et al. Combination treatment with arsenic trioxide and sulindac enhances
apoptotic cell death in lung cancer cells via activation of oxidative stress and
mitogen-activated protein kinases. Oncol Rep 20, 379-84 (2008).
65. Webster, W.S. & Leibovich, B.C. Exisulind in the treatment of prostate cancer.
Expert Rev Anticancer Ther 5, 957-62 (2005).
66. Seo, S. et al. Sulindac-derived reactive oxygen species induce apoptosis of human
multiple myeloma cells via p38 mitogen activated protein kinase-induced
mitochondrial dysfunction. Apoptosis 12, 195-209 (2007).
67. Resnick, L., Rabinovitz, H., Binninger, D., Marchetti, M. & Weissbach, H. Topical
sulindac combined with hydrogen peroxide in the treatment of actinic keratoses. J
Drugs Dermatol 8, 29-32 (2009).
105 68. Marchetti, M. et al. Sulindac enhances the killing of cancer cells exposed to
oxidative stress. PLoS ONE 4, e5804 (2009).
69. Lavine, T.F. The formation, resolution, and optical properties of the
diastereoisomeric sulfoxides derived from l-methionine. J. Biol. Chem. 169, 477-
492 (1947).
70. Minetti, G., Balduini, C. & Brovelli, A. Reduction of DABS-L-methionine-dl-
sulfoxide by protein methionine sulfoxide reductase from polymorphonuclear
leukocytes: stereospecificity towards the l-sulfoxide. Ital J Biochem 43, 273-83
(1994).
71. Rahman, M.A., Nelson, H., Weissbach, H. & Brot, N. Cloning, sequencing, and
expression of the Escherichia coli peptide methionine sulfoxide reductase gene. J
Biol Chem 267, 15549-51 (1992).
72. Sagher, D. et al. Thionein can serve as a reducing agent for the methionine
sulfoxide reductases. Proc. Natl. Acad. Sci. U.S.A 103, 8656-8661 (2006).
73. Vasak, M. Standard isolation procedure for metallothionein. Methods Enzymol 205,
41-4 (1991).
74. Klein, D., Sato, S. & Summer, K.H. Quantification of oxidized metallothionein in
biological material by a Cd saturation method. Anal Biochem 221, 405-9 (1994).
75. Li, T.Y., Minkel, D.T., Shaw, C.F. & Petering, D.H. On the reactivity of
metallothioneins with 5,5'-dithiobis-(2-nitrobenzoic acid). Biochem J 193, 441-6
(1981).
106 76. Hunt, J.B., Neece, S.H. & Ginsburg, A. The use of 4-(2-pyridylazo)resorcinol in
studies of zinc release from Escherichia coli aspartate transcarbamoylase. Anal
Biochem 146, 150-7 (1985).
77. Shaw, C.F., Savas, M.M. & Petering, D.H. Ligand substitution and sulfhydryl
reactivity of metallothionein. Methods Enzymol 205, 401-14 (1991).
78. Chen, Y. & Maret, W. Catalytic selenols couple the redox cycles of metallothionein
and glutathione. Eur. J. Biochem 268, 3346-3353 (2001).
79. Chaudiere, J., Courtin, O. & Leclaire, J. Glutathione oxidase activity of
selenocystamine: a mechanistic study. Arch. Biochem. Biophys 296, 328-336
(1992).
80. Björnstedt, M., Hamberg, M., Kumar, S., Xue, J. & Holmgren, A. Human
thioredoxin reductase directly reduces lipid hydroperoxides by NADPH and
selenocystine strongly stimulates the reaction via catalytically generated selenols.
J. Biol. Chem 270, 11761-11764 (1995).
81. Kumar, S., Björnstedt, M. & Holmgren, A. Selenite is a substrate for calf thymus
thioredoxin reductase and thioredoxin and elicits a large non-stoichiometric
oxidation of NADPH in the presence of oxygen. Eur. J. Biochem 207, 435-439
(1992).
82. Sagher, D., Brunell, D., Brot, N., Vallee, B.L. & Weissbach, H. Selenocompounds
can serve as oxidoreductants with the methionine sulfoxide reductase enzymes. J.
Biol. Chem 281, 31184-31187 (2006).
107 83. Wizemann, T.M. et al. Peptide methionine sulfoxide reductase contributes to the
maintenance of adhesins in three major pathogens. Proc. Natl. Acad. Sci. U.S.A 93,
7985-7990 (1996).
84. Hassouni, M.E., Chambost, J.P., Expert, D., Van Gijsegem, F. & Barras, F. The
minimal gene set member msrA, encoding peptide methionine sulfoxide reductase,
is a virulence determinant of the plant pathogen Erwinia chrysanthemi. Proc. Natl.
Acad. Sci. U.S.A 96, 887-892 (1999).
85. Zhang, Chung & Oldenburg A Simple Statistical Parameter for Use in Evaluation
and Validation of High Throughput Screening Assays. J Biomol Screen 4, 67-73
(1999).
86. Simeonov, A. et al. Quantitative High-Throughput Screen Identifies Inhibitors of
the Schistosoma mansoni Redox Cascade. PLoS Negl Trop Dis 2, e127 (2008).
87. Lauer-Fields, J.L. et al. High throughput screening of potentially selective MMP-13
exosite inhibitors utilizing a triple-helical FRET substrate. Bioorg. Med. Chem 17,
990-1005 (2009).
88. Ellman, G.L. A colorimetric method for determining low concentrations of
mercaptans. Arch. Biochem. Biophys 74, 443-450 (1958).
89. Lee, S.C. & Renwick, A.G. Sulphoxide reduction by rat and rabbit tissues in vitro.
Biochemical Pharmacology 49, 1557-1565 (1995).
90. Kitamura, S. et al. Extremely High Drug-Reductase Activity Based on Aldehyde
Oxidase in Monkey Liver. Biological & Pharmaceutical Bulletin 24, 856-859
(2001).
108 91. Brot, N. & Weissbach, H. Biochemistry and physiological role of methionine
sulfoxide residues in proteins. Arch. Biochem. Biophys 223, 271-281 (1983).
92. Kim, H. & Kim, J. Thioredoxin as a reducing agent for mammalian methionine
sulfoxide reductases B lacking resolving cysteine. Biochemical and Biophysical
Research Communications 371, 490-494 (2008).
93. Maret, W. & Vallee, B.L. Thiolate ligands in metallothionein confer redox activity
on zinc clusters. Proc Natl Acad Sci U S A 95, 3478-82 (1998).
94. Zhang, S., Li, J., Wang, C. & Tsou, C. Metal regulation of metallothionein
participation in redox reactions. FEBS Letters 462, 383-386 (1999).
95. Uchida, Y., Takio, K., Titani, K., Ihara, Y. & Tomonaga, M. The growth inhibitory
factor that is deficient in the Alzheimer's disease brain is a 68 amino acid
metallothionein-like protein. Neuron 7, 337-347 (1991).
96. Hidalgo, J., Aschner, M., Zatta, P. & Vasák, M. Roles of the metallothionein family
of proteins in the central nervous system. Brain Res. Bull 55, 133-145 (2001).
97. Yang, Y., Maret, W. & Vallee, B.L. Differential fluorescence labeling of cysteinyl
clusters uncovers high tissue levels of thionein. Proc. Natl. Acad. Sci. U.S.A 98,
5556-5559 (2001).
98. Jacob, C., Maret, W. & Vallee, B.L. Selenium redox biochemistry of zinc-sulfur
coordination sites in proteins and enzymes. Proc. Natl. Acad. Sci. U.S.A 96, 1910-
1914 (1999).
109 99. Zhao, R. & Holmgren, A. A novel antioxidant mechanism of ebselen involving
ebselen diselenide, a substrate of mammalian thioredoxin and thioredoxin
reductase. J. Biol. Chem 277, 39456-39462 (2002).
100. Björnstedt, M., Xue, J., Huang, W., Akesson, B. & Holmgren, A. The thioredoxin
and glutaredoxin systems are efficient electron donors to human plasma glutathione
peroxidase. J. Biol. Chem 269, 29382-29384 (1994).
101. Antoine, M., Boschi-Muller, S. & Branlant, G. Kinetic characterization of the
chemical steps involved in the catalytic mechanism of methionine sulfoxide
reductase A from Neisseria meningitidis. J. Biol. Chem 278, 45352-45357 (2003).
102. Francesconi, K.A. & Pannier, F. Selenium metabolites in urine: a critical overview
of past work and current status. Clin. Chem 50, 2240-2253 (2004).
103. Moskovitz, J. & Stadtman, E.R. Selenium-deficient diet enhances protein oxidation
and affects methionine sulfoxide reductase (MsrB) protein level in certain mouse
tissues. Proc. Natl. Acad. Sci. U.S.A 100, 7486-7490 (2003).
104. Björnstedt, M., Kumar, S., Björkhem, L., Spyrou, G. & Holmgren, A. Selenium and
the thioredoxin and glutaredoxin systems. Biomed. Environ. Sci 10, 271-279
(1997).
105. Wilton, D.C. Studies on fatty-acid-binding proteins. The purification of rat liver
fatty-acid-binding protein and the role of cysteine-69 in fatty acid binding.
Biochem J. 261, 273–276 (1989).
110 106. Klein, M. & Mihaesco, C. Affinity chromatography: specific binding of
hemoglobin on agarose linked haptoglobin. Biochem. Biophys. Res. Commun 52,
774-778 (1973).
111