INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afreet reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
University Microfilms International A Bell & Howell Information Company 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 313/761-4700 800/521-0600 Order Number 9218983
Purification, characterization, and inhibition of S-adenosyl-X-methionine decarboxylase (E.C. 4.1.1.50) from Acanthamoeba castellanni N eff
Hugo, Eric R., Ph.D. The Ohio State University, 1992
UMI 300 N. ZeebRd. Ann Arbor, MI 48106 Purification, Characterization, and Inhibition of S-Adenosyl-L-Methionine
Decarboxylase (E.C. 4.1.1.50) from Acanthamoeba castellanii Neff
DISSERTATION
Presented in Partial Fulfillment of the Requirement for the Degree of Doctor of
Philosophy in the Graduate School of the Ohio State University
by
Eric R. Hugo, B.S., M.S.
* * * * *
The Ohio State University
1992
Dissertation Committee Approved by
Thomas J. Byers, Committee Chair
Berl O. Oakley
Donald H. Dean Adviser
Roy H. Tassava Department of Molecular Genetics To Vickie ACKNOWLEGEMENTS
I would like to thank Drs. Thomas and Sandra Byers for their gracious help during my tenure in Columbus. I would also like to thank Dr. Wayne Parrish for support during part of this study as well as for many helpful discussions. I am greatly appreciative of the all of the assistance Ms. Rebbecca Gast provided during my stay in Dr. Byers laboratory. Finally, I thank my wife Vickie for her love, kindness, and patience during the many long hours I spent working in the laboratory. VITA
November 27,1961 ...... Born Ft. Thomas, Ky.
May, 1983 ...... B.S. Northern Kentucky University Highland Heights, Ky.
1983-1986 ...... Graduate Teaching Associate Dept, of Microbiology The Ohio State University Columbus, OH
December, 1986 ...... M.S. The Ohio State University Columbus, OH
1986-1990 ...... Graduate Teaching and Research Associate, Dept, of Molecular Genetics, The Ohio State Univesity, Columbus, OH
1990-199 1...... Instructor, Denison University Granville, OH
1991-presen t ...... Research Associate, DepL of Biochemistry and Molecular Biophysics, Washington University School of Medicine St. Louis, MO PUBLICATIONS
Abstracts:
Byers, T.J. and E.R. Hugo. 1990. S-adenosylraethionine decarboxylase: a possible role in the encystment of Acanthamoeba castellanii. Invest. Opthal. Vis. Sci. Abstract 2066 31:420.
Hugo, E.R. and T.J. Byers. 1989. Purification and characterization of S-adenosyl-L-methionine decarboxylase from Acanthamoeba castellanii. J. Ceh Biol. Abstract 1403 109:255a.
Hugo, E.R. and T.J. Byers. 1989. S-adenosylmethionine decarboxylase of Acanthamoeba castellanii (Neff strain). J. Protozool Abstract 198 37(4) :42A
Papers and Proceedings
Byers, T.J., B.G. Kim, L.E. King, and E.R. Hugo. 1991. Molecular aspects of the cell cycle and encystment in Acanthamoeba. Rev. Infect. Dis.13(Suppl 5):S378-84.
Hugo, E.R., W.R. McLaughlin, K. Oh, and O.H. Tuovinen. 1991. Quantitative enumeration of Acanthamoeba for evaluation of cyst inactivation in contact lens care solutions. Invest. Vis. Ophthalmol. Res.32(3):655-57.
Byers, T.J., E.R. Hugo, and V. Stewart. 1990. Genes of Acanthamoeba: DNA, RNA and protein sequences (a review), J. Protozool. 37(4):17S-25S.
FIELDS OF STUDY
Major Field: Molecular Biology
v ABBREVIATIONS
apparent Km ab: antibody
agm: agmatine MTA: methylthioadenosine
Ber: berenil NEM: Neffs Encystment medium
BLOTTO Bovine-lacto transfer Norspd: norspermidine optimizer
CBB: Coumassie Brilliant Blue OGMA: optimal growth medium A
Dap: 1,3-diaminopropane Ora: ornithine
dcSAM: decarboxylated SAM ODC: ornithine decarboxylase
DFMO: a-difluoromethylomithine Pi: inorganic phosphate DGM-11: defined growth medium PA: polyamine
DTT: dithiothreatol Pent: pentamidine
EB: ethidium bromide Put: putrescine
EEA: encystment enhancing act. Pro: propamidine
HPLC: high performance liquid SAMDC: S-adenosylmethionine chromatography decarboxylase
HS: hydroxystilbamidine SAM: S-adenosylmethionine kDa: kilodalton TPCK: tosylphenylalanyl chloromethyl ketone
Kl: inhibition constant velocity TABLE OF CONTENTS
ACKNOWLEGEMENTS ...... iii
VTTA ...... iv
LIST OF FIGURES______xi
LIST OF TABLES______xiv
Introduction. ------1 1. Literature Review------3
l.L Polyamines ______3
1.1.1 Polyamine function ______3
1.1.2 Polyamine synthesis ...... 7 1.1.3 Regulation of polyamine biosynthesis ...... 12
1.1.4 Polyamine function and synthesis as targets of
chemotherapy ------14
1.2 Acanthamoeba ______19
1.2.1 General characteristics...... 19
1.2.2 Developmental biology ...... 21
1.2.3 Polyamine metabolism ...... -2 4
1.2.4 Molecular biology ...... 25 1.2.5 Pathogenicity ...... 26
2. Materials and Methods ...... 31
2.1 Cell culture and growth ...... 31
2.1.1 Cell lines ...... 31
2.1.2 Media...... 32
2.1.3 Cell growth ...... 32
2.1.4 Effect of diamidines on growth ...... 33
2.2 Enzyme purification ...... 34
2.2.1 Homogenization and extract preparation ...... 34
2.2.2 Precipitation ...... 35
2.2.3 Hydrophobic interaction chromatography ...... 35
2.2.4 DEAE anion exchange chromatography ...... 36
2.2.5 Gel filtration chromatography ...... 37
2.3 Electrophoresis ...... 37
2.3.1 Sodium dodecyl sulfate polyacrylamide gel electrophoresis 37
2.3.1 Agarose isoelectric focusing gel electrophoresis ...... 38
2.4 SAMDC assay...... 40
2.4.1 Quantitative microassay ...... 40 2.4.2 Qualitative microtiter plate assay ...... 43
2.4.3 Determination of pH optima ...... 44
2.4.4 Kinetic analysis ...... 44
2.5 Physical structure analysis ...... 44
2.5.1 Determination of a-subunit size ...... 44
2.5.2 N-terminal amino acid sequence determination ...... 46 2.5.3 Tryptic peptide map of SAMDC ...... 47
2.6 Developmental regulation of enzyme activity ...... 48
2.6.1 SAMDC activity during encystment ...... 48
2.6.2 Attempts to quantitate SAMDC protein ...... 48
2.6.3 In vivo substrate concentrations ...... 53
3. Results...... 53 3.1 Purification of S-adenosylmethionine decarboxylase ...... 53
3.1.1 Preparation of cell extracts ...... 53 3.1.2 Hydrophobic interaction chromatography ...... 54
3.1.3 Anion exchange chromatography ...... 57
3.1.4 Gel filtration chromatography ...... 57
3.1.5 Summary of the purification of SAMDC ...... 62
3.2 Development of SAMDC microassays ...... 62
3.3 Physical characteristics of SAMDC ...... 65
3.3.1 Subunit constitution and molecular weight ...... 65
3.3.2 Isoelectric point ...... 69
3.3.3 Optimal pH ...... 69
3.3.4 Special properties ...... 69
3.3.5 N-terminal amino acid sequence of the a-subunit ...... 76
3.3.6 Tryptic peptide map ...... 80
3.4 Biochemical properties of SAMDC ...... 85
3.4.1 Michaelis constant for SAMDC ...... 85
3.4.2 Competitive inhibition of SAMDC ...... 86
3.4.3 In vivo effects of diamidines ...... 90
ix 3.5 Regulation of SAMDC ...... 107
3.5.1 In vivo stability of SAMDC ...... 114
3.5.2 Developmental regulation ...... 114
3.6 Production of a polyclonal antibody to SAMDC ...... 120
3.7 S-adenosyl-L-methionine content ...... 120
4. DISCUSSION...... 127
4.1 Enzyme Purification ...... 127
4.1.1 Purification strategy and improvements ...... 127
4.1.2 Physical and kinetic parameters ...... 129
4.1.3 Amino acid sequence ...... 131
4.1.4 Tryptic peptide mapping of SAMDC ...... 132
4.1.5 Biochemical Characteristics ...... 133
4.2 Microassay Applications ...... 137
4.2.1 In vivo response to SAMDC inhibitors ...... 137
4.3 Regulation of SAMDC ...... 139
4.3.1 Regulation of SAMDC activity ...... 139
4.3.2 Intracellular SAM concentration ...... 140
4.3.3 Attempts to quantitate SAMDC protein ...... 141
4.3.4 Attempts to clone the SAMDC cDNA ...... 141
4.4 Future directions...... 142
4.5 Summary...... 144
5. Bibliography ...... 147
6. Appendix A ...... 164
7. Appendix B ...... 165
x LIST OF TABLES
Table Page
1. SDS-PAGE molecular weight marker proteins ...... 39
2. Standard marker proteins used for agarose IEF ...... 41
'3. HPLC gel filtration results for TSK-250 gel filtration ...... 63
4. Summary of the purification of SAMDC ...... 65
5. Determination of subunit Mr by SDS-PAGE ...... 66
6. Results of the agarose isoelectric focusing of SAMDC ...... 71
7. Effect of pH on in vitro SAMDC activity...... 74
8 . Amino acid sequences of SAMDC from various sources ...... 79
9. Tryptic digest fragments of SAMDC resolved by HPLC ...... 81
xi Summary of the kinetic properties of SAMDC, 87
Effect of MGBG on the in vitro activity of SAMDC 91
Effect of ethidium bromide on the in vitro activity of SAMDC 93
Effect of berenil on the in vitro activity of SAMDC 95
Effect of propamidine on the in vitro activity of SAMDC 97
Effect of pentamidine on the in vitro activity of SAMDC 99
Effect of hydroxystilbamidine on the in vitro activity of SAMDC 101
Summary of the effects of diamidines on the in vitro activity of SAMDC 104
Effects of diamidines on Acanthamoeba survival 105
Effects of G418 on protein synthesis and SAMDC activity 108
Effects of cycloheximide on protein synthesis and SAMDC activity 111
Effects of spermidine on the in vitro activity of SAMDC 115
xii 22. SAMDC specific activity during growth ...... 117
23. SAMDC specific activity during encystment ...... 118
24. Structures and retention properties of seven adeneine-containing compounds resolved by HPLC ...... 122
25. SAM quantitation by HPLC results ...... 125
xiii LIST OF FIGURES
Figure Page
1. Polyamine biosynthetic pathway for A. castellanii...... 9
2. SAM metabolism ...... 16
3. Quantitative microassay reaction tube ...... 42
4. Results of the SAMDC qualitative microassay ...... 45
5. Results of hydrophobic interaction chromatography ...... 55
6. Results of anion exchange chromatography ...... 58
7. HPLC gel filtration Mr standards 59
8. HPLC gel filtration of SAMDC ...... 61
9. Calibration curve for TSK-250 gel filtration ...... 64
xiv 10. SDS-PAGE electrophoresis of SAMDC ...... 67
11. Plot of the relative mobility of denatured standard proteins separated by SDS-PAGE on 16% solids gel ...... 68
12. Wide-range isoelectric focusing of labeled SAMDC ...... 70
13. Plot of IEF standards isoelectric point (pi) versus relative mobility ...... 72
14. Histogram of SAMDC activity at various H+ concentrations ...... 73
15. Hypothetical mechanism for the formation of the N-terminal pyruvate of SAMDC ...... 75
16. Hypothetical mechanism for the reductive labeling of SAMDC with radioactive SAM ...... 77
17. Wide-range isoelectric focusing of l^C- labeled SAMDC ...... 78
18. HPLC chromatogram of amoeba SAMDC tryptic peptides resolved by RPIP chromatography ...... 82
19. HPLC chromatogram of bovine serum albumin tryptic peptides resolved by RPIP chromatography ...... 83
20. HPLC chromatogram of cytochrome C tryptic peptides resolved by RPIP chromatography ...... 84
21. Kinetic properties of Acanthamoeba S AMDC...... 88
xv 22. Effect of MGBG on the kinetic properties of SAMDC ...... 92
23. Effect of ethidium bromide on the kinetic properties of SAMDC ...... 94
24. Effect of berenil on the kinetic properties of SAMDC ...... 96
25. Effect of propamidine on the kinetic properties of SAMDC ...... 98
26. Effect of pentamidine on the kinetic properties of SAMDC ...... 100
27. Effect of hydroxystibamidine on the kinetic properties of SAMDC 102
28. Effect of diamidines on Acanthamoeba survival ...... 105
29. The effects of G 418 inhibition of protein synthesis on SAMDC activity ...... 109
30. The effects of cycloheximide inhibition of protein synthesis on SAMDC activity ...... 112
31. The effect of spermidine on the in vivo specific activity of SAMDC 116
32. SAMDC in vivo specific activity during encystment ...... 119
33. Separation of SAM, SAH and MTA by reversed-phase ion-pair chromatography ...... 123
xvi 34. Graph of SAM concentration versus relative peak area ...... 124
35. Structures of compounds that interact with SAMDC ...... 135
xvii INTRODUCTION AND GOALS
Acanthamoeba is a genus of small free-living amoebae (SFLA) that are ubiquitous in nature. The organisms have been isolated from around the globe—from surface soil and water sources to deep marine sites (Sawyer, et al., 1977). Originally thought to be primarily an opportunistic pathogen at best, the amoeba recently has been shown to be a more significant human pathogen (Visvisvara and Stehr-Green, 1990). It is responsible for a painful and debilitating eye disease known as Acanthamoeba keratitis. Although rare,the disease is important because it is associated with contact lens wear, is difficult to diagnose and treat and can cause blindness (Stehr-Green et al 1987).
Currently, the most successful means of treating human Acanthamoeba infections is surgery, which typically entails the removal of the infected cornea and replacement with donor tissue (Osata, et al., 1991). Because of the drastic measures needed to cure many cases of Acanthamoeba keratitis, a variety of chemotherapeutic methods have been evaluated. Initially, the only effective chemotherapeutic agents tested were antifungal compounds such as miconazole and ketokonazole (Visvesvara, et al., 1983; Hirst, et a l,
1984). Successful treatment of Acanthamoeba keratitis was, however, limited to very few cases. Recent research in several laboratories has focused on polyamine metabolism as a possible target for chemotherapy. Inhibitors of polyamine synthesis and transport have been used to cure several other protozoan infections. Bacchi, et al. (1981) first demonstrated that inhibition of polyamine metabolism could completely eliminate an infection by Trypanosoma brucei brucei. Since this early work, a number of protistan 2 pathogens have been effectively treated with antipolyamine agents. Organisms that have been tested include: Trypanosoma spp., Leishmania spp., Plasmodium sppEim eria spp.,
Pneumocystis carinii,, Cryptosporidium spp., and Acanthamoeba spp. (Bacchi and
McCann, 1987).
The most effective inhibitors of polyamine metabolism used against protozoans have been inhibitors of ornithine decarboxylase (e.g. DFMO or eflomithine) and inhibitors of S-adenosyl-L-methionine decarboxylase (e.g. pentamidine and propamidine). Previous studies from this laboratory examined the first enzyme and the effects of its inhibitors on cell multiplication and differentiation (Kim et al. 1987a, b).
Antimetabolites that affected ornithine decarboxylase arrested cell multiplication without inducing cell differentiation. In contrast, our laboratory demonstrated that agents known from studies on other organisms to inhibit S-adenosyl-L-methionine decarboxylase
(SAMDC) (e.g. pentamidine (Byers, Kim et al. 1991) and berenil (Akins and Byers,
1980)), could both arrest multiplication and induce encystment. These two agents and related compounds were generally much more effective inhibitors of cell replication than the inhibitors of ornithine decarboxylase. Their tendency to stimulate the transformation of metabolically active amoebae into dormant cysts, however, seems counterproductive to chemotherapy. Nevertheless, these inhibitors may be good probes for exploring the molecular basis of encystment.
This study focuses on SAMDC and its inhibitors. It describes purification and characterization of the enzyme including its sensitivity to a number of potential chemotheraputic agents. It also begins a study of the enzyme's regulation with an eventual goal of testing its possible role in encystment. 3
Project Goals
The major goals of this study are summarized as follows:
1. To purify S-adenosyl-L-methionine decarboxylase (SAMDC, E.C. 4.1.1.50) to homogeneity.
2. To completely characterize the physical parameters of SAMDC.
3. To determine the biochemical properties of SAMDC.
4. To examine the effects of SAMDC inhibitors that were known to induce encystment on in vitro SAMDC activity.
5. To compare the in vivo and in vitro effects of SAMDC inhibitors.
6. To describe the regulation and in vivo stability of SAMDC activity.
7. To examine the regulation of SAMDC activity during differentiation.
10. To isolate a cDNA for SAMDC with hopes of using this cDNA as a probe for determining gene activity and as a tool to eliminate SAMDC activity from Acanthamoeba by antisense oligonucleotide inhibition. LITERATURE REVIEW
1.1. Polyamines 1.1.1. Polyamine function
Polyamines are small, organic polycationic molecules. In most eucaryotes and
procaryotes putrescine, spermidine, and spermine are the major polyamines (Tabor and
Tabor, 1984a). Additionally, in some organisms, less commonly encountered polyamines such as cadaverine, diaminopropane, norspermidine, and norspermine exist.
All of these molecules share a common feature—they possess a strong positive charge at physiological pH. This strong positive charge allows the polyamine to interact with nucleic acids, acidic proteins, and phospholipids to name but a few of the many constituents of a cell that are negatively charged. In all eucaryotic organisms where it has been possible to eliminate synthesis of polyamines by mutation, it has been shown that they are absolutely required for growth. Those eucaryotes that have been tested include: Chinese Hamster ovary cells (CHO) (Pohjanpelto, et al., 1981), Neurospora crassa (Paulus, et al, 1982), and Saccharomyces cerevisiae (Cohn, et al., 1980). All of the previously mentioned cases were those in which ornithine decarboxylase (ODC), the first enzyme in polyamine biosynthesis, had been eliminated by null mutations. ODC mutants demonstrate an obligate requirement for exogenous putrescine. Because of these results it would appear that polyamines are essential for growth in these organisms.
Polyamines are not absolutely required for growth in procaryotic organisms. In one experiment, Tabor, et al. (1980) created a strain of Escherichia coli that was unable
4 to produce any of the enzymes necessary for polyamine biosynthesis. This strain was not auxotrophic for polyamines, but grew, in the absence of exogenously added polyamines, at one third the rate of polyamine supplemented cultures. One phenotypic trait of this strain was resistance to production of bacteriophage X after infection or induction of prophages. It has been possible in at least one instance to create strains of E. coli that are auxotrophic for polyamines under certain conditions (Tabor, et al., 1981). In this case, a strain of E. coli lacking all of the polyamine biosynthetic enzymes was transformed with a mutant ribosomal protein
allele (rpsL or strA). The transformed strain was auxotrophic for polyamines at temperatures above 42° C.
Additional biochemical evidence supporting the claim that polyamines are necessary for growth comes from the fact that for most eucaryotes polyamine levels become elevated during proliferation (Tabor and Tabor, 1984a). The opposite of this is true in resting cells. In mammalian lymphocytes one of the first responses to stimulation by mitogenic agents is a rapid increase in polyamine biosynthesis (Mustelin, et al, 1988).
A similar stimulation of polyamine biosynthesis can be brought about in response to hormones (Scalabrino, et al, 1991). In rapidly proliferating tumor cells large pools of polyamines exist in comparison to the normal progenitor cells (Heby, et al., 1987).
Polyamines have been shown to interact with nucleic acids. Sakai, et al. (1975) reported that the interaction between spermidine and t-RNA is specific. Each t-RNA specifically binds 16-17 spermidine molecules. Two to three molecules bind at high affinity sites whereas the remaining spermidine molecules bind at sites with approximately 10-fold lower affinities. These results indicate that it is not merely interaction based on strong positive vs strong negative charges on the molecules since the charge distribution on nucleic acids is essentially uniform. Polyamines have also been shown to cause condensation of DNA (Gosule and Schellman, 1978; Fredericq, et al,
1991) as well as B-DNA to Z-DNA transitions in vitro (Behe and Felsenfeld, 1981).
Computer modeling studies have shown that, at least in theory, spermine, interacts with
DNA by binding within the major groove of the double helix (Feuerstein et al, 1991).
This binding reduces the size of the groove presumably by reducing the repulsive force at work between the phosphate-containing backbones of the molecule.
The physiological function of polyamine/nucleic acid interaction is less understood than the actual mechanism of interaction. Spermidine and spermine stabilize the structure of t-RNAs as well as increase the efficiency of aminoacylation of these molecules (Lbvgren, et al, 1978). Polyamines have long been implicated as an essential part of ribosomes (Cohen and Lichtenstein, 1960). Spermine increases the rate of protein synthesis primarily by effecting the codon-directed binding of charged tRNAs to the A and P sites of the ribosome (Kudan, et al., 1988). Spermidine is known to interact with proteins that act upon nucleic acids. These proteins are DNA gyrase, which is stimulated by spermidine, and ©, which is inhibited by spermidine (Gellert, et a l, 1976). This is interesting as gyrase introduces negative supercoiling while © removes these supercoils.
Polyamines can interact with proteins by other than allosteric means. Several instances of the covalent attachment of polyamines to proteins are known. One of the better understood mechanisms is the synthesis of the novel amino acid residue hypusine, an amino acid that has been found in only one known protein—eucaryotic initiation factor
4D (eIF-4D) (Park, et a l, 1991). This amino acid residue, which is created post- translationally, is derived from the reaction of spermidine and lysine. In this reaction spermidine is metabolized to diaminopropane and 4-aminobutyraldehyde. The butylamine moiety is covalently attached via a Schiff base reaction to a lysyl residue to form deoxyhypusine which is then hydroxylated to hypusine. As the actual function of eIF-4D is poorly understood, it is very difficult to determine the role of this novel amino acid. It has been postulated, however, that the synthesis of hypusine, which closely parallels the state of proliferation of the cell, may be a polyamine regulated control point.
Another type of covalent attachment of polyamines to proteins is known.
Putrescine, spermidine, and spermine can all be covalently attached to glutamyl residues in proteins (Beninati and Folk, 1988). These modifications are formed by a series of enzymes known as transglutaminylases. Polyamines are transferred to the y-carboxyl moieties of specific glutaminyl residues and serve as precursors for the formation of protein crosslinks. The crosslinks formed through this modification are e-(y-glutamyl)lysine links. Although the function of these crosslinked amino acid residues in unknown, it is known that they are found to a greater degree in proteins isolated from epidermal envelopes. Because of this localization, it is postulated that e-(y-glutamyl)lysine crosslinks serve a structural role similar to the hydroxyproline- derived crosslinks found in collagen or the desmosine crosslinks of elastin (Zubay,
1983).
Other less characterized interactions between polyamines and proteins include the inhibition of adenylate cyclase by interaction with the inhibitory GTP-binding protein (Gj) (Clo, e ta l, 1988), the activation of protein kinase C (Moruzzi, e ta l, 1988), and regulation of uptake, transport, and synthesis of polyamines themselves (Hem&idez-
Yago, 1988). In the case of adenyl cyclase and cAMP synthesis, spermine negatively modulates synthesis of cAMP both in vivo and in vitro in a chick embryonic heart muscle system (Clo, et al., 1988). Feedback regulation of polyamine synthesis will be discussed in Section 1.1.3. Protein kinase C (PKC) exists both as a soluble enzyme and in a membrane-bound form. PKC is a key regulatory enzyme, acting in the secondary messenger pathways necessary for stimulation of proliferation and differentiation (Rubin 8 and Rosen, 1975). The enzyme interacts with phospholipids as part of its role in the message pathway. Moruzzi, et al (1988) demonstrated that spermine at physiological concentrations was inhibitory to the activity of membrane-associated PKC. Although no direct evidence supports the observations made in this study, it is thought that spermine may interfere with the binding of PKC to acidic phospholipids in the membrane.
In vitro, spermine specifically associates with phospholipid liposomes (Tadolini and Hakim, 1988). This interaction is dependent upon the type of phospholipid, with the strongest association being between spermine and phosphatidic acid and the weakest with cardiolipin. In vivo polyamines may act to neutralize strong negative charges on these lipids, a function that could not be fulfilled by inorganic cations due to the hydrophobic environment. Because there appears to be a certain specificity to the binding of spermine to various phospholipids, a clustering of specific lipid molecules may result.
Polyamines may also be part of the osmoregulatory mechanism of the cell. From very early on Tabor (1960) recognized the potential for spermidine and spermine to prevent the swelling of isolated mitochondria suspended in hypotonic solutions.
Polyamines act in a protective manner against a variety of mitochondrial damaging agents (Toninello, et al, 1988). Spermine also protects against damage by high concentrations of Ca++ and phosphate (Toninello, et al. 1984), and exposure to hydroperoxides. The actual mechanism of this protective activity is unknown but is thought to be involved with neutralization of membrane anionic charges.
1.1.2. Polyamlne synthesis
The most commonly encountered pathway for the biosynthesis of polyamines begins with the synthesis of putrescine from the amino acid ornithine (Fig. 1). In all systems that have been examined the enzyme necessary for this step, ornithine Figure 1. Polyamine biosynthetic pathway for A castellanii (Kim, etal., 1987a). The enzymatic steps necessary for the production of norspermidine are speculative and indicated by a single question mark.
9 NH ornithine COOH CO,
ornithine decarboxylase dcSAM putrescine OHOH
spermidine synthase
MTA
OH OH spermidine HgN
polyamine oxidase or spermidine dehydrogenase
-NHg ^ h2n ' NHg dcSAM ^ diaminopropane h^N IJQ S - C H j HjO
NH
I J L S> H jC -S -C tfe MTA norspermidine H2N decarboxylase, occurs (Tabor and Tabor, 1984). In E. coli and plants an alternate
pathway for the synthesis of putrescine exists. This pathway, which forms putrescine from arginine, involves two enzymes in procaryotes, arginine decarboxylase which catalyzes the formation of agmatine, and agmatine ureohydrolase which forms putrescine from agmatine (Bowman, et al., 1973). In plants agmatine is not directly converted to putrescine but alternatively N-carbamoylputrescine is formed by agmatine iminohydrolase (Hiatt, etal., 1986). Putrescine carbamoyl transferase then converts the
N-carbamoylputrescine to putrescine and carbamoyl phosphate. Higher polyamines are synthesized from putrescine.
Spermidine is synthesized from putrescine and the aminopropyl donor
(dcSAM=decarboxylated S-adenosylmethionine) S-adenosyl-5'-3-methylthiopropylamine
(Fig. 1). The enzyme which catalyzes this reaction is spermidine synthase (Bowman, et al., 1973). Spermine is synthesized from spermidine and dcSAM by spermine synthase.
Both spermidine and spermine synthases are constitutively expressed in all organisms that synthesize the respective polyamines (Tabor and Tabor, 1984).
The key rate-limiting enzymes of polyamine biosynthesis are ornithine decarboxylase (ODC) and S-adenosyl-L-methionine decarboxylase (SAMDC) (JSnne, et al., 1978; Jhnne, et al., 1985). ODC is a pyridoxal phosphate-requiring enzyme. It has been purified to homogeneity from a number of sources including E. coli (Tabor and
Tabor, 1984b), Physarum polycephalum (Mitchell, eta l, 1988), Mus, Rattus, and Homo sapiens (Kameji and Pegg, 1987a), and S. cereviseae (Tyagi, etal., 1981). The enzyme is inducible in most systems and is negatively modulated by the product of the reaction putrescine. Unlike ODC, SAMDC is a pyruvoyl-containing enzyme. The pyruvoyl moiety is generated in the mature enzyme by the autocatalytic cleavage of a proenzyme into two peptides (Pegg, 1984). Procaryotic SAMDC requires Mg++ for optimal activity 12 whereas most eucaryotic SAMDCs do not require divalent cations (Tabor and Tabor,
1984b). Eucaryotic SAMDCs are stimulated by, but do not require putrescine. The genes for both ODC and SAMDC have been cloned from procaryotes (E. coli) (Tabor and Tabor, 1987), and a variety of eucaryotes (Mach, etal., 1986; Pajunen, etal., 1988;
Balasunduram, etal., 1991).
Other less commonly encountered polyamines include cadaverine
(1,5-diaminopentane), diaminopropane, norspermidine, as well as N-acetylated derivatives of most polyamines. Cadaverine, which appears to be specifically produced by procaryotes, is synthesized from lysine by lysine decarboxylase. Unlike the major polyamines, putrescine, spermidine, and spermine; cadaverine is only produced under select conditions (Boeker and Fisher, 1983). Synthesis of the other polyamines has not been well studied. Diaminopropane can be produced from spermidine or spermine by polyamine oxidase or spermidine dehydrogenase with the concomitant production of
A^-pyrroline (Tabor and Tabor, 1984a) (Fig. 1). Although no reports on the synthesis of norspermidine could be found, it is thought that this polyamine could be synthesized from diaminopropane by aminopropylation (Kim, etal., 1987a).
N-acetylated derivatives of the major polyamines occur at relatively low concentrations in most living systems under conditions of normal growth (Pegg, et al.,
1981). The actual function of these derivatives is unknown, but specific enzymes for the formation of the N-acetylated forms exist (N1 -acetyltransferases). These enzymes appear to have a great deal of substrate specificity and are rapidly turned over, thus, indicating that there may be a specific function for these derivatives. One possible use for N-acetylated polyamines is that they represent both a storage form as well as an excretory form as they greatly increase in concentration when polyamine synthesis is induced (Oka, etal., 1981). 13
In addition to the previously mentioned polyamines many polyamine and polyamine derivatives have been found that are specific to a particular genus of organisms. These include thermine, thermospermidine, caldine, sym-homospermidine, homospermidine, caldopentamine, caldohexamine, homocaldopentamine, and homocaldohexamine, all of which are found in extremely thermophilic organisms such as
Thermus spp. (Oshima, etal., 1988). In trypanosomes, two novel polyamine derivatives have been found. These polyamines are N1 -glutathionylspermidine and
Nl-N^-bis-(glutathionyl)spermidine (trypanothione) (Fairlamb, 1988; Fairlamb, etal.,
1986). Both of these compounds are derivatives of glutathione and spermidine. In trypanosomes it is believed that these compounds function as superoxide scavengers.
1.1.3. Regulation of polyamine biosynthesis
Ornithine decarboxylase is the best characterized enzyme, in terms of regulation, of all polyamine biosynthetic enzymes. The regulation of ODC activity occurs at the transcriptional and translational levels as well as post-translationally in mammals (Heller, et al., 1976; Hayashi, et al., 1988). ODC is rapidly turned-over—exhibiting a half-life on the order of 10 - 30 minutes. The enzyme can also rapidly accumulate after a variety of stimuli to levels 10 to 600-fold higher than in unstimulated cells (Pajunen, etal., 1982).
Conditions that stimulate ODC accumulation in mammalian systems are: androgen stimulation in mammalian renal tissue (Pajunen, etal., 1982), relaxation of serum starvation (Morris and White, 1988; Morris, 1991), activation by mitogens and hormone stimulation (Janne, et al., 1988). Nearly every condition which stimulates growth in mammalian cells will, to some degree trigger a stimulation of ODC production.
The rapid changes in ODC activity are made possible by both the rapidity with which the gene for the protein can be induced as well as the rapid turn-over of the protein. ODC turnover is in part regulated by the ODC antizyme (Panagiotidis, et al., 14
1988; Hayashi, etal., 1988; Brosnan and Wu, 1988). ODC antizyme is a small, acidic protein that has been found in both procaryotic and eucaryotic systems. ODC antizyme production is induced by polyamines (Heller, et al., 1976). It rapidly complexes with
ODC, inactivating the enzyme, and leading to a rapid degradation of the protein. ODC antizyme itself appears to be regulated by another small protein which has been termed antizyme inhibitor (Fujita, et al., 1982). Antizyme inhibitor is induced in a fashion parallel to ODC induction and acts to prevent and reverse the formation of ODC-ODC antizyme complexes. The production of ODC is also regulated at the translation level.
Polyamines specifically decrease the translational efficiency of the ODC mRNA by at least 90% of control levels in cultured mouse cells (Kahana and Nathans, 1985). In addition to the previously mentioned regulatory mechanisms there is limited evidence for regulation of ODC by phosphorylation. In the slime mold P. polycephalum, there is evidence for a polyamine stimulated ODC specific protein kinase (Atmar and Kuehn,
1981). Additionally, there is in vitro evidence for the phosphorylation of ODC by casein kinase II (Meggio, etal., 1984) and dephosphorylation by specific phosphatases
(Mitchell, et al., 1988). Less is known concerning the regulation of SAMDC, the second major control point in polyamine biosynthesis. It is the only known eucaryotic enzyme to have a covalently-bound pyruvoyl group (Anton and Kutney, 1987b). In mammalian systems the cellular content of SAMDC appears to be regulated four ways: 1. mRNA synthesis, 2. protein synthesis, 3. conversion of the proenzyme to the active enzyme, and finally 4. turnover (Pegg, et al., 1988b). SAMDC is turned-over at a reasonably high rate (ti/2=l-
2h) although not as rapidly as ODC (Pegg, 1984). During induction of polyamine synthesis or stimulation of SAMDC by inhibition of ODC, mRNA levels for SAMDC increase approximately 4-fold while the rate of synthesis of the enzyme increases 15
approximately 10-fold (Mach, et al., 1986; Pegg et al., 1988c). An increase in the size of
SAMDC mRNA-containing polysomes is observed as well during stimulation. It has
also been shown that the rate of conversion of SAMDC proenzyme to mature SAMDC is
stimulated by putrescine (Shirahata and Pegg, 1986). To this date, no SAMDC
antizyme-like activity has been reported in the literature.
SAMDC, in addition to being essential to the biosynthesis of polyamines,
provides a metabolic bridge to yet another part of metabolism—that involving
S-adenosylmethionine (SAM) (Fig. 2). SAM metabolism is critical to the cell because it
provides substrate for single carbon reactions. SAM can donate methyl groups for the
methylation of both nucleic acids and proteins (Zubay, 1983). Any perturbations in the
metabolism of SAM could have multiple effects within living systems (Zhu, et al.,
1989).
1.1.4. Polyamine function and synthesis as targets of chemotherapy
Because polyamines appear to be specifically needed for proliferation, the polyamine metabolic pathway was thought to be suitable as a target for chemotherapy.
The most commonly used chemotherapeutic agent that specifically targets this pathway is a-difluoromethylomithine (DFMO). DFMO is a suicide inhibitor of ODC (Metcalf, et al., 1978). Bacchi et al. (1981) discovered that DFMO can totally cure Trypanosoma brucei brucei infections in mice. This discovery led to the rapid adaptation of DFMO in the treatment of human trypanosomiasis and provided cures in what would have previously been fatal infections (Schechter et al., 1987). The efficiency of DFMO treatment in trypanosomiasis may be due to the trypanosome specific polyamine derivative trypanothione (Fairlamb etal., 1985). This novel compound has only been found in trypanosomes, appears to be essential for their survival and the synthesis of tryptathione is very sensitive to DFMO inhibition. DFMO is also useful as a Figure 2. SAM metabolism adapted from Zubay (1983). SAMDC is highlighted by underscoring. Specific transmethylation reactions are not shown.
16 methionine S-adenosylmethionine S-adenosylhomocysteine COOH , . . COOH NHo COOH NH2 h2n - c h nfbonme ' 1 H2-Ch H
SAM decarboxylase c o .
decarboxylated SAM homocysteine COOH NH-, 1 h2n - c h HgC CH2 6 3 CHp1 £ SH
H2C ^Serine cystathionine- OH / OH transaminase / ft-synthase spd or norspd synthase HO y cystathionine CQOH
HoN-CH* 1 methylthioadenosine NHp c h 2 CHp N L L N> N c h 2 H3C -S -C H 2 h2n - ch COOH
OH OH a-ketobutyrate phosphorylase cystathionine- l-phospho-4- "*■ adenine f lyase methylthioribofuranoside 0 c h 2 - s - c h 3
0 V cysteine O-P-O^ COOH OH OH NH, HoN-CH^ 1 CHp1 * OII methyithio-a- SH h3c - s - c h 2 - c h 2 - c - c o o h ketobutyrate
Figure 2. 18
chemotherapeutic agent in mammals because of its low toxicity. Mammals, including
humans are able to tolerate high doses of the drug, on the order of 100 uM bloodstream
concentrations, with no lasting side effects (Bacchi and McCann, 1987). Unfortunately,
many pathogenic organisms are similarly unaffected by DFMO. Even protozoans as
closely related to T.b. brucei as Trypanosoma cruzi are relatively unscathed by exposure
to DFMO (Bacchi and McCann, 1987). However, because inhibition of polyamine
metabolism appears to have the greatest effect on rapidly proliferating cells the search for
more effective inhibitors of the pathway continues.
Other inhibitors of ODC have been synthesized. These inhibitors include
ornithine analogs, putrescine analogs, as well as transition state analogs (Edwards, et al,
1991). By far ODC is the most sensitive to transition state analogs such as l-aminooxy-3-aminopropane (AOAP) (Kramer, etal., 1989). Inhibition constants (Kj)
for these transition state analogs are on the order of 10-100 nM, which is 100-1000 times
more effective than any of the ornithine or putrescine derivatives tested (Bey, et al.
1978). Unfortunately the most potent of these inhibitors, AOAP, has significant
detrimental side effects in vivo (Hyvonen, et al, 1988).
Inhibition of SAMDC has also been targeted as a means of chemotherapy. Much
of the research into SAMDC inhibition has been in relationship to cancer chemotherapy.
The first known inhibitor of SAMDC tested for use as an antineoplastic agent was methylglyoxal bis(guanylhydrazone) (MGBG) (Williams-Ashman and Seidenfeld, 1986).
MGBG was originally used as an antileukemia agent and later shown to be a potent inhibitor of SAMDC. Treatment of tumor cells with MGBG leads to derangement of organellar structure followed by cell death (Porter et al., 1979). In many cases, however,
MGBG is not immediately cytotoxic. MGBG has been shown to stabilize SAMDC in some cell lines (Williams-Ashman and Pegg, 1981) and lead to an increase in putrescine 19
levels (Williams-Ashman and Seidenfeld, 1986). MGBG is actively transported into
mammalian cells by the same mechanism as polyamines and can accumulate to relatively
high concentrations (Pegg, 1986). The drug is also known to have other targets than
SAMDC, including polyamine aminotransferases and diamine oxidase (Pegg, et al.,
1986).
Chemotherapeutic diamidines were originally developed as antiprotozoal agents
(Newton, 1975). These agents, which include berenil, pentamidine and propamidine, are
structurally related to ethidium bromide (EB) and MGBG. Both berenil and ethidium
bromide were originally thought to function mainly as inhibitors of organellar nucleic
acid metabolism as both agents have been shown to inhibit mitochondrial DNA and RNA
synthesis (Newton, 1975; Waring, 1975). Pentamidine, EB, and MGBG later were
shown to inhibit the synthesis of spermidine from putrescine in Leishmania by Bachrach,
et a l (1979). These data suggested that these compounds interfered with polyamine biosynthesis. Bitonti, et al. (1986) first demonstrated that the diamidines were potent irreversible inhibitors of SAMDC. Recently it has been shown that these inhibitors are not specific for SAMDC; for example, diamine oxidase is 100-fold more sensitive to these berenil than SAMDC in mammals (Balana-Fouce, et al., 1986a; Balana-Fouce, et al., 1986b).
1.2. Acanthatnoeba
1.2.1. General characteristics
Acanthamoeba is a genus of small free-living amoebae. The amoeba typically ranges from 15-40 urn in diameter when adherent to a substrate. Usually, the amoebae have a single nucleus, 4-8 (un in diameter, with a well defined nucleolus. Multinucleate cells occur with varying frequency depending on the growth state of the organism (ie. log 20
vs stationary phase and monolayer vs suspension culture). Members of the genus
characteristically produce spike-like pseudopods called acanthapodia and broad
pseudopodia referred to as Umax pseudopods (Neff, 1957; Page, 1967). The amoeba
occurs in two physical forms—the dormant cyst and the active trophozoite.
Acanthatnoeba castellanii was first isolated by Dr. A. Castellani as a contaminant in a
Cryptococcus culture (Volonski, 1931). The most commonly encountered laboratory strain of this species is the Neff strain. A. castellanii Neff was originally isolated from a
soil sample in Pacific Grove, CA (Neff, 1957).
The amoeba has been isolated from numerous natural sources throughout the world including: soU, fresh, brackish and salt water, the surface of vegetables and fungi, and hot springs (Sawyer, et al., 1977). The organism has also been found Uving in association with the products of civiUzation such as in hot tubs, swimming pools, power plant (both nuclear and conventional) cooling towers, air conditioning units, sewage, and various medical equipment (Samples, etal., 1984; DeJonkheere, 1991). Acanthatnoeba has been isolated from apparently healthy individuals in nasal and throat swabs. The organism has also been isolated from patients with respiratory disease (bronchial secretion), ear infections (auditory canal), and from fecal samples from patients with diarrhea (Cerva, etal., 1973; Lengy, e ta l, 1971; Wang and Feldman, 1961). More recently the amoeba has been isolated from contact lens care solutions and lens storage units (Stehr-Green e ta l, 1987), eye wash stations (Paszkokolova, etal., 1991), and intrauterine devices (Arroyo and Quinn, 1989).
Identification of Acanthamoeba has been problematical. While classification of the amoeba as to genus can be accomplished using morphological characters, identification of species as well as strain is more difficult (Page, 1967; Stratford and
Griffiths, 1978). Originally strain identifications were based on morphological 21 characteristics such as structure of the mitotic spindle, cyst morphology, and nuclear morphology. Identification based on morphology is subject to errors, however, due to the great deal of variability caused by growth conditions (Griffiths, et al., 1978). More recently identification of the organism has been accomplished using immunofluorescent microscopy with specific antiserum or monoclonal antibodies (Willaert and Stevens,
1976; VisvesvaraandBalamuth, 1975; Visvesvara, etal., 1983).
Acanthatnoeba can be cultivated in a number of ways. For isolation of the organism from the environment, the amoeba is typically grown as monoxenic cultures using Escherichia coli or Enterobacter aerogenes as feeder organisms on a low nutrient agar medium (Richards, 1968; Hugo, et al., 1991). Cultures of the organism can be adapted to axenic growth in a variety of semidefined media (Neff, 1957).
1.2.2. Developmental biology
The use of A. castellanii as a model laboratory organism for research into aspects of eucaryotic cell differentiation began with the work of R. J. Neff and associates (Neff, etal., 1964). Because of the organism's simple life cycle and ease in cultivation, it was chosen to examine the mechanism of cell differentiation. Evidence of cellular differentiation in this organism is very simple to quantitate in the laboratory due to the gross morphological differences between the metabolically active trophozoite and the dormant cyst. Differentiation can be induced in the laboratory by a number of means—the simplest being starvation (Neff and Neff, 1969). When actively growing amoebae are placed in an isoosmotic nonnutrient aqueous environment differentiation is typically induced. Transformation from trophozoite to cyst can occur in 24 hours, but typically requires 72-120 h for maximal encystment to occur in the Neff strain (unpublished observation). Under optimal conditions, 70-90% of the trophozoites will synchronously 22 encyst. Cell lines can be selected by cloning for optimal response to encystment-inducing conditions (Akins and Byers, 1980; Akins, et al., 1985).
A variety of other stimuli can induce encystment. Low oxygen concentrations, accumulation of metabolic wastes, and changes in osmolarity can also induce differentiation in the amoeba (Neff and Neff, 1969). A variety of inhibitors of various metabolic processes can induce encystment (reviewed in Byers, etal., 1990; Byers 1979;
Neff and Neff, 1972). DNA and RNA synthesis inhibitors such as fluorodeoxyuridine, mitomycin C, methotrexate and actinomycin D can cause the amoeba to encyst. Specific inhibitors of mitochondrial protein synthesis such as chloramphinicol and erythromycin can also induce encystment (Akins and Byers, 1980). Compounds which can perturb polyamine biosynthesis and function, in particular the diamidines berenil, pentamidine, propamidine, and hydroxystilbamidine, will induce differentiation in Acanthatnoeba
(Akins and Byers, 1980; Kim, et al., 1987b). Both energy metabolism and cytoplasmic protein synthesis appear to be necessary components of the encystment process.
Inhibitors that block ATP production such as cyanide and azide or compounds blocking protein synthesis on 80s ribosomes such as emetine, chloramphinicol, and G 428 prevent encystment and lead to cell death (Neff and Neff, 1969; Akins and Byers, 1980; E. Hugo, unpublished results).
The response of cells to encystment-inducing stimuli is variable and is dependent on a variety of conditions including growth phase, population density, and nutrient conditions as well as the cell line used (Akins, 1981, Ph.D dissertation). Akins and
Byers (1980) observed that berenil would only induce 10-15% encystment in low density cultures whereas 70-90% of the cells in high density exponential phase cultures would encyst in response to this drug. Cell-free culture medium from high density cultures contained a low molecular weight factor, that while not a direct inducer of encystment, 23 could enhance the response of low density cultures to berenil. This compound (termed
EE A for encystment enhancing activity) has not been identified but was partially characterized by Akins and Byers (1980). Attempts to completely characterize this factor have been stymied by the instability of strains used in the bioassay for EEA (Akins, et al., 1985). Clones that have been selected for development of high levels of encystment in response to EEA and berenil lose the ability to respond well in a short period of time
(1-6 months).
Attempts to analyze the cell cycle in A. castellanii, particularly with respect to commitment to development have yielded conflicting data. Neff and Neff (1972) originally reported that in observation of pairs of recently divided single cells only cells that had completed 15-80% of the cell cycle were competent to encyst. From these results it was concluded that encystment competency occurred sometime during S-phase, based on the observation that the first 10-20% of the cell cycle is spent in Gj. Evidence for the existence of a specific point within the cell cycle from which cells are competent to undergo development (developmental transition point- DTP) was provided in a study utilizing flow cytometry (Stdhr, et al., 1987). In this study nuclear DNA content was measured in order to determine where in the cell cycle members of a population were. In this study it was shown that less than 10% of an exponentially growing population of the amoebae were competent to encyst under conditions of starvation. When stationary cultures were used, however, 50-70% of the cells encysted. It was also shown that when stationary phase cultures were diluted into fresh medium, a 4 hour lag followed by synchronous division of the cells occurred. These two observations tend to support the hypothesis that there is a specific DTP in the cell cycle of Acanthamoeba. 24
1.2.3. Polyamine metabolism
The relationship of polyamine metabolism to growth and development in the
amoeba has been examined. Induction of encystment by diamidines and ethidium
bromide provided the first indication that polyamines might be involved in this process.
Dr. S. Hutner (personal communication to Byers) was the first person to point out that in
addition to effects on organellar metabolism, berenil was known to be a potent inhibitor
of polyamine biosynthesis. Kim, etal. (1987a) analyzed the polyamine content and the
biosynthetic enzymes used by A. castellanii (Fig. 1). This study confirmed the finding
by Poulin, et al. (1984) that the most abundant polyamine in the amoeba is
diaminopropane. Kim, et al. (1987a) also found trace amounts of putrescine and
norspermidine as well as significant amounts of spermidine. Spermine was only found in
amoebae grown in OGMA and thus may be taken up from the medium rather than
synthesized. A. castellanii had both ODC and SAMDC activities (Kim, etal., 1987a;
Hugo and Byers, 1989).
The effects of inhibitors of specific enzymes of polyamine biosynthesis in
Acanthamoeba have been tested in several laboratories. Gupta, et al. (1987a) reported
that a-methylomithine, a competitive inhibitor of ODC enhanced starvation-induced
encystment in Acanthamoeba culbertsonii. Kim, et al. (1987b) examined the effects of a
variety of specific inhibitors on the induction of encystment. This study found that
specific inhibition of ODC by a number of compounds (DFMO, R,R'-MAP,
A-monofluoromethylomithine methyl ester) was not in itself sufficient to induce encystment. These compounds would block multiplication and did reduce cellular polyamine levels, however, encystment was only seen when the cultures were supplemented with 30 mM Ca++ or Mg++. Additionally, polyamines added to the medium could restore multiplication in ODC inhibited cultures. This finding was in 25 contradiction to that of Gupta, et al. (1987b), but the difference in results may be due to
differences in experimental technique. Kim, et al. (1987b) tested a variety of SAMDC inhibitors. He found that all SAMDC inhibitors except for MGBG were capable of inducing encystment. More importantly, Kim, et al. (1987b) found that exogenously added polyamines were able to block induction of development but were unable to restore multiplication in these experiments. These results indicate that multiple targets may be involved in polyamine inhibitor-induced encystment.
1.2.4. Molecular biology
Very little is known about the molecular biology of Acanthamoeba as compared to other model organisms such as Saccharomyces or Dictyostelium. This is in part due to the lack of a suitable system for genetic study (recently reviewed in Byers, et al.,
1990). Evidence for polyploidy has accumulated but has not been conclusively shown.
Kinetic complexity measurements combined with measurements of nuclear DNA content tend to indicate that the organism possesses a 12-25n genetic complement. Also, it has been observed that mutations in nuclear genes are extremely difficult to isolate and very unstable (Akins, Ph.D. dissertation 1979). Observation of chromosomal content in the organism has been hindered by the extremely small size of the chromosomes and the lack of clearly defined structures in microscopic preparations (reviewed in Byers, 1986).
Recently, attempts to analyze chromosomal complements by pulse-field gel electrophoresis and clamped, homogeneous electrophoresis have been made (Rimm, et al., 1988; Jantzen, H. and I. Shultz, abstr. Vth International Conference on the Biology and Pathogenicity of Free-Living Amoebae). Although separation of chromosome-sized
DNAs was achieved, it was unlikely that all of the chromosomes were resolved by the techniques used based on the observation that some DNA molecules were unresolved
(DNA remained in sample wells). 26
By far the best characterized genes of Acanthamoeba are those for contractile proteins and rRNAs (reviewed in Byers, et ah, 1990). Acanthamoeba has been used as a model organism for the study of nonmuscle movement. The cDNAs for actin, and actin binding proteins such as myosin and a variety of myosin-like proteins, as well as profilins, have been cloned and sequenced (Nellen and Gallwitz, 1982; Hammer, et al.,
1987; Pollard and Cooper, 1986). The main reason Acanthamoeba has been used in the study on nonmuscle motility has been the ease in preparation of large quantities of biological material (Hammer, personal communication).
The ribosomal RNA genes of Acanthamoeba have been well characterized. The
18s, 5.8s, and 5s genes have been cloned and sequenced from a number of strains
(D'Alessio, etal., 1981; Gunderson and Sogin, 1986; MacKay and Doolittle, 1981; Gast and Byers, unpublished results). Paule, et al., (1991) has examined the mechanism for transcriptional control of the rRNA transcription unit in some detail (Kownin, et al.,
1987; Kownin, et al., 1988). In addition to analysis of the promotor regions of the rRNA genes, Paule, etal. (1984), have also characterized RNA polymerase I from this organism.
1.2.5. Pathogenicity
The discovery of amoebae as etiological agents in human disease occurred in
1875 when Fedor Aleksandrovich characterized the obligate parasite Entamoeba histolytica as the cause of a human dysentery (Martinez, 1985). In 1892, small free- living amoebae were implicated as the cause of a jawbone abscess in a 60-year-old
Virginia woman (Flexner, 1892). Unfortunately, the techniques needed to isolate and cultivate pathogenic amoebae were not available to researchers in these early cases and positive identification was not possible. Early work in proving the pathogenicity of small free-living amoebae also was complicated by the misidentification of leucocytes as 27
amoebae (Culbertson, 1971). It was not until the mid-1950s that strong evidence was
obtained supporting the role of small free-living amoebae in human disease. Jahnes, et
al (1957) discovered cytopathic effects in African Green Monkey kidney cell cultures
caused by small free-living amoebae. Culbertson (1958) discovered, as part of a routine
safety check of cultures used to produce polio vaccine, that contaminated media from
these cultures could destroy central nervous system (CNS) tissue. In this safety check,
Culbertson inoculated mice and Rhesus monkeys intravenously, intranasally, and
intracerebrally with the suspected culture media. All of the test animals died of a severe
meningoencephalitis. Microscopic examination of the media and tissues from the test
animals revealed trophozoites and cysts of Acanthamoeba (Culbertson, 1958).
A. castellanii was later identified as a contaminant in human tissue cultures (Chi,
et al., 1959). Human disease was subsequently linked to this amoeba by several
independent groups. Small, free-living amoebae were isolated from the throat and nasal
mucosa of healthy individuals. The amoebae isolated were cytopathic to human cells in
vitro (Wang and Feldman, 1961; Chang, Pan, and Rosenau, 1966), suggesting a
pathogenic potential against humans. The first documented case of infection in humans
by small free-living amoebae was reported in 1965 and 1966 (Fowler and Carter, 1965;
Butt, 1966). These investigators described an acute, fatal meningoencephalitis. The etiological agent in this disease was identified as Naegleria, a relative of Acanthamoeba.
This led to the discovery of two distinct diseases associated with small free-living amoebae, Primary Amebic Meningoencephalitis (PAM) and Chronic Granulomatous
Amebic Encephalitis (GAE) (Willaert, et al., 1978; Martinez, 1985). PAM is an acute, usually fatal, encephalitis caused by Naegleria sp.. The disease occurs in seemingly healthy young individuals after swimming in fresh water. Occurrences of the disease have been reported world-wide. The mode of entry for Naegleria is through the nasal 28
neuroepithelium. GAE, the Acanthamoeba-associated CNS disease, occurs in
immunocompromised persons. The disease is characterized by a chronic encephalitis.
Individuals at risk of acquiring GAE include patients receiving chemotherapy or radiotherapy, splenectomy patients and allograft recipients (Borochovitz, etal., 1981).
Several cases of the disease have been seen as complications to the Acquired
Immunodeficiency Syndrome (AIDS) (Ma, 1987). As of 1990, a total of ten cases associated with AIDS patients had been reported to the CDC (Visvesvara and Stehr-
Green, 1990). By far, the majority of cases of GAE are diagnosed during postmortem examination of brain tissue (Willaert, et al., 1978). Post-mortem examination of GAE fatalities reveals a classical granulomatous reaction with syncytia production in the affected tissues. As of 1989,27 cases of GAE had been reported in the United States and
50 cases had been reported world-wide (Martinez, 1991). The mode of entry of
Acanthamoeba in GAE is through the lower respiratory tract (Martinez, 1985). Since most cases of GAE are diagnosed only upon autopsy, very little is known as to the efficacy of chemotherapy (Martinez, 1991).
Although Naegleria is only known to cause PAM, Acanthamoeba has been associated with human diseases other than GAE. Infections of the ear (Lengy, et al.,
1971), uvea (Jones, e ta l, 1975), and jawbone (Visvesvara, etal., 1983) have been reported. Several cases of Acanthamoeba keratitis, an eye infection, acquired from hot tubs and swimming pools have been reported (Jones, et al., 1975; Key, et al., 1980;
Samples, et al., 1984). Between April of 1985 and August of 1986 40 cases of
Acanthamoeba keratitis were reported to the Centers for Disease Control (Givens and
Underwood, 1986). In twenty-five of the thirty confirmed cases (83%) the patients used contact lenses. Keratitis may be increasing in frequency as only 31 cases were documented in the previous decade. The amoebae, which were identified by 29
morphological characteristics and in some cases indirect immunofluorescence using
species-specific antisera, were identified as A. castellanii, 9 cases; A. polyphaga, 8 cases;
A. rhysodes, 4 cases; A. culbertsoni, 3 cases; A. hatchetti, 1 case; and 6 cases in which the
amoebae were not identified as to species (Centers for Disease Control, 1986). By the
end of 1988 (the most recent available data) Visvesvara and Stehr-Green (1990) reported
a total of 208 confirmed cases of keratitis in the United States. The geographic incidence
of Acanthamoeba keratitis reported in this study was unusual in that the majority of the
cases were reported from California, Texas, and Pennsylvania. There were no factors
that would explain this geographic distribution such as temporal clustering that would
have suggested a common source of infection such as contaminated contact lens care
solutions. Visvesvara and Stehr-Green (1990) suggest that the great number of reports
from these states might be due to better diagnosis of the disease rather than an actual
increase in the number infections since these states have active eye research centers.
Acanthamoeba keratitis is a very serious condition which often leads to permanently
damaged vision. Previously, the only successful treatment of the disease was with
penetrating keratoplasty (corneal transplant). Recently, however, ketoconazole,
miconazole, propamidine, and pentamidine have been successful in affecting a cure in a small number of cases (Hirst, et al., 1984; Osata, et al, 1991). Probably the most
significant problem with amebic keratitis is the failure of the medical community to correctly diagnose the disease. Often Acanthamoeba keratitis is diagnosed as Herpes-
Simplex keratitis and treated as a viral infection. This only worsens the condition and leads to permanent damage to the eye.
Most cases of acanthamoebiasis are diagnosed either by autopsy or biopsy of living tissue. Trophozoites in the CNS fluid or purulent exudates from diseased tissues can be cultured. These trophozoites can be identified as Acanthamoeba by indirect 30 immunofluorescence (Willaert and Stevens, 1976). Circulating complement-fixing anti -Acanthamoeba antibodies are found in diseased individuals, but are not a basis for a valid test as most normal individuals possess these antibodies (Visvesvara, et al., 1977). MATERIALS AND METHODS
2.1. Cell culture and growth 2.1.1. Cell lines
Acanthamoeba castellanii strain Neff was used for all experiments. This strain
was originally isolated from a soil sample from Pacific Grove, CA by Dr. R.J. Neff
(Neff, 1957). Two clonal isolates were used, clone OS4-7B and clone NIH-BB. Clone
OS4-7B (hereafter called 7B) was subcloned through several generations from an
original clone (112) sent to the Byers’ lab by Dr. Neff in 1967. Clone 7B was isolated by
screening for maximal encystment in the presence of the drug diminazine aceturate
(berenil) (Akins and Byers, 1980). This strain grows well both in defined (DGM-11) and semidefined (OGMA) growth media. The 7B strain adheres well to plastic surfaces and grows well in monolayer cultures, however, the growth rate drops considerably when it is grown in suspension culture. Typical generation times are = 12h for monolayers, and =
18h for suspension cultures. NIH-BB is another subclone of the Neff strain. This subclone was obtained from Dr. B. Bowers at the National Institutes of Health, Heart,
Lung, and Blood Institute, and was selected for its ability to grow to high cell densities in suspension culture. Typically, the NIH-BB strain will grow with a generation time s 14h and to a final cell density approaching 1 X 10^ cells per ml in suspension in a semidefined medium (NIH medium). For experiments requiring large quantities of cells, such as the purification of enzymes, the NIH-BB strain is ideal. NIH-BB does not grow well in defined medium and does not adhere well to plastic tissue culture flasks.
31 32
2.1.2. Media
Three different media formulations were used for cell cultivation. Two of these formulations were used for cell growth while the third was a starvation medium used for encystment induction. Defined growth medium, DGM-11, contains eleven amino acids, glucose, and salts (Byers, et al., 1980; Appendix A). This medium was originally selected because it does not contain exogenous polyamines. It was thought that exogenous polyamines might reduce the levels of SAMDC, however, it was later discovered that there was no significant difference for SAMDC found in amoebae grown in either defined or semidefined media. For the first two years of this project our modification of Neffs Optimal Growth Medium designated OGMA was used as a rich, semidefined growth medium (Appendix A). The medium was prepared according to the method of Byers, Rudick, and Rudick (1980) with one change-the amount of yeast extract was reduced by 50% for economy. This medium also appears to contain at least spermine and possibly other polyamines when analyzed by HPLC (Kim, et al., 1987a).
Another semidefined medium was used to support the growth of NIH-BB and is called
NIH medium (Appendix A). NIH medium is a less complex formulation than OGMA and does not have the problem of precipitate formation as seen with the more complex medium.
2.1.3. Cell growth
All strains used in this study were maintained axenically. Stock cultures were grown in 25 cm^ tissue culture flasks (Data Packaging Corporation) containing 3.0 ml of medium. All monolayer cell growth was in either a modified Neffs Optimal Growth
Medium (OGMA) or DGM-11 with an incubation temperature of 30° C.
For large scale production of cell mass, NIH-BB cells were grown in suspension culture in NIH medium. Three liter cultures of amebas were grown in magnetically 33 stirred tissue culture flasks (Bellco Biotechnology) while 1.51 cultures were grown in 2.8
1 Fembach flasks (Coming 4420, Coming Glass Co.). Magnetically driven flasks were stirred at 160 RPM and incubated at ambient temperature (26° ± 4° C), whereas,
Fembach flasks were shaken on an orbital shaker (New Brunswick Scientific model G25) at 275 RPM and incubated at 30° ± 0.2° C. Suspension cultures were inoculated to a density of 1 X 10^ cells / ml (10% innoculum) from late log-phase seed cultures.
Generation time in suspension culture varied from 12-15 hours (mean 14 h).
Population growth was monitored by removing a 3 ml aliquot, fixing the cells with 0.5% formaldehyde, and counting in a modified Neubauer hemocytometer.
Amoebae were harvested when cultures reached a density of 7-9 X 10^ cells/ml.
Cells were sedimented from culture medium by centrifugation at 1000 X g for 15 minutes in 500 ml centrifuge bottles using an RC-2B refrigerated centrifuge with a GS-2 rotor (Ivan Sorvall Co., Norwalk, CT). The resulting cell pellet was resuspended at 4° C in 5 ml of 0.15 M KC1 per liter of medium harvested and was centrifuged at 1000 X g for
10 minutes using the GS-2 rotor. The KC1 wash was repeated twice. The washed cell pellet was resuspended in an equal volume of lysis buffer (50 raM Tris HCL, pH 7.2,
2mM dithiothreitol (DTT), 2 mM putrescine, 0.5 mM phenylmethyl sulfonyl fluoride, and 10'^M pepstatin A) and was either frozen and stored at -80° C (in a thin layer in a pyrex dish) or placed on ice and processed immediately. Typical yields from six 1.51 cultures were 130-210 g wet weight of packed cells.
2.1.4. Effect of dlamidlnes on growth
The in vivo effects of diamidines were tested using the plaque assay described by
Hugo, et al. (1991). This assay was used to quantitate the number of viable amoeba after exposure to berenil, propamidine, or pentamidine in culture medium for various lengths of time. Trophozoites were grown in 0.5 ml cultures in a 24 well tissue culture plate 34
(Costar 6645) using OGMA as the culture medium. When the cultures reached a density of 800 cells / mm2 the diamidines were added to the medium. All diamidines were dissolved in ddH^O and sterile filtered. Stock concentrations were made at 100X so that a minimal volume (5 ul) could be added to each well. Berenil was tested at a final concentration of 4 nM, propamidine 300 nM, and pentimidine 50 nM. These concentrations were chosen as they approximated the Kj values for these drugs with respect to Acanthamoeba SAMDC (Section 3.4.2). Duplicate wells were harvested at 0 h, 36 h, 72 h, and 120 h after the addition of each drug. The cells were gently washed free of the medium by first centrifugation at 1000 x g for 10 min at ambient temp, followed by two rounds of resuspension and centrifugation in 3 ml of sterile, ice-cold
0.15 M KC1. After the final wash the cells were resuspended in 0.5 ml of the KC1 solution and placed on ice. Decimal dilutions of the cultures were made using KC1 solution. Aliquots were plated on nutrient limited plates (SM/5 agar, Appendix A) using standard spread plate techniques (Hugo, et al., 1991). A feeder layer was made by mixing 50 \il of an overnight saturated culture of Enterobacter aerogenes with 3 ml of molten top agarose (Appendix A)(50° C). The molten agarose was then poured onto the surface of the plate. After the agarose had solidified the plate was then inverted and incubated at 30° C for 5-7 days. Plaques in the resulting bacterial lawn were counted and the number of viable amoebae in the original sample was determined.
2.2. Enzyme purification 2.2.1. Homogenization and extract preparation
Cells were lysed either by freezing at -80° C and thawing at ambient temperature twice or by a single pass through a French pressure cell at 14,000 psi (Aminco, Inc).
Cell breakage was monitored using a phase contrast microscope (American Optical).
Lysis was termed successful when >90% of the original cells were seen to be ruptured. 35
Cyst extracts were always prepared using a French pressure cell. An aliquot of not more
than 1% of the total volume sample was removed and frozen at -80° C for subsequent
analysis.
2.2.2. Precipitation
The cell homogenate was placed in a prechilled beaker on ice and stirred using a
magnetic stirrer. The pH of the solution was continuously monitored using a pH meter
fitted with a combination probe. Sodium acetate/acetic acid (1M of pH 4.5) was added
until the homogenate pH was 5.0-5.2. At this point the solution was transferred to centrifuge tubes (Nalgene #3118-0050) and centrifuged for 20 min at 14,000 X g. The supernatant was decanted and the pellet of denatured protein and insoluble material discarded. The solution was stirred on as previously described and the pH was returned to 7.2-7.3 by the dropwise addition of 1M tris (pH 9.5). After neutralization solid (NH4 )2S0 4 was added to 35% saturation. The mixture was stirred for 20 min on ice and then centrifuged in an HB-4 rotor at 16,000 X g for 20 min. The supernatant was decanted and the pellet discarded. Solid (NH 4 )2SC>4 was added to 65% saturation and the mixture was stirred for 20 min and centrifuged as before. The supernatant was discarded and the pellets were dissolved in l/10th the original volume of lysis buffer.
This solution was held overnight at 4° C to allow any aggregates to form, centrifuged as above, and the supernatant was filtered through a 0.2 nm filter.
2.2.3. Hydrophobic interaction chromatography
Phenyl-Sepharose CL-4B (Pharmacia) was equilibrated in standard column buffer
(SCB - 50 mM tris, pH 7.2,2mM putrescine, 2 mM dithiothreitol). The resin was degassed by gentle swirling of the flask at reduced pressure for 10 min or until outgassing was no longer visible. The degassed resin (50 ml packed volume) was packed into a glass liquid chromatography column (1.6 X 60 mm) using a packing reservoir and 36 a peristaltic pump (Isco Corp.) connected to the column outlet. The column was packed at a flow rate of 1 ml / min until the bed volume remained constant (typically 10 h). The column was equilibrated in binding buffer (SCB + 0.3 M (NH 4 >2S0 4 for preadsorption;
SCB + 0.8 M (NH 4)2SC>4 for SAMDC adsorption; see Section 3.3.2) by passing 10 column volumes (500 ml) of the appropriate degassed buffer through the column at a flow rate of 1.0 ml/min. The filtered SAMDC-containing solution was passed through the column at a flow rate of 0.2 ml / min. After applying the SAMDC, the column was washed with 5 volumes of binding buffer. SAMDC was eluted with a descending gradient of (NH 4 )2S0 4 (0.8M-0.3M) in a total volume of 600ml. Absorbance at 280 nm was continuously monitored with an LKB UV flow monitor attached to a Kipp and
Zonin two-channel chart recorder. Thirty 20 ml fractions were collected using an Isco fraction collector (Isco Corp.). Fractions with detectable SAMDC activity were pooled, concentrated ten-fold by (NH 4)2S0 4 precipitation, and dialyzed for 24 h at 4° C in
Spectropore 7 dialysis tubing (M.W. cutoff 5000 Da) against three changes of 21 each of
SCB.
2.2.4. DEAE anion exchange chromatography
Column preparation was essentially the same as used for phenyl-sepharose chromatography except that DEAE Sepharose-CL4B (Pharmacia) was used as the packing matrix. The column was equilibrated with 10 volumes of SCB at a flow rate of
1 ml / min. The dialyzed SAMDC sample was applied at a flow rate of 0.2 ml / min and the column was washed with 5 column volumes of SCB. Absorbance was monitored as before. SAMDC was eluted from the matrix by an ascending gradient of 0.0-0.5M KC1 in SCB in a total volume of 600 ml. One hundred 6 ml fractions were collected during gradient elution. Fractions were assayed for the presence of SAMDC activity and active 37 fractions were pooled and concentrated either by (NH^SCty precipitation or ultrafiltration in a 30,000 Da cutoff centrifugal ultrafilter (Centricon 30).
2.2.5. Gel filtration chromatography
Gel filtration chromatography (size exclusion chromatography) was performed using one of two systems. An HPLC system (Beckman-Altech) equipped with a Biorad
TSK-250 gel filtration column (10 X 250 mm) was used to determine the holoenzyme molecular weight for SAMDC as well as provide the final purification step. The column was equilibrated at 1.0 ml / min with degassed SCB plus 0.2 M NaCl for a total of 10 column volumes (140 ml). The column eluate was continuously monitored at 254 nm with a Hitachi spectrophotometer. Data was collected with a Hewlett-Packard model
3396A recording integrator. Sample volumes were controlled by a 20 (d injection loop.
The column was calibrated using proteins of known molecular weight provided in a kit from Sigma. SAMDC fractions were manually collected by observing the UV absorbance reading of the eluate displayed digitally on the spectrophotometer.
FPLC (fast protein liquid chromatography) with a Superose 12 gel filtration column (10 X 150 mm) was also used to purify and size SAMDC. Conditions for FPLC were the same as HPLC except that a recording integrator was part of the system and fractions were collected with a Pharmacia fraction collector instead of manually.
2.3. Electrophoresis 2.3.1. Sodium dodecyl sulfate polyacrylamide gel electrophoresis
Polypeptides were analyzed by gel electrophoresis using the method of Laemmli
(1970). Protein samples, typically in SCB, were combined with an equal volume of 2X
SDS sample buffer (Appendix A), heated to 100° C for 5 min, and then chilled on ice.
The samples were then applied to polyacrylamide gels using an Eppendorf micropipettor.
A maximum volume of 10 nl / well was applied to minigels and a maximum volume of 38
25 |xl / well was applied to full sized gels. The protein concentration in sample buffer
was adjusted to approximately 0.1 mg/ml and 1-5 ng of total protein / lane was loaded
onto the gels. Proteins were electrophoresed through 0.75 mm thick 16% acrylamide
gels (19:1 acrylamide:bisacrylamide) using either an 8 X 10 cm minivertical apparatus
(Hoefer model SE 250) at 15 ma constant current or a 14 X 16 cm apparatus (Biorad
model Protean I) at 30 ma constant current. Electrophoresis was continued until the
bromphenol blue tracking dye was within 1 cm of the bottom of the gel. Standard low
molecular weight markers proteins (Tbl. 1) were applied at the same total protein
concentration as the sample proteins into two lanes on every gel.
After electrophoresis the resolved polypeptides were either visualized by staining
with Coomassie Brilliant Blue (CBB) or transferred electrophoretically to solid-phase
supports (Immobilon-P, PVDF membrane, Millipore Corp.) (Choli, e ta l, 1989).
Electrophoretic transfer of separated polypeptides to PVDF membrane was accomplished using a transblot apparatus (Biorad Trans-Blot System) and Towben’s buffer (Appendix
A) following the procedure outlined by Choli, et al. (1989). Electrophoretic transfer was done at 15° C at 100 v for 1 h. Polypeptides that had been transferred to a solid-phase support were visualized by either staining with CBB or India Ink (Pelikan Corp.) or immunostained using the methods described in Section 2.7.2.
2.3.2. Agarose isoelectric focusing gel electrophoresis
Samples were prepared for isoelectric focusing (IEF) by removal of all salts from the sample using either dialysis or ultrafiltration followed by concentration under vacuum in a centrifugal concentrator (Savant Corp.). Precast IEF agarose gels were purchased from the FMC corporation. Two types of gels, were used pH 3-10 and pH 3-
7. The anodlyte was 0.1M acetic acid for both types of gels (Isogel IEF Technical
Reference Guide, FMC Corp., 1990). For the pH 3-10 gels 0.1M NaOH was used as the 39
Table 1. SDS-PAGE molecular weight marker proteins.
Standard 8 Mrb Bovine Serum Albumin 66.0 Ovalbumin (Hen) 45.0 Glyceraldehyde-3-phosphate dehydrogenase 36.0 Bovine Carbonic Anhydrase 29.0 Trypsinogen 24.0 Soybean Trypsin Inhibitor 20.1 Lysozyme (Hen) 14.2 Bromophenol blue < 1
a Sigma Chemical Co. b kilodaltons 40
cathodlyte while 0.1M histidine (free-base) was used with the pH 3-7 gels. Proteins
were applied to the gel using the provided sample applicator mask. A total of 1-5 ng of
protein in no more than 5 nl was placed in each slot in the sample mask. One or two
lanes per gel contained 1 nl of a mix of IEF protein standards (Tbl. 2). Proteins were
drawn into the gel by electrophoresis at 1 W constant power for 10 min. The proteins
were then focused with constant voltage of either 1000 v for the pH 3-7 gels or 1500 v
for the pH 3-10 gels for 1 h. A Pharmacia flat-bed IEF apparatus was used for all
separations. Focused proteins were either visualized by staining with CBB using the
provided protocol or transferred by capillary action to Immobilon P membrane for
immunostaining.
2.4. SAMDC assay 2.4.1. Quantitative microassay
A microassay vial was developed to scale-down the larger volume assay of
Shirahata and Pegg (1985) (Fig. 3). This vial allowed reproducible assay of volumes as small as 50 ul. It also isolated the 14CC>2 collection membrane from the assay medium,
thus, limiting artifacts caused by the splashing of radioactive substrate onto the filter.
The standard reaction mixture for SAMDC kinetic assays consisted of 50 mM tris, pH 7.2,2 mM DTT, and 2 mM putrescine. Various concentrations of SAM, ranging from 5-400 jxM, were used in kinetic assays. l^COOH-SAM (Amersham) was added to a specific activity of 0.2 mCi/mole. The reaction cocktail was made so that 90 jxl of cocktail plus 10 jd of SAMDC would give the desired concentration of reactants. The reaction cocktail was placed into the reaction vessel portion of the assay vial. The collection vessel contained a 1 X 1 cm piece of Whatman #1 filter paper onto which 25 \d of a saturated solution of Ba(OH )2 was spotted. The complete assay vial was assembled by attaching the upper collection vessel to the lower reaction vial via a 41
Table 2. Standard marker proteins used for agarose isoelectric focusing.
Band® Protein 5 Mrc pi 1 Cytochrome C 13.0 10.2 2 Myoglobin - major band 17.5 7.4 3 Myoglobin - minor band 17.5 7.0 4 Carbonic Anhydrase 31.0 6.1 5 0 - Lactoglobin (B) 35.0 5.5 6 0 - Lactoglobin (A) 35.0 5.4 7 Ovalbumin 45.0 4.8 8 Glucose Oxidase 186.0 4.2 9 Amyloglucosidase 97.0 3.6
a Band number assignments based on migration away from cathode. Band one migrates the least amount. b Standard values from Harper (1981) c kilodaltons 42
ro o 3 3
oo 3 3
00 CM
Figure 3. Quantitative microassay reaction tube. The assay tube was constructed from two modified 0.6 ml microcentrifuge tubes, 0.25 in ID. amber latex tubing (A), and a 10 x 10 mm square of Whatman #1 filter paper containing 25 pi of a saturated solution of Ba(OH )2 (B). The filter paper was folded at a 90° angle lengthwise to allow it to be positioned in the upper tube. 43
1-1.5 cm segment of 1/4 inch I.D. amber latex tubing. The entire assembly was then preequilibrated to 37 0 C in a water bath for 10 min. SAMDC (typically 2-3 X 10'4 IU) was added to the cocktail by injection of 10 id through the latex connector that separated the reaction vessel from the collection vessel. An Eppendorf repeating pipettor with a 21 ga hypodermic needle was used to inject reactants into the vials. The reaction was allowed to proceed for 30 min when 50 id of 6N H2 SO4 was injected to stop the reaction. The tubes were incubated at 37° C for an additional 15 min to allow for the complete collection of The filters were removed from the collection vessels and placed in 20 ml scintillation vials containing 10 ml of Scintiverse E scintillation cocktail
(Fisher Scientific) and radioactivity was determined in a Beckman LS2400 liquid scintillation counter.
2.4.2. Qualitative mlcrotiter plate assay
A microtiter plate assay was adapted from a procedure described by Tabor, et al.
(1976) which was used to screen mutagenized Escherichia coli colonies for SAMDC (-) mutants. This assay was used in this study to screen large numbers of liquid chromatography fractions for SAMDC activity. The assay was performed in a 96 well microtiter plate (Costar #6027). The plate was placed on ice and 10 id of sample plus 40 id of reaction cocktail was put into each well. The final concentration of reactants in each well was: 50 mM tris, pH 7.2; 2 mM DTT; 2 mM putreseine; and 20 nM
[14coOH]~SAM (specific activity 0.02 mCi/mole). The plate was then quickly covered with a piece of Whatman 3mm chromatography paper that had been cut to the exact dimension of the plate, wetted in saturated Ba(OH) 2 , and vacuum dried over KOH pellets. The lid to the microtiter plate was replaced and sealed with Parafilm. Total setup time for one plate was approximately 5 min. The assembled assay plate was placed 44 in a 37° C incubator for 2 h. After the 2 h incubation, the plate was disassembled and the filter was allowed to air dry. All portions of the assay where exposure to was possible were performed in a chemical fume hood. Upon drying, the filter was sprayed with Enhance (New England Nuclear), an autofluorographic enhancing agent, dried, and exposed to Kodak XAR-5 X-ray film for 24-48 h at -80° C. Upon development of the film, an image of those wells containing SAMDC was visible (Fig. 4).
2.4.3. Determination of pH optima
Optimum pH for SAMDC was determined using the quantitative microassay
(Section 2.4.1). There were no major modifications to the assay except for the substitution of 50 mM sodium phosphate for the tris buffer in the reaction cocktail.
Assays were performed at 37° C for 30 min. Buffer H+ concentration was varied in 0.2 pH unit intervals from pH 5.0 through pH 9.0.
2.4.4. Kinetic analysis
The quantitative microassay was used for all experiments dealing with SAMDC kinetics and inhibition. There were no major modifications to the assay except that inhibitors were added in no more than 10 nl of aqueous solution. A final volume of 100 nl was used in all assays. For all quantitative data reported, assays were repeated at least twice with 3 replica assays per experiment (n =6 to 9).
2.5. Physical structure analysis 2.5.1. Determination of a-subunit size.
Analysis of the physical structure of SAMDC used the electrophoretic techniques described in Section 2.3.1. Two major sets of experiments were done to analyze some of the physical propertites of this enzyme. First, the presence of an N-terminal pyruvate was tested by the covalent attachment of l^C-dcSAM. This was done by the incubation 45
Figure 4. Autoradiographic image obtained from the qualitative SAMDC microassay. Wells A l and H12 contained purified SAMDC to serve as control and alignment wells. Wells A2 through E12 contained 10 pi samples of fractions obtained from HIC chromatography. Wells FI through HI 1 contained 10 pi samples of fractions from DEAE chromatography. 46
of the purified enzyme under standard reaction conditions (Section 2.4.1) in the presence of 2-4 nCi of [ 14CH3] SAM and 200 mM NaCNBH 3 (Shirahata, et al., 1985). After
removal of the unreacted dcSAM by dialysis, the now labeled enzyme was processed
using SDS-PAGE methodology (Section 2.4.1). Labeled SAMDC was detected in either
dried polyacrylamide gels or on Immobilon-P membrane by autoradiography. Kodak
XAR-5 x-ray film (Eastman Kodak) was exposed to the l^C-containing samples in X-ray
cassettes at -80° C for 2 -10 days. X-ray film was developed with an automatic film
processor.
2.5.2. N-terminal amino add sequence determination
The last experiment investigating the physical structure of SAMDC was the
determination of its partial amino acid sequence. An Applied Biosystems Model 470
gas-phase peptide sequencer was used to determine the actual N-terminal amino acid
sequence of the reductively aminated a-subunit of SAMDC (Anton and Kutney, 1987b).
The sequencing was performed by the Ohio State University Biochemical
Instrumentation Center. HPLC-purified SAMDC was reductively aminated by reaction of 100 ug of the protein in the presence of 0.2M (NH 4 )2C0 3 with 0.15 M NaCNBH 3 providing the reducing agent (Anton and Kutney, 1987a). The reaction (100 jd) was allowed to proceed for 2 h at 37° C. The preparation was desalted by dialysis against three 1 liter changes of ddH 2 0 for a total of 24 h. The dialysed material was concentrated in a SpeedVac vacuum concentrator (Savant Instruments) to a volume of 60 nl. The contrated enzyme was mixed with 60 jd of 2 X SDS-PAGE sample buffer
(Appendix A) and processed by SDS-PAGE (Section 2.4.1). The resolved, reductively aminated proteins were electrophoretically transferred to Immobilon-P membrane, and 47 stained with CBB (Choli, et al., 1989). The band containing proteins 32-33 kDa in Mr
was excised from the membrane and used for automated sequencing.
2.5.3. Tryptic peptide map of SAMDC
Polypeptide fragments produced by the treatment of SAMDC with trypsin were
analyzed by HPLC (Hancock and Sparrow, 1984). A total of 100 pg of HPLC-purified
SAMDC was reduced with DTT and SH residues were carboxymethylated with
iodoacetamide (0.5 M DTT, 100 mM iodoactamide, 37° 12h). Reactants were removed
by ultrafiltration and the reduced and carboxymethylated SAMDC was digested with 100
BAEE units of TPCK-treated bovine trypsin (Sigma Chemical Co.) in a volume of 100 pi. The reaction was buffered by 50 mM ammonium bicarbonate, pH 8.0, and was allowed to proceed for 16 h at 37° C. The sample was then evaporated to dryness in a
SpeedVac apparatus (Savant Corp.). Tryptic peptides were resuspended in 100 pi of ddH2 0 and dried again in the SpeedVac. The water washes were repeated two additional times to assure that all of the volatile buffer salt was removed. The sample was then resuspended in 100 pi of an aqueous solution containing 0.1% CF 3COOH and 10%
CH3CN. Particulate matter was removed from the sample by centrifugation at 16,000 X g for 10 min at ambient temperature. HPLC analysis was performed using a 4.6 X 150mm Cig-silica reversed-phase column (Alltech Corporation). The mobile phase was an aqueous solvent gradient of 0.1% CF3COOH in which the % CH 3CN changed linearly from 10-60% during a 60 min elution profile at a flow rate of 1 ml / min. A Beckman/Altex HPLC system with a
Hitachi UV/visible spectrophotometer as a detector was used. Absorbance of the eluant was monitored at 210 nm and recorded using a Hewlett-Packard recording integrator. 48
The particulate free sample was applied using a 20 pi sample loop, with 20 pg of tryptic peptides being injected per analysis.
2.6. Developmental regulation of enzyme activity 2.6.1. SAMDC activity during encystment
SAMDC activity in actively growing and starvation-induced, encysting amoebae was measured using the quantitative microassay. Cultures were grown in 25 cm^ tissue culture flasks in OGMA until a density of 800 cells/mm^ was reached. At this point the medium was removed from half of the cultures and replaced with nutrient-free encystment medium (Appendix A). At various time points amoebae were harvested by gently scraping the cells from the flasks with a rubber policeman. The suspended cells were sedimented at 1000 X g in a clinical centrifuge in 10 X 100 mm centrifuge tubes.
The cell pellets were washed twice by resuspension in 0.15 M KC1 at 4° C followed by sedimentation as above. The final cell pellets were resuspended in 1 ml of lysis buffer
(Section 2.1.3) and placed on ice. Because cysts are resistant to gentle lysis methods, it was necessary to disrupt them using a French pressure cell at 14,000 psi. Intact cells and particulate matter were removed by centrifugation at 16,000 X g at 4° C for 30 min. The resulting supernatants were assayed for SAMDC content as before and protein concentrations were determined using the method of Bradford (1976).
2.6.2. Attempts to quantitate SAMDC protein
A polyclonal antibody to SAMDC was prepared by immunization of a rabbit. A single New Zealand White rabbit (female, 4.7 kg) was immunized by subcutaneous injection of 250 ng of HPLC purified SAMDC. The enzyme (in SCB) was emulsified with an equal amount of Freund's complete adjuvant in a total volume of 1 ml. A 10 X
10 cm patch on the animals back was prepared by shaving away the fur. A 2 ml blood 49
sample (preimmune control) was removed from the lateral ear vein prior to injection of
the antigen. The antigen emulsion was administered in five separate sites to minimize
pain to the animal. Each site received approximately 0.2 ml of the emulsion
subcutaneously (Harlow and Lane, 1988). Six weeks after the original injection of
antigen, SAMDC a-subunit that had been purified by electrophoresis through Prosieve
agarose, was administered. Purification of the subunit was performed in the same .
manner as SDS-PAGE except that Prosieve agarose was the solid matrix (ProSieve Manual, FMC Corp., 1989). Prosieve gels were stained with CBB and the a-subunit band was excised with a razor blade. The agarose containing the protein antigen was melted at 65° C and mixed with Freund’s incomplete adjuvant to a final volume of 1 ml.
Injection of the antigen was performed as previously described. An estimated 50-100 ng of purified a-subunit was injected. Six weeks after the second immunization, a third immunization was performed in the same manner as the second. Eleven days after the third immunization blood (approx. 20 ml) was collected from the central ear vein of the rabbit by venipuncture following anesthesia with apromazine (1 mg / kg). The blood was allowed to clot at room temperature for 2 h followed by an overnight incubation at 4° C.
The clot was removed from the serum and the serum was cleared by centrifugation at
10,000 X g for 20 min. The serum was stored at 4° C. Six weeks after the third immunization the animal was immunized for a fourth and final time as before. On days eleven and fourteen postimmunization, blood was collected and serum prepared as before. A total of 30 ml of serum was prepared from these two collections.
The antiserum specificity was tested by immunoprecipitation, inhibition of
SAMDC by the antiserum, and Western blotting with immunostaining. For the immunoprecipitation and enzyme inhibition experiments, SAMDC was assayed after 50
exposure to various amounts ( 0 .01- 10% v/v) of both preimmune and immune antiserum
in the standard assay buffer. For the immunoprecipitation experiments SAMDC was
assayed both with and without centrifugation at 16,000 X g for 20 min. The ability of
the antiserum to recognize SAMDC was determined by the ability to make SAMDC
sedimentable.
Several different systems for immunostaining were tested with both immune
serum and affinity purified IgG from immune serum. Purified IgG was prepared using a
Protein A solid phase matrix (Acti-Disk, FMC corp.). IgG was removed from serum that
had been diluted ten-fold in 0.01 M sodium phosphate, pH 7.2, by passing the dilute
solution through the matrix at a flow rate of 3 ml/min for 1 h with recirculation. A total
of 10 ml of serum diluted to 100 ml was processed. The Protein A matrix was washed
with 100 ml of sodium phosphate buffer without serum. IgG was eluted with 0.1M
glycine:HCl, pH 2.8. Fractions were collected (1 ml) and the absorbance at 280 nm was
determined. Fractions with absorbance readings > 0.05 were pooled, the pH brought to 7
with 1 M tris pH 9.6, and dialyzed against three changes of 11 each of phosphate- buffered saline (PBS) at 4° C over 24 hours. The concentration of IgG was adjusted to
0.5 mg / ml and aliquots were frozen at -80° C.
The different types of immunostaining techniques tried were: gold-labeled secondary antibody (GAR-gold, Biorad), horseradish peroxidase labeled secondary antibody (Cappell), and biotin ABC-alkaline phosphatase system (Vector Laboratories).
In all cases the manufacturer recommended procedures were followed. All of the above procedures shared common blocking and washing protocols. After electrophoretically transferring the polypeptides of interest to PVDF membrane the membrane was blocked for 2 h in 0.5% nonfat dry milk in PBS (BLOTTO) at 37° C (Choli et al, 1989). The 51
membrane was then either dried or used immediately. For immunostaining the
membrane was incubated in dilute (1:500-1:2000) immune serum or purified IgG (5
lag / ml) with BLOTTO being the diluent. Membranes were typically incubated 30 min
at room temperature with gentle agitation. After this incubation the membranes were
washed three times in PBS for 5 min / wash. For the biotin-ABC system the membranes
were incubated for 30 min in biotinylated mouse antirabbit IgG. The washing protocol
was repeated and the membranes were placed in the avidin-alkaline phosphatase reagent
for 30 min. The wash was repeated and the blots were developed in the color reagent for
5-30 min. The biotin-ABC system was modified by using biotinylated antiSAMDC
rabbit IgG prepared by the method of O'Shannessy and Quarles (1985) using biotin-LC-
hydrazide (Pierce Fine Chemicals). The secondary antibody (mouse antirabbit) was
unnecessary with this modification although it was necessary to increase the incubation
time with the primary antibody to 8-10 h. The remaining steps in the protocol remained
unchanged. 2.6.3. In vivo substrate concentrations
Reversed-phase ion pair high performance liquid chromatography (HPLC) was used to analyze the adenine nucleotide content of Acanthamoeba. Chromatography used the system described by Wagner, et al. (1982). The stationary phase was aCjg
(octadecylsilane) reversed-phase HPLC column (4.6 X 250 mm) with a 5 urn matrix porosity. The solvent system consisted of a gradient elution with two mobile phases. Mobile phase A contained: 0.1M NaH 2PC>4 , 2% (v/v) acetonitrile, 8.0 mM octane sulfonic acid (OSA), 0.1 mM EDTA, adjusted to pH 2.55 with 3M H 3PO4 . Mobile phase B consisted of a 70:30 mixture of 0.2M NaH 2P0 4 and acetonitrile (v/v) with 8 mM OSA, 0.1 mM EDTA, adjusted to pH 3.10 with 3M H 3PO4 . A linear gradient was 52
used in the chromatographic system consisting of a starting condition of 85% A: 15% B
and ending after 30 min with 15% A:85% B at a flow rate of 1 ml / min. The column
was held at the final concentrations of A and B for 30 min and then brought to the
starting conditions for 30 min for each sample processed. Standards were dissolved in
mobile phase A, filtered though 0.46 urn filter before being injected. Acanthamoeba extracts were prepared by homogenizing the cells in 10% perchloric acid with a Dounce
homogenizer (40 strokes), followed by centrifugation at 16,000 X g in a microcentrifuge.
Cell extract supernatants were filtered through a 0.46 ^m filter before HPLC analysis.
Detection of the nucleotides was accomplished with an Hitachi spectrophotometer measuring absorbance of the column eluate at 254 nm. Data was recorded and analyzed with a Hewlett-Packard recording integrator. Standard concentrations were confirmed by measurement of the absorbance at 259 nm in a Bausch and Lomb Spectronic 2000 spectrophotometer. A molar extinction coefficient of 15,200 was used to calculate standard concentrations. RESULTS
3.1. Purification of S-adenosylmethionine decarboxylase 3.1.1. Preparation of cell extracts
Several methods were tested for the preparation of cell extract from large
amounts (>100 g wet weight) of acanthamoebae. Initially cells were lysed using a
French pressure cell. Although greater than 99% of the cells were lysed by this method, the extracts prepared tended to have a lower specific SAMDC activity due to the
solubilization of additional unrelated proteins, although total activity was unaffected.
The use of the French press was not completely abandoned, however, because it was needed to lyse preparations of amoeba cysts.
The method that proved to be the most useful for the isolation of active SAMDC was freeze-thawing. When the cell paste was frozen in a thin layer in a pyrex baking dish, it was possible to freeze completely the cells at -80°C and completely thaw the cell paste in less than 30 min. Typically, with two freeze-thaw cycles >90% of the cells were lysed. Additionally, the presence of two protease inhibitors, phenyl methyl sulfonyl fluoride and pepstatin A, added during the first thaw cycle may have stabilized the enzyme.
Initially, the crude cell lysate was cleared by two centrifugation steps (10,000 x g for 20 min, followed by 100,000 x g for 30 min). The 100,000 x g step proved to be too time consuming because only a fraction of the total homogenate could be centrifuged at a single time (180 ml from a total volume ranging from 500-1000 ml). I found that the lysate could be adequately cleared of particulate matter, also removing a large amount of 53 54
soluble protein through the addition of an acid precipitation step. SAMDC is stable at
pH 5 and above. At this pH, many proteins found in the crude cellular extract
precipitated and formed a flocculent precipitate that aided in the removal of very fine
suspended particles. The precipitated material was removed by a 10,000 x g
centrifugation step and the pH was returned to 7.2-7.5. Some enzyme loss did occur at
this step, but I was unable to determine if the loss was due to irreversible denaturation of
the enzyme or simply trapping of the enzyme in the precipitated material. The final step
of preparing the cell extract was done by ammonium sulfate precipitation. A slight
purification was achieved by this step, but the major purpose of ammonium sulfate
precipitation was to reduce the total volume of the extract by a factor of 10.
3.1.2. Hydrophobic interaction chromatography
Binding and elution characteristics of SAMDC on phenyl-Sepharose CL-4B were
initially determined by binding the enzyme to small amounts of the matrix in a test tube experiment Once optimal binding conditions were found (0.8 M (NH 4 )2 SC>4 in SCB),
standard low pressure liquid chromatography was used to partially purify the enzyme.
Because the amount of matrix available was limited, it was necessary to pass the cell extract through the HIC column under conditions in which SAMDC did not bind (0.2 M (NH4 >2S0 4 in SCB). The flow-through material was collected from this step and solid
(NH4 )2S0 4 was added to a final concentration of 0.8M. The column was then washed free of bound proteins and the flow-through was passed through the column so that
SAMDC could bind to the matrix. These steps were necessary to maximize the amount of SAMDC that could be absorbed by the HIC matrix, otherwise much of the SAMDC was displaced by proteins with a greater degree of hydrophobicity (Fig. 5). Four- to five fold degrees of purification were achieved with the HIC step. 55
Figure 5. Results of hydrophobic interaction chromatography (HIC) on phenyl sepharose CL-4B for Acanthamoeba castellanii proteins. A. UV absorbance trace onbtained from the flow-through chromatography spectrophotometer used for all preparative LC in this work. The dotted line represents the ideal descending salt gradient used to elute absorbed proteins. B. Histogram representing SAMDC activity eluted from this system. Typically, 20 ml fractions were collected and every third fraction was assayed for activity. 56 A.
CO 0.20 LL D < J U [|/S0|O CM 0.16 ■ O
E c 0.12 s )] n h T s o " CM 0 o C 0.08 0 nv— o 0.04 0 n < 0.00 l , l_j o .O 0 200 400 600 800 1000 Elution Volume (ml)
B.
0.8
0.6
0.4
0.2
0.0 0 200 400 600 800 1000 Elution Volume (ml)
Figure 5. 57
3.1.3. Anion exchange chromatography
Optimal conditions for the binding and elution of SAMDC to the anion exchange
resin, diethylaminoethyl (DEAE) sepharose were determined in the same fashion as HIC
optima. SAMDC was optimally bound to DEAE sepharose in SCB and optimally eluted
by SCB with 0.2-0.3M KC1. Standard liquid chromatography was used to purify pooledconcentrated (by (NH 4 )2S0 4 precipitation) HIC fractions containing SAMDC.
Before chromatography was possible on the DEAE matrix, it was necessary to remove the residual salts ((N H ^SC ty) to promote binding of the enzyme. Desalting was
accomplished by dialysis against SCB. Binding and elution typically took 14-18 h and
yielded 100 assayable fractions (Fig. 6 ). Anion exchange chromatography typically
resulted in an additional five- to tenfold degree of purification of SAMDC.
3.1.4. Gel filtration chromatography
Gel filtration chromatography was used as the final step in the purification of
Acanthamoeba SAMDC. This form of chromatography was accomplished using one of
two chromatographic systems, either Fast Protein Liquid Chromatography (FPLC,
Pharmacia) or High Pressure Liquid Chromatography (HPLC). Both types of
chromatography had their own unique advantages and disadvantages. The advantages of
the HPLC system were high resolution and rapid sample throughput whereas the
advantages of the FPLC system were that of sample handling (2-3 times larger samples could be processed) and the ability to perform the separations at reduced temperatures.
Both systems were used in this project This form of chromatography typically resulted in an additional five-fold purification of the protein and provided the Mr of the
SAMDC holoenzyrae. Calibration of the chromatographic system was done with proteins of known molecular mass (Fig. 7). The pooled and concentrated DEAE fractions containing SAMDC typically yielded one major and at least three other UV absorbing peaks (Fig. 8 ). The major UV absorbing peak contained the greatest amount Figure Figure 6 . Results of anion exchange chromatography on DEAE Sepharose for for Sepharose DEAE on chromatography exchange anion Resultsof . Absorbance^^ (0.1 AUFS) 0.2 1 a | 0 I 1*° £ 0.6 g o Acanthamoeba castellanii Acanthamoeba proteins. B. Histogram representing SAMDC activity eluted from this from eluted SAMDC activity representing Histogram B. proteins. system. absorbed elute to used ascendinggradient salt ideal the represents line dashed The spectrophotometer. chromatography the flow-through from
.00.0 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.0 0-4 °-8
5 10 5 20 5 30 5 40 450 400 350 300 250 200 150 100 50 0 5 10 5 20 5 30 5 40 450 400 350 300 250 200 150 100 50 0 Elution Volume (ml) Elution Volume (ml) proteins. A. UV absorbance trace obtained trace absorbance UV A. proteins. 0.1 . r 0.2 © 0.3 0.4 0.5 58 59 0.10
0.08
E 0.06 c in 0.04 CM CD O 0.02 C CD 0.00 .Q 0 2 4 6 8 10 12 14 16 O CO 0.10 JQ < 0.08
0.06
0.04 -
0.02
0.00 4 0 1 Retention time (min)
Figure 7. HPLC integrator traces for TSK-250 gel filtration size standards. Arrows indicate the peaks of interest. A. cyanocobalamin, B. carbonic anhydrase, C. ovalbumin, D. bovine serum albumin, and G. urease. Molecular weights for these proteins are given in Table 3. Figure 7 (continued) 7 Figure 0.08 W in Absorbance 0.10 C 0.00 0.02 0.04 0.06 0.08 0.10 0.00 0.02 0.04 0.06 0.00 0.02 0.04 0.06 0.08 0.10
2 6 1 1 1 16 14 12 10 8 6 4 2 0 0 Retention time (min) time Retention 2 4 6 8 0 2 4 16 14 12 10 4 16 14 0 .1 0
E 0.08
CM 0.06
0.04 .Q 0.02
0.00 0 2 4 6 8 10 12 14 16 Retention Time (min)
SAMDC Activity Profile
£ ■> 4—• o < 0 > 0.4 - JO 0 oc
Retention Time (min)
Figure 8 . HPLC gel filtration results for partially purified A. castellanii SAMDC. The upper plot (A) represents the UV absorbance at 254 nra, 0.1 AUFS, for the sample as eluted from the TSK-250 column. The lower plot (B) represents the relative SAMDC activity found in each of the peaks. 62
of SAMDC activity. From the elution characteristics of this peak and data from the protein Mr standards (Tbl. 3), a value for relative Mr of the SAMDC holoenzyme was
determined by first calculating kav using the following formula:
v Ve - Vo , n kav - v t_ v 0 feQ*
Ve=elution volume, V0=void volume, and Vt=column volume
A plot of kav versus logio Mr was constructed and the SAMDC Mr was interpolated from the plot of marker proteins. The Mr was 88.0 ± 8.8 kilodaltons (Fig. 9). Similar values for Mr were produced using the FPLC system.
3.1.5. Summary of the purification of SAMDC
Typically, the enzyme was purified 1200 tol50Q-fold using the procedure outlined in Table 4. Various modifications of the purification protocol such as changes in the order of chromatographic steps, did not increase the general purity of the enzyme.
This procedure was optimized for purity rather than yield of active enzyme and, thus, is one reason yields of SAMDC were low ( s 15%). For all subsequent experiments, except for the in vivo studies, SAMDC was purified by this protocol.
3.2. Development of SAMDC microaesays The first steps in examining SAMDC activity in Acanthamoeba were the development of two rapid and quantitative assays for the enzyme. Previous work in our laboratory with decarboxylases had relied on the use of 20 ml scintillation vials as reaction vessels for enzyme assays (Kim, et al., 1987a). This type of assay presented several drawbacks. First, large reaction volumes were needed in this assay to ensure adequate mixing of the reagents and to minimize changes in reactant concentrations due to evaporation. Second, I wished to minimize both setup time and the amount of Table 3. HPLC gel filtration results for TSK-250 gel filtration of proteins of known molecular weight and SAMDC.
Standard 8 Mrb Elution Volume 0 Kavd Urease 1 545.0 4.94 0.00 Urease 2 272.0 5.67 0.084 Bovine Serum Albumin 1 132.0 6.75 0.209 Bovine Serum Albumin 2 66.0 7.88 0.339 Ovalbumin (Hen) 45.0 8.55 0.417 Carbonic Anhydrase 29.0 9.25 0.498 Cyanocobalamin (B 12) 1.1 13.59 1.00
peak 1 177.8 6.31 0.158 SAMDC 88.0 7.43 0.288 peak 3 38.5 8.78 0.444 peak 4 10.3 10.91 0.690
a Sigma Chemical Co. b kilodaltons c milliliters deq. 1 Figure 9. Plot of the calibration curve for the TSK-250 column. Arrows indicate indicate Arrows column. the TSK-250 for curve calibration of the Plot 9. Figure
Molecular Weight (kDa) 0 0 1 10 kilodaltons. The correlation coefficient for the best fit line is r=0.987. is line fit best the for coefficient correlation The kilodaltons. the position of the SAMDC containing peak, The average Mr = 88.0 = Mr average The peak, SAMDC containing the of position the 0.0
0.2
0.4
0.6
0.8
1.0 64 65
radioactive solid waste. And last, it was desirable to minimize the use of 14C [COOH]
SAM due to the high cost of the substrate. With the microassay I was able to use
ten-fold less radiolabled substrate than previously reported for quantitative analysis and
100-fold less in qualitative analysis (Shirahata and Pegg, 1985).
The minimum detectable amount of enzyme using the quantitative assay was
10“5 IU / ml (= lOpM per min) released, when assaying 10 ul of sample in a total reaction volume of 50 ul. The qualitative microtiter plate assay was used primarily to
screen column fractions during the purification of the enzyme. This assay was developed from a procedure described by Tabor, et al. (1976) and allowed for the overnight detection of as little as 300 dpm of 14CC>2 released from a single assay, when coupled with a fluorographic enhancing agent. This amount of released 14C0 2 is= 1.5 x 10"H moles released during the 2 h assay period (Fig. 4).
3.3. Physical characteristics of SAMDC
3.3.1. Subunit constitution and molecular weight
The subunit composition of purified SAMDC was analyzed using denaturing SDS polyacrylamide gel electrophoresis. Three major bands typically were resolved on a 15% acrylamide gel. Using a curve constructed from protein molecular weight marker data (Tbl. 5), the Mr of the three proteins found in SAMDC preparations was determined
(Fig. 10 and 11). The largest of the three proteins, with an average Mr of 32.46 kilodaltons, was the catalytic subunit of SAMDC (see Section 3.3.4). This size is comparable to the Mr of 32 kilodaltons reported for the mammalian enzyme (Pajunen, et al., 1988). There was a discrepancy between the holoenzyme Mr, the active subunit Mr, and the putative subunit constitution. If the holoenzyme is composed of two a- subunits and two smaller subunits as has been determined for both the mammalian and yeast
SAMDC, then the molecular weight of the 0-subunits should be approximately 11 kDa 66
Table 5. Determination of subunit Mr by SDS-PAGE.
Standard 8 Mrb migration 0 Rf Bovine Serum Albumin 66.0 37.13 .240 Ovalbumin (Hen) 45.0 54.54 .353 Glyceraldehyde-3-phosphate dehydrogenase 36.0 63.53 .411 Bovine Carbonic Anhydrase 29.0 82.56 .534 Trypsinogen 24.0 86.77 .562 Soybean Trypsin Inhibitor 20.1 107.30 .694 Lysozyme (Hen) 14.2 129.94 .838 Dye Front 154.50 1.000
SAMDC a-subunitd 32.8 76.33 .494 minor band 1 28 84.01 .544 minor band 2 27 89.15 .577
8 Sigma Chemical Co. b kilodaltons c mm from bottom of sample well d only protein labeled by reduction in the presence of ^C-SAM 67
SAMDC*
Figure 10. Composite photograph of SAMDC ( 14C-labeled) analysis by autoradiography, immunostaining and protein staining of SDS-PAGE resolved protein. Lanes 1 & 2 autoradiographic exposures of lanes 3 & 4. Lanes 3 4 immunostained SAMDC, lane 5 immunostained molecular weight markers. Lanes 6 and 7 coomassie blue-stained SAMDC (1 pg and 0.1 pg protein per lane respectively), lane 8 coomassie blue-stained molecular weight markers lane 9 autoradiographic image of lane 7. Only two of the three polypeptides present in purified SAMDC preparations are visible in this photograph. 68
1 0 0 90 80 70 60 g 50
32.46 Is 4 0 a 30 13 0 © 1 20
0.0 0.2 0.4 0.6 0.8 1.0
R,
Figure 11. Plot of the relative mobility of denatured standard proteins separated by SDS polyacrylamide gel electrophoresis on a 16% solids gel. The arrow indicates the level to which the major polypeptide species seen in purified SAMDC preparations migrates. This polypeptide is also the only protein labeled by [ 14CH3]-SAM in the presence of NaCNBH 3, thus, indicating that it is the catalytic (a) subunit of SAMDC. 69
(Pegg, et al., 1988c; Tabor and Tabor, 1984b). However, polypeptides in the size range expected for the smaller subunits have not been detected.
3.3.2. Isoelectric point
The isoelectric point (pi) for Acanthamoeba SAMDC was determined by analysis on thin-layer agarose isoelectric focusing (IEF) gels. The purified and desalted holoenzyme gave a single major band on both wide range (pH 3-10) and narrow range
(pH 3-7) IEF gels (Fig. 12). By constructing a standard curve based on the migration of proteins with known pi and comparing this to the migration of the amoeba SAMDC holoenzyme, the pi was 4.8 (Tbl. 6 and Fig. 13).
3.3.3. Optimal pH Acanthamoeba SAMDC had a wide pH optimum (Fig. 14). At pH values less than 7, however, activity of the enzyme rapidly dropped (Tbl. 7). This loss of activity at lower pH was reversible until the enzyme was exposed to conditions of pH < 4.5.
Presumably irreversible denaturation of the enzyme occurred at high [H+]. The optimum pH for the amoeba protein was not significantly different from the optima reported for mammalian, yeast, and E. coli SAMDCs (Pegg, 1984; Xie, etal., 1988; Bowman, etal,
1973). 3.3.4. Special properties
One significant property of all SAMDCs that have been examined is the presence of a covalently-bound pyruvoyl prosthetic group at the N-terminus of the catalytic subunit (Fig. 15) (Anton and Kutney, 1987b). This characteristic prosthetic group can be tentatively identified by the enzyme's sensitivity toward inactivation by reducing agents such as sodium borohydride and cyanoborohydride. The presence of a covalently- bound pyruvate also renders the enzyme unstable. Anton and Kutney (1987a) showed that SAMDC a-subunit becomes naturally reductively aminated in vitro. The rate Figure 12. Wide range isoelectric focusing of 14C-labeled SAMDC. Lane 1- marker proteins, 2 - immunostained SAMDC. Multiple bands seen in the SAMDC lanes could arrise from isoelectric varients due to post- translational modification (Dr. C. Brooks, personal communication) of SAMDC or because focusing was performed in the absence of SAMDC substrate (Dr. L. Kurz, personal communication) 71
Table 6. Results of agarose isoelectric focusing (pH 3-10) of standard IEF marker proteins and SAMDC.
Band 3 Proteinb Mrc pi migration 41 1 Cytochrome C 13.0 10.2 0.00 2 Myoglobin - major band 17.5 7.4 42.04 3 Myoglobin - minor band 17.5 7.0 52.00 4 Carbonic Anhydrase 31.0 6.1 69.00 5 P - Lactoglobin (B) 35.0 5.5 80.07 6 P - Lactoglobin (A) 35.0 5.4 81.98 7 Ovalbumin 45.0 4.8 89.77 8 Glucose Oxidase 186.0 4.2 96.72 9 Amyloglucosidase 97.0 3.6 102.71
SAMDC Acanthamoeba SAMDC 88.0 4.8 89.72
3 Fig. 11 b standard values from Harper (1981) c kilodaltons d mm migrated from cathode 72
10
8
6
4
2
0 20 40 60 80 100 Distance Migrated from Cathode (mm)
Figure 13. Plot of IEF standards isoelectric point (pi) versus relative mobility. Separation was on a pH 3-10 agarose IEF gel at 15° C. Arrow indicates mobility of amoeba SAMDC (calculated pi = 4.8). iue1. itga fSMCatvt tvrosH o ocnrtos Ap concentrations. ion H+ at various activity SAMDC of Histogram 14. Figure DPM released/30 min 000 200 0 0 8 600 optimum pH observed was pH 7.4. pH was observed optimum pH adjusted to the required pH by the addition of H of addition the by pH required adjusted tothe assay medium in used SAMDC was of purified IU 0.01proximately pH 3 PO 4 o aH The NaOH.or 73 Table 7. Effect of pH on in vitro SAMDC activity.
______pH ______DPMa______STD. DEV. 6.4 48 10 6.6 59 12 6.8 55 10 7.0 852 36 7.2 1498 36 7.4 1786 35 7.6 1604 47 7.8 1467 41 8.0 1645 42 8.2 1447 123 8.4 852 23 8.6 478 32
a DPM ^CC >2 released from SAM / 30 min @ 37° C. pyruvate OHO 0 O I I! II II H aN -C -C " H3C - C - C .... OH, a- subunit NK.+
Hypothetical mechanism for the formation of the N-terminal pyruvate of SAMDC. Adapted from Rescei and Snell (1984). 76 for this phenomenon is approximately 5000, that is, only half the total molecules of
SAMDC will survive long enough to catalyze 5000 reactions.
The pyruvoyl moiety is typically generated from the autocatalytic cleavage of a proenzyme polypeptide and arises from a specific serine residue in the polypeptide
(Anton and Kutny, 1987b). The existence of a bound pyruvoyl group can be indirectly shown in three ways. First, enzymes with a bound pyruvate are rapidly inactivated by reducing agents such as sodium cyanoborohydride (NaCNBH 3>. Secondly, Edman sequencing of a polypeptide containing an N-terminal pyruvate is not possible unless it is first converted to an alanyl residue by reductive animation. Finally, a direct test for the presence of an N-terminal pyruvate was the NaCNBH 3-induced covalent attachment of radiolabeled dcSAM to the a-subunit of the enzyme (Fig. 16). Acanthamoeba SAMDC was inactivated by NaCNBH 3. The activity was reduced to below the level of detection (at least 500-fold in this experiment) by incubation of the enzyme with 200 mM NaCNBH 3 for 15 min at 37° C. Reductively aminated SAMDC had an N-terminal alanyl residue. Additionally, radiolabeled dcSAM (synthesized from l^C [CH 3] SAM by SAMDC) was covalently attached to the a-subunit in the presence of NaCNBH 3 (Fig. 11) by the described mechanism (Fig. 16).
According to this experiment, the exact size of the a-subunit was 32.5 kDa (Fig. 10).
The absolute migration of the holoenzyme as resolved by IEF was confirmed using the radiolabeled preparation (Fig. 17). In both SDS-PAGE and agarose IEF experiments, only a single species of polypeptide could be detected, thus, the specificity of the labeling process was demonstrated. 3.3.5. N-terminal amino add sequence of the a- subunit
The catalytic subunit of Acanthamoeba SAMDC was completely purified using
SDS-PAGE electrophoresis followed by electrophoretic transfer to PVDF membrane
(Immobilon P, Millipore Corp.). Before purification the enzyme was treated with 77
SAM NHj NHj NH2
’tH. J 06 'tHj J © S-CHj ffiS-CH, „ CH, NaCNBH„ I C< H > | 9^'/N vk «H Figure 16. Hypothetical mechanism for the reductive labeling of SAMDC with radioactive SAM. Adapted from Rescei and Snell (1984). B. « SAMDC Figure 17. Wide range isoelectric focusing of 14C-labeled SAMDC. Tn order to provide a better visualized signal an agarose EEF gel was grossly over loaded with radiolabled SAMDC. Panel A is the coomassie blue-stained image of focused SAMDC transferred to PVDF membrane. Panel B is the autoradiographic image of the membrane. 79 Table 8 . Comparison of the amino acid sequences for the N-terminal 20 amino acids of the a-subunit of SAMDC from Acanthamoeba, Sacchromycesa, Homob , and Rattus. Exact matches are indicated by (I), conservative substitutions by (o), and nonidentity by (-). source 1 2 3 4 5 6 7 Mammalian Ser0 Ser Met Phe Val Ser Lys 1 1 1 Amoeba Set*1 Ser Met Phe Val Trp Asn 1 1 0 1 1 o - Yeast Ser Ser Leu Phe Val Phe Asp source 8 9 10 11 12 13 14 Mammalian Arg Arg Phe De Leu Lys Thr - 0 - 1 1 1 1 Amoeba Thr Lys Leu De Leu Lys Thr - 1 1 - 0 1 1 Yeast His Lys Leu Thr Met Lys Thr source 15 16 17 18 19 20 Mammalian Cys Gly Thr Thr Leu Leu i 1 Amoeba Cys*5 Gly Thr Thr ndf Leu 1 1 1 1 m Yeast Cys Gly Thr Thr Thr Thr 8 Genbank Acession M38434, (Kashiwagi, et al., 1988). b The amino arid sequence for rat and human were identical for this region, thus they are combined under the single heading mammalian (Pajunen, et al., 1988). c Pyruvoyl residue in mature enzyme anises from this residue. d Inferred amino acid alanine was the residue actually determined. e Residue inferred actual residue was not identifiable. f not determined. 80 NH4HCO3 and NaCNBH 3 to reductively aminate any pyruvoyl residues present. After electrophoretic transfer to the solid phase matrix the proteins were stained with Coomassie Brilliant Blue and the bands corresponding to the 32.5 kilodalton a- subunit were excised. These bands were subjected to gas phase Edman degradation in an Applied Biosystems 430 Sequenator (Choli, et al., 1989). Sequencing was performed by the Ohio State Center for Biochemical Instrumentation. A total of 25 cycles of degradation was performed from which eighteen amino acid residues could be positively identified. The first residue identified was an alanyl residue thus supporting the hypothesis that Acanthamoeba SAMDC is a pyruvoyl enzyme. The amoeba enzyme has two regions of sequence homology (100% identity for residues 1-5 and 11-20) and one area (residues 6-10) with only 20% sequence similarity to the mammalian enzyme (Tbl. 8 ). Overall sequence homology between Acanthamoeba SAMDC and mammalian SAMDC was 73.7% (14 exact matches/19 residues). Yeast SAMDC was only 63.2% homologous to the amoeba sequence but this figure does not include the three conservative substitutions. This high degree of homology at the amino acid level should not be overinterpreted as this region of the enzyme may be very important to its catalytic function and less tolerant of changes in structure. 3.3.6. Tryptic peptide map Purified SAMDC was analyzed by proteolytic digestion followed by separation of products by reversed-phase HPLC. Polypeptide fragments produced by cleavage withtrypsin, were separated from each other based primarily on hydrophobic properties. The separated peptides were detected by absorbance at 214 nm, the wavelength of light most strongly absorbed by peptide bonds, which was recorded on a recording integrator (Tbl. 9, Fig. 18). Bovine serum albumin and equine cytochrome C were similarly treated for use as comparative controls (Figs. 19 and 20). Trypsin cleaves proteins at lysine and arginine residues. 81 Table 9. Tryptic digest fragments of SAMDC as resolved by reversed-phase HPLC. Peak Number® Retention timeb Area 0 Percent Aread 1 0.266 162206 1.431 2 0.385 170513 1.505 3 2.370 447425 3.948 4 3.056 221306 1.953 5 3.310 690944 6.097 6 4.831 49251 0.435 7 7.605 265943 2.347 8 12.654 93713 0.827 9 18.688 5889331 51.972 10 24.046 134579 1.187 11 29.172 21601 0.191 12 33.225 192621 1.700 13 36.165 504263 4.450 14 38.565 468870 4.138 15 39.863 77371 0.623 16 41.183 569041 5.022 17 42.610 111837 0.987 18 43.260 729840 6.440 19 45.405 65013 0.573 20 48.190 45978 0.406 21 48.794 86545 0.763 22 51.199 223161 1.969 23 65.137 62254 0.549 24 73.703 39333 0.347 a see Fig. 18 bmin 0 relative area as determined by HP integrator. d % of total peak area. 0.10 E 0 .0 8 c o aT g 0 .0 6 (0 -e | 0 .0 4 < 0.02 0.00 0 10 20 30 40 50 60 70 Retention time (min) Figure 18. HPLC chromatogram of amoeba SAMDC tryptic peptides resolved by reversed-phase ion pair chromatography. Detection was at 210 nm, 0.1 AUFS. Approximately 50 pg of trypsin digested SAMDC was applied to the column. oo to 0.10 0 .0 8 £ c 0.00 0 10 20 30 40 50 60 70 Retention time (min) Figure 19. HPLC chromatogram of bovine serum albumin tryptic peptides resolved by reversed-phase ion pair chromatography. Detection was at 210 nm, 0.1 AUFS. Approximately 50 pg of digested BSA was applied to the column. w00 0.10 0 .0 8 E g 8 0 .0 6 c CO € 8 0 .0 4 .o < 0.02 10 20 30 50 60 7040 Retention time (min) Figure 20. HPLC chromatogram of bovine cytochrome C tryptic peptides resolved by reversed-phase ion pair chromatography. Detection was at 210 nm, 0.1 AUFS. A total of 150 pg of trypsin digested protein was applied to the column. oo 85 On a comparative basis, fewer peptides were resolved from the digested SAMDC than from the BSA sample (24 versus 45 fragments), but this was not unexpected as SAMDC is between 60-65% of the size of BSA. A similar number of tryptic fragments was seen for cytochrome C (Mr =13 kDa), a polypeptide much smaller than SAMDC. Because the number of peptides resolved is based on, 1. the total number of peptides produced, and 2. the ability of the chromatographic system to separate the different species of molecules, one cannot draw general conclusions based simply on the number of fragments observed. The data generated in this experiment are useful for two reasons. First, the tryptic peptide fragment elution patterns (Figs. 18,19, and 20) are useful as fingerprints for these specific proteins. The BSA and cytochrome C patterns provide an internal control for deciding the effectiveness of the chromatographic system as well as affording a means to compare this chromatographic system with other systems. Second, the tryptic fragments themselves offer a starting point in obtaining additional information about SAMDC. To get more amino acid sequence from this protein it would be necessary to purify fragments of the polypeptide. HPLC analysis of tryptic fragments of SAMDC provides the necessary technology to get these purified fragments. It would be a simple procedure to collect individual peak fractions from the HPLC. Although it is possible that a given peak fraction would contain a single species of fragment, further purification by HPLC using an alternative solvent system would be necessary to insure this. 3.4. Biochemical properties of SAMDC 3.4.1. Mlchaells constant for SAMDC The Michaelis constant (Km) determined for Acanthamoeba SAMDC was 30.3 nM. The value was calculated from an Eadie-Hofstee plot of velocity/substrate concentration versus velocity, where the slope of the best fit line is equivalent to -Km (Tbl. 10, Fig. 21). This relationship can be shown mathematically by rearrangement of the Michaelis-Menten equation (eqs. 2-4): VmaxtSl (eq.2) v= Km+[S] VmaxtSl (eq. 3) v - -Km j-jj] + Vmax (eq. 4) When both sides of the equation are divided by the substrate concentration, as shown in equation 3, and then rearranged (eq. 4), an equation for a straight line with slope of -Km and y-intercept equal to maximum velocity is obtained. The Km value, which represents the affinity of the enzyme for the substrate, falls within the range of those values reported for other eucaryotic SAMDCs (20-40 nM) (Pegg, 1984). 3.4.2. Competitive inhibition of SAMDC One of the major goals of this study was to determine the effects various diamidine inhibitors had on the in vitro activity of amoeba SAMDC. Aikins and Byers (1980) had previously observed that both ethidium bromide and berenil could induce differentiation in Acanthamoeba. Kim, et al. (1987b) had shown that the diamidines pentamidine and hydroxystilbamidine could also induce development. Because pentamidine and the related compound propamidine had been suggested as chemotheraputic agents, and cyst formation was an undesirable side-effect, an effort was made to determine if induction of encystment was due to SAMDC inhibition. Kim, et al. (1987b) had observed that at least two targets were involved in diamidine effects on the 87 B 10. Summary of kinetic properties of SAMDC. [SI8 v (x 10' 4)b v/[S] (x 10'6)c n SDd 2 0.1083 5.415 4 0.504 8 0.3188 3.985 8 0.376 20 0.6408 3.204 8 0.279 30 0.6648 2.216 8 0.222 40 0.8298 2.075 8 0.153 120 1.872 1.560 8 0.273 160 2.032 1.270 6 0.177 200 2.144 8.933 4 0.086 300 2.653 8.843 4 0.035 400 2.894 7.235 4 0.079 8 fiinoles*]'1 b pmoles* min*l»mg‘l c Umin'^mg"^ d v/[S] (x lO"6) Figure 21. Kinetic properties of Acanthamoeba SAMDC showing both first and zero order kinetics. Inset is an Eadie-Hofstee plot of the first order data from which Km and Vmax were determined. 88 T 1------r 2 - 6 4 1 - 2 0 0 10 20 30 40 50 0 0 50 100 150 200 250 300 350 400 Figure 21. oo [S] so 90 amoeba. Induction of encystment in diamidine-treated cultures could be blocked by exogenous polyamines, however inhibition of multiplication by diamidines was unaffected by polyamines in the medium. To examine this phenomenon at the biochemical level a variety of compounds were tested for their effects on the kinetics of SAMDC. All of the compounds tested, except MGBG, were known to induce encystment in the amoeba. All compounds were tested with at least five different concentrations of inhibitor. Two concentrations that reduced activity but permitted 10- 90% of the uninhibited enzyme activity were then used in the determination of the inhibitor dissociation constant Kj. Kinetic values for MGBG (Tbl. 11, Fig. 22), ethidium bromide (Tbl. 12, Fig. 23), berenil (Tbl. 13, Fig. 24), propamidine (Tbl. 14, Fig. 25), pentamidine (Tbl. 15, Fig. 26), and hydroxystilbamidine (Tbl. 16, Fig. 27) were determined and plotted. To determine Kj it was necessary to find out how great an effect each inhibitor had on the apparent Km for each concentration of inhibitor tested. This was done using the quantitative microassay and determining the reaction velocity (v) at different concentrations of both substrate and inhibitor. Analysis of kinetic dataobtained for inhibition by MGBG, berenil, propamidine, pentamidine, hydroxystilbamidine, and ethidium bromide, supported the hypothesis that all of these compounds are competitive inhibitors of SAMDC (Bitonti, et al„ 1986, Pegg, et al., 1988a; JSnne, et a l, 1985). Dissociation constants (Kj), are a measure of the relative stability of the enzyme- inhibitor complex. In competitive inhibition the enzyme can form two different complexes, either E + S to give an ES complex or E + 1 to give an El complex. Whereas the Michaelis constant is a measure of the ratio of free enzyme and substrate to enzyme- -substrate complex, the dissociation constant is a measure of the ratio of free enzyme and inhibitor to enzyme-inhibitor complex. By substitution of the Eadie-Hofstee equation with variables accounting for inhibitor concentration, (eq. 4) an equation describing the effect of a competitive inhibitor was derived (eq. 5). 91 Table 11. Effect of MGBG on the in vitro activity of SAMDC. !AMla Inhibitor [I? v/[S]b v° Z 10 none 0.764 7.635 0.521 20 tl 0.628 12.565 0.397 40 ti 0.38 15.19 0.72 60 ft 0.277 16.616 0.566 10 MGBG 250 0.599 5.99 0.545 20 tt 0.462 9.23 1.012 40 tt 0.313 12.54 0.633 60 tt 0.265 15.88 0.532 10 tt 500 0.325 3.254 0.47 20 tl 0.263 5.265 0.789 40 tt 0.2411 9.644 0.61 60 it 0.189 11.329 0.52 a |imoles/l b •min'l»mg"l 92 3 0 H CH3 H HN , NH CN -N —C—C~N —N—C 25 NH 20 >15 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 v/[S] Figure 22. Effect of methylglyoxal(bisguanylhydrazone) (MGBG) on the kinetic properties of SAMDC. Each point represents triplicate determinations for uninhibited enzyme (# ), 250 pM MGBG (V ), and 500 pM MGBG (▼). 93 Table 12. Summary of the effect of ethidium bromide on the in vitro activity of SAMDC. All concentration values are in |xmoles/l. fSAM? Inhibitor fl1a ______v/[S]b______vc STD. DEV. 5 none 0.6980 3.490 0.41 10 tt 0.5914 5.914 0.22 25 tt 0.4380 10.956 0.33 50 tt 0.2770 13.856 0.31 5 ethidium 200 0.5666 2.833 0.25 10 tt 200 0.4781 4.781 0.12 25 tt 200 0.3410 8.514 0.31 50 tt 200 0.2680 13.442 0.39 5 tt 1000 0.3416 1.708 0.30 10 tt 1000 0.3259 3.259 0.67 25 tt 1000 0.2424 6.056 0.78 50 tt 1000 0.2328 9.756 0.21 a (imoles/1 b min‘l-mg'1 c nmole- min‘1- mg'* 94 30 - 20 - 10 - 0 0.0 0.2 0.4 0.6 v/[S] Figure 23. Effect of ethidium bromide on the activity of SAMDC. Each point represents triplicate determinations for uninhibited enzyme (# ), 200 pM ethidium bromide (V ), and 1.0 raM ethidium bromide (▼). 95 Table 13. Summary of the effect of berenil on the in vitro activity of SAMDC. All concentration values are in nmoles/ 1. rSAMI Inhibitor \l\m ______v/[S]______v STD. DEV. 10 none 0.096 0.961 0.250 20 tt 0.090 1.800 0.076 30 tt 0.082 2.480 0.009 40 tt 0.067 2.690 0.089 20 0.2 0.047 1.416 0.090 30 0.2 0.042 1.688 0.056 40 0.2 0.039 1.968 0.052 20 0.5 0.033 0.625 0.053 30 0.5 0.032 0.964 0.006 40 0.5 0.029 1.148 0.070 96 NH NH ii 6 NHN=N C-NH, 0.00 0.05 0.10 v/[S] Figure 24. Effect of diminazine aceturate (Berenil) on the kinetic properties of SAMDC. Each point represents triplicate determinations for uninhibited enzyme ( • ) , 0.2 pM Berenil (v),and 0.4 pM Berenil (▼). Table 14. Summary of the effect of propamidine on the in vitro activity of SAMDC. [SAMla Inhibitor ______[T£______v/fS1b______STD. DEV. 10 none 0.16 0.086 0.1 20 tt 0.125 3.09 0.4 30 tt 0.103 5.13 0.38 40 it 0.072 7.18 1.33 10 propamidine 100 0.106 0.53 0.05 20 tt 100 0.119 1.19 0.04 30 tt 100 0.097 2.43 0.05 40 tt 100 0.064 3.62 0.11 10 tt 250 0.06 0.3 0.04 20 it 250 0.068 0.68 0.08 30 tl 250 0.0588 1.47 0.28 40 it 250 0.042 2.11 0.19 a pmoles/l b •min"*»mg'* 98 1 NH NH >—C-NH. 1 0 > 0 0.00 0.05 0.10 0.15 0.20 v/[S] Figure 25. Effect of propamidine isethionate on the kinetic properties of SAMDC. Uninhibited enzyme ( • ) , 100 pM propamidine (V ), 250 pM propamidine (T ). 99 Table 15. Summary of the effect of pentamidine on the in vitro activity of SAMDC. The velocity data for the unhibited and 5 pM pentamidine were averaged and plotted together as very low concentrations of pentamidine had no effect on the SAMDC activity. rSAMla Inhibitor m v/fSlb______vc STD. DEV. 10 none 0.698 3.49 0.41 20 ti 0.591 5.91 0.22 30 It 0.438 10.95 0.33 40 it 0.333 20.19 0.31 10 pentamidine 5 0.626 3.13 0.52 20 tt 5 0.671 6.71 0.32 30 tt 5 0.482 12.05 0.44 40 tt 5 0.346 17.30 0.12 10 tt 25 0.521 2.57 0.96 20 tt 25 0.515 5.21 0.82 30 tt 25 0.393 9.83 0.25 40 tt 25 0.347 14.56 0.44 a jxmoles/1 b •min'^nng'l c nmoles»min"^«mg'^ 100 NH NH C-NH,HnN-C 20 5 0 5 0 0.0 0.2 0.4 0.6 v/[S] Figure 26. Effect of pentamidine isethionate on the kinetic properties of SAMDC. Uninhibited enzyme ( • ) , 25 pM pentamidine (V). Unihibited data was the average of the unhibited enzyme data and SAMDC activity data in the presence of 5 pM pentamidine 101 Table 16. Summary of the effect of hydroxystilbamidine on the in vitro activity of SAMDC. [SAMja Inhibitor [I]* ______v/fS?______vc STD. DEV. 10 none 0.096 0.961 0.250 20 it 0.090 1.800 0.076 30 it 0.082 2.480 0.009 40 It 0.067 2.690 0.089 10 HS 100 0.041 0.408 0.062 20 tt 100 0.036 0.72 0.098 30 It 100 0.038 1.236 0.13 40 tt 100 0.034 1.379 0.03 10 It 500 0.021 0.249 0.002 20 tt 500 0.020 0.401 0.061 30 it 500 0.019 0.576 0.039 40 tt 500 0.015 0.615 0.078 a pmoles/l b -min‘l-mg’1 c nmoles- min'l- mg‘l 102 HO NH n C-NH,C= 0.00 0.05 0.10 W[S] Figure 27. Effect of hydroxystilbamidine on the kinetic properties of SAMDC. Each point represents triplicate determinations for uninhibited enzyme ( • ) , 100 pM hydroxystilbamidine (V ), and 500 pM hydroxystilbamidine (▼). 103 K r —K m O + i} ) (eq.5) rearranging to solve for Kj KW[I] Ki=ic7T ( e q ' 6 > Substituting the values for Km obtained in the presence of inhibitor, Ki values were calculated for the five compounds tested (Tbl. 17). Of the five inhibitors, berenil was the most potent with a dissociation constant of between 0.25 and 4 nM. 3.4.3. In vivo effects of diamidines The effects of three diamidines on culture viability were tested in order to determine whether concentrations that inhibit SAMDC activity in vitro would be inhibitory in vivo. The in vivo effects of berenil, propamidine, and pentamidine were tested by exposing Acanthamoeba to these compounds for fixed lengths of time and then testing culture viability. Late log-phase amoebae were exposed to concentrations of berenil, propamidine, and pentamidine equal to the Kj values determined in Section 3.4.3. Samples were taken at 0, 36,72, and 124 hours after addition of the inhibitors. The results are summarized in Tbl. 18 and Figure 28. It was apparent that there is a very rapid decrease in culture viability followed by a plateau level of viability. The decrease was at least 100-fold, ie. 99% cell death. It is possible that the plateau level represents the portion of the culture that was mature cysts. It should be noted that the concentrations of inhibitors used, with the exception of berenil were at least 10-fold higher than concentrations observed to induce encystment. 104 Table 17. Summary of the effects of diamidines and related compounds on the kinetics of SAMDC. Inhibitor m a | . W Km *1 kI ethidium 0 25.67 1304 tt 200 33.83 829 tt 500 58.56 1780 MGBG 0 16.99 604 tt 250 27.53 522 tt 500 62.85 685 berenil 0 29.01 0.17 tt 0.2 30.43 0.19 tt 0.5 87.05 0.14 hydroxystilbamidine 0 28.44 357 rr 100 109.3 135 tt 500 206.5 579 propamidine 0 46.07 334 tt 100 70.58 288 tt 250 134.6 380 pentamidine 0 43.78 19.4 tt 25 58.52 19.4 a concentration in jiM 105 Table 18. Effect of diamidines on Acanthamoeba survival. Amoebae were exposed to 4 pM berenil, 334 pM propamidine, or 57 pM pentamidine for the times indicated. Timea No inhibitor* berenilb propamidine 6 pentamidine 6 0 9500 9500 9500 9500 36 24500 160 130 90 72 18000 95 90 45 128 11500 95 30 55 a Hours after addition of diamidine 6 Plaque-forming units (pfu) / ml Figure 28. The effects of berenil (□), pentamidine (v), and propamidine propamidine and (v), pentamidine (□), berenil of effects The 28. Figure Cells/ml 10OOOir 1000 survival of stationary-phase trophozoites (•). trophozoites stationary-phase of survival 100 0 20 40 60 Time 80 100 120 t ( ), on the ), 106 107 3.5. Regulation of SAMDC 3.5.1. In vivo stability of SAMDC The stability of SAMDC activity in vivo was tested by inhibition of protein synthesis followed by assay for SAMDC activity. Two inhibitors of protein synthesis were tested in these experiments, Geneticin (G 4 ig, Gibco/Bethesda Research Labs.), and cycloheximide. Both of these compounds are known inhibitors of eucaryotic cytoplasmic protein synthesis. Both compounds were effective in inhibition of [l^C] amino acid incorporation in the amoeba. Geneticin appeared to inhibit protein synthesis after a several hour delay (Tbl. 19, Fig. 29a). Cycloheximide, while causing a more immediate effect on protein synthesis, did not cause any greater degree of inhibition (Tbl. 20, Fig. 30a). Most likely these differences between Geneticin and cycloheximide were due to transport differences. Loss of SAMDC activity was not observed until loss of incorporated radioactivity occurred. Once inhibition of protein synthesis was observed, SAMDC activity disappeared with a half-life of 2.3 h for both inhibitors (Tables 19 and 20, Figs. 29b and 30b). This observation was consistent with the hypothesis that turnover of SAMDC occurs rapidly. The observed half-life for the amoeba enzyme was greater than that reported for the mammalian enzyme (~1 h) (Shirahata and Pegg, 1985). The short half-life of this enzyme may be in part due to the lability of the protein itself (see Section 3.4.4.) or it may be due to the presence of specific amino acid sequences (PEST) that have been shown to target SAMDC for rapid turnover in other organisms (Loetscher, e ta l., 1991). Because inhibitors of protein synthesis can cause artifactual changes in enzyme half-life, an experiment was designed that used conditions previously shown in other systems that could specifically reduce SAMDC levels (Kameji and Pegg, 1987b). Exogenously added spermidine can cause a similar loss of SAMDC activity in other eucaryotes (Pajunin, et al., 1988). Spermidine was added to actively growing cultures of 108 Table 19. Summary of the effect of G 418 on amino acid incorporation (A.) and SAMDC specific activity (B.). A. Time (h) control DPMa STD. DEV. G418 DPMa STD. DEV. 0 4427 479 4427 479 2 5707 501 5650 493 4.5 7107 294 7503 78 6 6723 341 8938 412 8 6720 649 10624 730 10 6550 442 11872 654 12 6123 321 13001 423 B. control SAMDCb G 418 Time (h) x iq-5 STD. DEV. SAMDCb STD. DEV. x lO ’5 0 5.630 0.190 7.754 0.325 2 7.480 0.090 7.987 0.222 4 7.310 0.180 7.386 0.098 6 3.980 0.360 6.921 0.114 8 1.070 0.110 7.443 0.231 10 ndc nd 7.345 0.112 12 nd nd 7.778 0.421 3 DPM 14 C-labeled amino acids incorporated. b SAMDC specific activity in pM CO2 released min'l mg’* protein. c nd ■ not detectable. 109 Figure 29. The effects of G 4 ig inhibition of protein synthesis on SAMDC activity. A. amino acid incorporation. B. SAMDC activity. In the absence (v) andpresence ( • ) of the drug. 14000 E* 8000 6000 2000 0 2 4 6 8 10 12 Time (h) o 10 o> 8 6 4 2 0 4 6 8 10 12 Time (h) Figure 29. I l l Table 20. Summary of the effect of cycloheximide on amino acid incorporation (A.) and SAMDC specific activity (B.). A. Time (h) PPM control* STD. DEV. PPM CHXa STD. DEV 0 2500 464 2500 464 2 3156 245 3054 502 4 4014 722 3171 129 6 4124 569 3345 571 8 4109 350 3639 328 10 4707 455 4129 450 12 6126 654 3007 143 B. Time (h) control SAMDCb CHX SAMDCb x10-5 STD. DEV xl0-5 STD. DEV. 0 7.386 0.098 7.329 0.021 2 6.921 0.114 6.040 0.893 4 7.443 0.231 4.703 0.721 6 7.345 0.112 4.044 0.031 8 7.778 0.421 4.413 0.123 10 7.754 0.325 3.827 0.075 12 7.987 0.222 3.643 0.100 a DPM labeled amino acids incorporated b SAMDC specific activity in pM CO2 released min"* mg" ^protein 112 Figure 30. The effects of cycloheximide inhibition of protein synthesis on SAMDC activity. A. *^C amino acid incorporation. B. SAMDC activity. In the absence (V ) and presence ( • ) of the drug. 6000 © 5000 o. 4000 3000 - 2000 1000 0 2 4 6 8 10 12 Time (h) •9 o 9 x o> 8 E c 7 E O 6 O 2 5 £ > 4-* 4 Q. 3 CO 4 6 8 10 12 Time (h) Figure 30. 114 the amoeba and the culture was sampled at various time points (Tbl. 21, Fig. 31). No significant loss of SAMDC activity was observed during this treatment. From this result it would appear the exogenous spermidine do not cause a detectable increase in the rate of degradation or decrease in the rate synthesis of SAMDC. 3.5.2. Developmental regulation During starvation Acanthamoeba will form a protective dormant cyst. SAMDC activity levels remain stable during log-phase growth, however, this activity was rapidly lost during the encystment process (Tbls. 22 and 23, Fig. 32). Within 24 hours of the start of starvation an observable loss of 20% in the specific activity of SAMDC occurred. At this point approximately 20% of the population was scored as mature cysts. By 48 hours post-starvation SAMDC activity had dropped below the limits of detection(representing a loss of at least 99% of the original specific activity) and approximately 70% of the population was encysted. Because SAMDC is an enzyme associated with growth, it is not unexpected that it is inactive in the dormant cyst. 3.6. Production of a polyclonal antibody to SAMDC All of the aforementioned observations of SAMDC regulation are based on the presence or absence of enzyme activity. A method to quantitate SAMDC protein was needed to determine whether these changes in activity were due to actual losses of SAMDC protein or modifications in the activity of the enzyme by allosteric effectors or other post-translational alterations. The method chosen was to produce polyclonal antibodies to SAMDC. The antibody reacted with the a-subunit polypeptide in Western blots of SDS-PAGE gels as well as Western blots of IEF gels containing the holoenzyme (Fig. 11 and 12). Purified IgG from the immune serum could both specifically precipitate SAMDC activity out of solution at 0.1 ng/ml IgG and inhibit the enzyme at 5 ng/ml as compared to preimmune control IgG. Unfortunately, the immune IgG could not be used to quantitate SAMDC levels in crude cell extract using Western blotting 115 Table 21. Summary of the effect of spermidine on the in vivo activity of SAMDC. Time (h) control SAMDC* STD. DEV. Spd SAMDC 8 STD.DEV 0 1.270 0.120 0.870 0.110 2 1.010 0.090 1.101 0.160 4 0.972 0.160 0.880 0.090 6 1.020 0.100 0.690 0.130 8 1.250 0.090 0.930 0.080 10 1.300 0.210 1.250 0.210 a SAMDC specific activity in pinoles CO2 released min'l mg’lprotein xlO'^ Figure 31. The effects of spermidine on the on spermidine of effects The 31. Figure Sp. Activity (n,M C 0 2 / min /mg X1 O'5) 0.50 0.75 2.00 1.25 1.50 1.75 Treatment with 1 spermidine mM with Treatment 0 2 4 Time (h) Time in vivo in ( a ), control (•). ),control specific activity of SAMDC. of activity specific 6 8 10 116 117 Table 22. SAMDC specific activity levels during growth. Cell Density 8______SAMDC specific activity** ______STD. DEV. 49 2.560 0.123 140 3.480 0.220 904 3.040 0.168 552 2.890 0.335 1784 3.770 0.312 2683 2.970 0.189 a cells/mm^ b pM min‘ 1 mg‘1 protein x 10'^ 118 Table 23. SAMDC specific activity during starvation induced encystment. Time relative to starvation (h)a SAMDC specific activity1* STD. DEV. % cysts -36 2.560 0.123 <1 -13 3.480 0.213 <1 0 3.070 0.156 <1 8 4.200 0.259 <1 24 3.400 0.395 20 48 n.d.c 0.000 70 a time 0 = 1000 cells/mm^ b |iM min‘ 1 mg'* protein x 10"^ c not detectable < 1 pM min"* mg"* protein x 10'^ iue3. SAMDC 32. Figure Sp. Activity (jxM C 0 2 / min /mg X10-5) specific activity activity specific 0 2 3 4 2 40 20 0 0 2 - 0 4 - Time Relative to Starvation (h) Starvation to Relative Time in vivo in specific activity during starvation induced encystment. encystment. induced starvation during activity specific (•), %css( ) (O cysts % 20 30 40 50 70 805 60 119 % % Cysts 120 techniques. Crude extracts contained many cross-reacting proteins. The number and mobility of these crossreacting proteins was such that SAMDC reactivity was masked. Several attempts were made to reduce this nonspecific reactivity including: biotinylation of the primary antibody to eliminate the need for secondary antibody, preabsorption of the antibody with various cell fractions from the amoeba, the use of an avidin-biotin blocking step, and, finally, the testing of four different detection systems (gold-labeled 2° ab, HRP-labeled 2° ab, biotin-labeled 2° ab, and biotin-labeled 1° ab). All attempts to optimize the specificity of this IgG preparation failed. 3.7. S-adenosyl-L-methionine content One key control points in the biosynthesis of polyamines is the addition of aminopropyl moieties to simpler diamines. The addition of aminopropyl groups is rate limiting for the synthesis of spermidine and spermine and the source of these groups is S-5’-adenenosyl-(5')-3-methylthiopropylamine (dcSAM) that is derived from S-adenosyl-L-methionine (SAM) (Pegg, et al. 1986). Because of the dependence on dcSAM, polyamine metabolism is intimately associated with SAM metabolism. Besides polyamine biosynthesis, SAM is extremely important as a donor of methyl groups in nucleic acid metabolism and other forms of single carbon metabolism. It has been suggested that perturbations in polyamine metabolism may have a profound effect on other SAM-dependent metabolic activities. These influences could include such global consequences as alterations in DNA methylation patterns and thus a direct effect on gene expression (Zhu, et al., 1989). It appears possible, that SAM concentrations could be limiting for SAMDC activity and, conversely, that SAMDC activity might place a demand on SAM pools that could have broader effects on the cell. Thus, an effort was made in this study to examine SAM levels and their relationship to SAMDC activity. Reversed-phase ion-pair HPLC was used to identify and quantitate SAM and various compounds involved in adenine nucleotide metabolism. The system was also 121 used to determine if SAMDC activity could be measured by directly monitoring the production of dcSAM from SAM. Finally, HPLC was used to examine the intracellular pools of these compounds during growth and encystment to see if there were major changes during the life cycle of the organism. Seven major naturally occurring adenine- containing compounds were resolved using the reversed-phase ion-pair high pressure chromatography (Tbl. 23). Guanosine-containing compounds tested for possible interference during chromatography (data not shown) did not effect the analysis ofsulfur- containing adenosine compounds. Decarboxylated SAM (dcSAM), the product of the SAMDC-catalyzed reaction, was unavailable for use as a standard because of its chemical instability and commercial unavailability. Attempts to synthesize this compound enzymatically in sufficient quantities to be detectable failed. Because dcSAM is lacking a charged carboxyl group, this molecule should be more hydrophobic than both SAM and SAH, therefore it should be more strongly retained on die column matrix and thus resolved from these molecules (Fig. 33a). This resolution was necessary to quantitate accurately SAM levels. A typical chromatogram illustrates separation of the three major compounds of SAM metabolism (Fig. 33a). These mixtures of pure compounds were well resolved from each other and other purine-based molecules. The three S-adenosyl molecules, SAM, S-adenosylhomocysteine (SAH), and methylthioadenosine (MTA), could be quantitatively recovered from perchloric acid extracts of amoeba homogenates. These extracts were then analyzed using reversed-phase ion pair HPLC (Fig. 33b). To quantitate the intracellular level of SAM in the amoeba, a standard curve of peak area versus SAM concentration was constructed (Fig. 34). The relationship between relative peak is and SAM concentration was linear and regressional analysis of this data gave a correlation coefficient of r=0.998 (Tbl. 24). Equipment for detection and analysis of HPLC data is described in Section 2.6.3 122 Table 23. Structures and retention properties of the seven adenine-containing compounds tested for resolution by reversed-phase ion-pair HPLC. Compound Structure Retention time (min) c6oh nh2 S-adenosyl-L-methionine 10.70 OH OH OH OH nh2 NC r > Methylthioadenosine h3c - 8 - c^ 0 1 20.18 oh7*2 Oh 0 3 Adenosine 21.28 OH OH nh2 c t > o V N Adenosine monophosphate •O-P-O-CH j - 15.23 , ^ OH NHj OH c x > 0 0 Adenosine diphosphate •o - p - o - p - o - ch2 - 5.80 , , , p l OH OH q O. Nicotinamide adenine dinucleotide 0 OH OH 0 - P = 0 nh2 51.36 ° - p = ° L t ) 0 OH OH 123 0.10 0.08 0.06 SAH SAM MTA E 0.04 c is 0.02 c 0.00 -Q (0 § 0.10 SAM 0.08 0.06 0.04 SAH 0.02 MTA 0.00 0 2 4 6 8 10 12 14 16 18 20 Retention time (min) Figure 33. Separation of S-adenosyl-L-methionine (SAM), S-adcnosylhomocysteine (SAH), and methylthioadenosine (MTA) by reversed-pase ion-pair chromatography. These molecules are the major metabolites of SAM metabolism. Part A is the separation of 0.4 nmoles of a standard SAM/ SAH/MTA solution. Part B is the separation of these metabolites in a crude PCA extract of homogenized log-phase amoebae. Absorbance is measured at 254 nm with a range of 0.1 absorbance units full scale (AUFS). Figure 34. Graph of SAM concentration vs relative peak area. Peak area was area Peak area. peak relative vs concentration SAM of Graph 34. Figure Peak Area (x105) 0 6 fit straight line was r=0.998. was line straight fit SAM from 10 pmoles to 4 nmoles. The correlation coefficient for the best thebest for coefficient correlation The nmoles. to4 10pmoles SAM from of amounts actual representing injected SAMwas of of each concentration pi 20 of Atotal integrator. recording Hewlett-Pacckard a by determined 0 5 [SAM |xM] [SAM 10 15 20 124 125 Table 24. SAM quantitation by HPLC results. sample [SAMI M-Ma retention timeb areac moles SAM (xlQ-10)d STD 1 20.00 10.07 5684600 ------STD 2 10.00 10.22 2591000 ------STD 3 5.00 10.32 1299300 ------STD 4 2.50 10.38 692290 ------STD 5 1.25 10.43 344280 ------STD 6 0.63 10.47 183030 ------STD 7 0.31 10.46 96940 ------ mid-loge 10.69 374390 5.07 late-logf 10.70 551150 3.54 cysts* 10.76 354270 3.37 a a 20 pi sample of these std. concentrations was applied to the column. b min c relative peak area as determined by HP integrator. d per 1 x 10^ amoebae e 500-700 cells/mm2 f 1800-2000 cells/mm2 * >90% mature (double-walled) cysts, < 1% of the population were adherant trophozoites. 126 From these data the mean intracellular concentration of SAM could be calculated. Intracellular water content was estimated by using values of 60-95% of the total mass of the cell being water (Prescott, 1989). Dry mass of the cell was determined by Byers, et al.. (1969) and an average value of 6.23 x 10"*0 g/cell was used. By using these values it was possible to calculate a total cellular mass of 1.56 x 10"^ to 1.25 x 10"8 g. Subtracting the mass due non-aqueous components values of 9.37 x 10"*0 to 1.19 x 10"8 g of total cellular water are calculated. If 1 liter of H 2O = lOOOg, total cellular water volume ranges from 9.37 x 10'*3 to 1.19 x 10"^ 1. Using these values for water volume and the amounts of SAM per cell as determined by HPLC, intracellular SAM concentrations ranging from 1.36 x 10'4 to 2.71 x 10'5 M for early log-phase cells, 1.95 x 10-410 3.85 x 10"5 m for late log phase, and 1.61 x 10"4 to 2.57 x 10"5 M for mature cysts were calculated (Tbl. 24). These values are well within the values of SAM concentration determined for optimal enzyme activity (Km=2.9 x 10"5 M, Section 3.4.1) It is also apparent from this data that SAM concentrations do not significantly vary during the normal life cycle of the amoeba. DISCUSSION 4.1. Enzyme Purification 4.1.1. Purification strategy and improvements Procedures that had been used to purify SAMDC from a variety of other sources had to be significantly modified to be of use in the purification of the amoeba enzyme. The major difference between the amoeba SAMDC purification protocol and previously reported methods is the lack of an MGBG affinity chromatography step. When used as a chromatographic ligand, MGBG is very well suited for the purification of SAMDC from other eucaryotes and procaryotes. Unfortunately, one significant difference between Acanthamoeba SAMDC and all other SAMDCs yet characterized is the ability to bind MGBG. Acanthamoeba SAMDC has the lowest affinity for MGBG of any SAMDC examinined including the E. coli (Tabor and Tabor, 1984b), yeast (Kashiwagi, et al., 1988) and mammalian (Pegg, 1984) enzymes. Mammalian SAMDC has a Kj for MGBG of ^ 1 mM (Williams-Ashman and Seidenfeld, 1986), whereas, the amoeba SAMDC has a Kj of @ 600 mM (Tbl. 17). This 600-fold difference in the dissociation constants accounts for the lack of selectivity in using MGBG affinity chromatography for the purification of Acanthamoeba SAMDC. Berenil and ethidium bromide were tested in this study for use as alternative ligands to MGBG. Ethidium bromide did not sucessfully bind SAMDC to a significant degree, whereas, binding of the enzyme to berenil was too strong. In berenil affinity chromatography, protein could not be eluted except under highly denaturing conditions (3-5 M guanidine HC1) so that active SAMDC could not be obtained in the eluted fractions. Additionally, more than 30 discrete bands were seen 127 128 when proteins purified by berenil affinity chromatography were examined by SDS- PAGE. Because of these results the search for a suitable affinity matrix was abandoned. Due to the lack of a suitable affinity matrix, additional purification steps were added to the "standard" protocol. The additional steps were needed to obtain a highly purified holoenzyme, but the yield was significantly reduced. A slightly less homogenous final product was obtained, though, as determined by standard chromatographic procedures. I have qualitatively estimated the purity of the holoenzyme based on available N-termini in amino acid sequence determination to be 20-40%. This value was calculated using the known amount of protein applied to the gel to calculate the total number of pmoles of SAMDC a-subunit in the original sample. These values are only estimated as the gas-phase sequenator can only measure those molecules which can be sequenced (degradation products and modified N-termini are excluded). The significant improvements applied to the purification protocol were the addition of HIC and HPLC gel filtration steps. These additions led to almost a 100-fold greater degree of purification than that reported by Gupta, et al. (1987) for Acanthamoeba culbertsonii (~1300-fold versus 15-fold). We additionally found that the initial ultracentrifugation step used in other protocols could be replaced by a low pH precipitation step followed by medium speed centrifugation. Information from IEF experiments could be used to refine the purification protocol developed for SAMDC. Either preparative IEF or a chromatographic technique known as chromatofocusing could be used in the purification of this enzyme. With the pi known, optimization of the procedures would be rapid. Additionally, it might be possible to improve the yield of SAMDC, thus increasing the amount of available protein for experiments or reducing the amount of starting material needed. This effect would be greatly desired as currently production of cellular starting material is limited to 129 approximately 10-15g wet weight per day. In order to purify sufficient material for the N-terminal sequencing experiments approximately 4 months of preparation time was used (3.5 months to collect cells, 2 weeks to purify the enzyme) 4.1.2. Physical and kinetic parameters Physical parameters determined here for SAMDC fall within the range seen for other eucaryotic SAMDCs. The optimum pH range (Fig. 14) of 7.2-7.6 is nearly identical with that for yeast (7.3) (Kashiwaga, et al., 1988) and mammalian (7.2) (Pegg, 1984) SAMDCs. The size (Fig. 8 ) of the holoenzyme is the same as that of S. cereviseae (Kashiwaga, et al., 1988), but is nearly 25% higher than the 66.2 kDa measured for the mammalian enzyme (Pajunin, et a l, 1987) and 20% less than the 110 kDa found for the SAMDC of A. culbertsonii (Mj=l 10 kDa) (Gupta, et al., 1987b). The discrepancy with the value of Gupta, et al. (1987b) could be due to actual differences between the SAMDCs from these two species, but I feel that the difference is due to higher resolution measurements made by HPLC with the A. castellanii enzyme compared to lowere resolution measurements made by chromatography through Sepharose S-300 made for the A. culbertsoni enzyme. Determination of subunit Mr and composition by SDS-PAGE was less definitive. If the structure of the amoeba holoenzyme is similar to that found for other eucaryotes, with a subunit composition of aabb, and given that the catalytic (a) subunit has a Mr of 32.8 kDa, then the small (b) subunit of A. castellanii should have a Mr of 11.2 kDa ([ 88 - (2 x 32.8)]/2). No polypeptides were seen in this size range for any of the experiments performed even though two protein size standards were resolved in this range (lysozyme and cytochrome b) (Fig. 10 and 11). There are several explanations that would account for the apparent absence of the small b-subunit. First, it is possible that the a-subunit size was underestimated due to 130 anomalous migration in the SDS-PAGE system. It is known that extremely negative net charges can cause proteins to migrate more rapidly through a gel matrix. Secondary structure also can play a role in the migration of polypeptides in PAGE systems. Based upon the isoelectric focusing results, SAMDC probably would not have a net charge great enough to affect the migration. With a pi of 4.8 for the holoenzyme, SAMDC had a net charge in this buffer system very similar to at least two of the molecular weight marker proteins (ovalbumin pl=4.8, and carbonic anhydrase pl=6.1). It could be argued that the a-subunit has an extremely large negative charge that is partially.counteracted by an extremely positively charged b-subunit. If this were the case I would have expected to see some evidence for this in IEF experiments (Fig. 12). Second, it is also possible that the holoenzyme Mr is an overestimate. Given that the measured Mr varied by 10%, this is highly probable. If a Mr of 10% less is assumed (79.2 kDa), then the expected b- subunit Mj-= 6.8 kDa. A polypeptide of this size would not have been seen using this SDS-PAGE system. Several different SDS-PAGE protocols were used to look for polypeptides in this size range but none were seen. This may be due to the low degree of staining associated with this size protein. Third, the amoeba SAMDC may have a different subunit composition from that seen for other eucaryotes. Circumstantial evidence for this hypothesis is that in most preparations of the enzyme, western blots reveal three immunologically reactive bands corresponding to MfS of 32.8,28, and 27 kDa for a sum of 88 kDa (Fig. 11). This evidence would support a subunit composition of abg. This appears to be an unlikely explanation in that the two smaller peptides often appeared in variable amounts. It is most likely that proteins of 28 and 27 kDa are degradation products of the a-subunit or immunologically cross-reactive proteins. The question of subunit composition possibly could be answered by either in vivo labeling of SAMDC to increase the sensitivity of detection or through the isolation and 131 sequencing of the cDNA for this protein's mRNA. As indicated later, efforts to do the latter have not yet been successful. Acanthamoeba SAMDC is not the first SAMDC in which subunit composition proved to be difficult to determine. The true subunit composition of mammalian SAMDC was not determined until 10 years after it had been first purified, and was not confirmed until the cDNA was sequenced (Williams-Ashman and Pegg, 1981; Pegg, e ta l, 1981; Pajunin, etal., 1988). 4.1.3. Amino add sequence Analysis of the amino acid sequence of the a-subunit N-terminus provided data for comparison of the amoeba SAMDC to other eucaryotic SAMDCs (Tbl. 8 ). It was not surprising to find that sequence divergences among Acanthamoeba, mammals and yeast are nearly equivalent. Similar results have been seen for contractile proteins and ribosomal rRNAs (Hammer, etal., 1986; Pollard and Cooper, 1986; Gunderson and Sogin, 1986). The most surprising finding was the degree of homology between the proteins from such widely diverged organisms. As was previously stated, this region of the enzyme probably is very near the active site (Pajunin, et al., 1988). Additionally, the sequence may be essential for the proper processing of the proenzyme to the mature protein (Tabor and Tabor, 1987; Pegg, eta l., 1988c). From the limited amount of data available, one could hypothesize that there are two regions near the N-terminal end of the protein that are intolerant of sequence change. One possible way to test this hypothesis would be the use of site directed mutagenesis of specific residues in this region. This could be done using either the mammalian or S. cereviseae sequence as both have been cloned and are now available. Based on comparison between the yeast and mammalian data one finds the overall sequence homology to be much lower than the sequence homology between the specific regions presented in Tbl. 8 (Kashiwagi, et al., 1988). 132 Because of this observation, I would expect the entire Acanthamoeba amino acid sequence to have a much lower overall sequence homology to these other SAMDCs. 4.1.4. Tryptic peptide mapping of SAMDC The results of the tryptic digestion experiments were useful in the sense that a method for tentatively identifying SAMDC that is independent of activity is available. This type of analysis would be useful for identification of possible fusion proteins should the cDNA for SAMDC ever be cloned and expressed. More importantly, the tryptic digest work is a starting point for gaining more amino acid sequence for the enzyme. With additional sequence, it also should be possible to design oligonucleotide primers for use in either PCR amplification of the gene encoding this protein or probes for identifying the cDNA in a library of cloned cDNAs. The major drawback of continued tryptic peptide analysis is the requirement of substantial quantities of absolutely pure protein. Given that the peptides produced would have to be purified through two HPLC separations to assure homogeneity and die associated losses with such a protocol, at least an additional 500 g of cells would be needed. An estimated 5 mg of pure enzyme would be needed to provide the 50 pmole amount of each peptide recommended for analysis. Conservatively estimating HPLC yields at 25% for each solvent system, a minimum of 80 mg of pure SAMDC would be required as starting material. Additional stages of purification would have to be added to the current protocol to achieve homogeneity of the enzyme. It is possible that preparative nondenaturing gel electrophoresis with in situ tryptic digestion (Choli, et al., 1989) could accomplish the final purification needed but this procedure would not be without inherent losses, thus increasing the amount of starting material needed. 133 4.1.5. Biochemical Characteristics Acanthamoeba SAMDC exhibited kinetic properties typical of those reported for other eucaryotic SAMDCs. These properties are Km in the 20-30 mM range, stimulation by putrescine (50-300% increase in activity), and competitive, yet almost completely irreversible, inactivation by berenil (Pegg, 1984; Kameji and Pegg, 1987a; and Pegg, et al., 1988a and 1988c). Based on the reported measurements of intracellular SAM concentration (Tbl. 24) and the observed Km 30.3 mM for SAMDC (Section 3.4.1), it is apparent that the substrate concentrations within amoebae should support optimal enzyme activity (50-100% of Vmax). This observation, however, is based on in vitro measurements of the enzyme activity. An important difference between the in vitro assay and what is known about intracellular conditions is the concentration of an allosteric effector of the enzyme, putrescine. Kim, et al. (1987a) reported that the intracellular amount of putrescine in the amoeba was 2-3 x 10' moles. Using this range of values plus the same values for cellular water content used in calculating intracellular SAM concentrations, calculated putrescine concentrations were 1.0-18 mM. It has been shown here for the amoeba SAMDC and elsewhere for the mammalian and yeast enzymes that putrescine at concentrations greater than 1000 mM stimulate the activity of the enzyme (Kameji and Pegg, 1987a; Tabor and Tabor, 1984b). I have observed a 1.5 to 2.5-fold increase in the Vmax of the enzyme when comparing dialyzed SAMDC without putrescine to SAMDC with 2 mM putrescine (data not shown). For the mammalian enzyme, these relatively high levels of putrescine not only stimulate activity, but increase the rate of conversion of the proenzyme to the mature holoenzyme as well (Kameji and Pegg, 1987a). However, concentrations of putrescine within the amoeba do not seem sufficient to provide a major stimulatory function. Three other polyamines, 134 diaminopropane, spermidine, and norspermidine, found in this organism had no significant effect on SAMDC activity. Acanthamoeba SAMDC is notably different in its response to inhibitors as compared to other SAMDCs (Tbl. 17 for Acanthamoeba values). The greatest difference is in response to the drug MGBG. Amoeba SAMDC is 2000-fold less sensitive to this compound as compared to rodent or human SAMDC (Kj = 600 mM as compared to 0.3 mM) and at least 20-fold less sensitive than E. coli SAMDC (Pegg, 1984a). Gupta, et al. (1987b) reported similar findings for the SAMDC of A. culbertsonii with a reported Kj > 100 mM. SAMDC from another protozoan, Trypanosoma brucei bruceU also has altered kinetic responses to inhibitors of SAMDC (Bitonti, et al., 1986). T. brucei SAMDC is much less sensitive to MGBG than mammalian SAMDC (Kj = 30 mM versus 0.3 mM), but is still 20 times more sensitive to the drug than the amoeba SAMDC. Although T. b. brucei SAMDC is more sensitive to MGBG Him Acanthamoeba SAMDC, like the amoeba enzyme, this protozoan's SAMDC is poorly resolved by MGBG affinity chromatography (B. Tekwani, personal communication). Unlike MGBG, aromatic diamidines were able to induce differentiation in the amoeba. These compounds include: berenil (diminazene aceturate), propamidine, pentamidine, and hydroxystilbamidine. In addition to these four diamidines, an aromatic polycyclic compound, ethidium bromide was also an inhibitor of SAMDC as well as an inducer of encystment. Upon casual examination, ethidium bromide may not appear related to these other compounds, but after close examination it is possible to find some structural similarity to other diamidines and to spermidine (Fig. 35). Of these drugs, berenil is the most potent inhibitor of SAMDC by several orders of magnitude, while ethidium bromide is the weakest. All of these compounds share a single common feature, the presence of at least 2 amine groups at the ends of the molecule. All of the NH. H,C Spermidine 7 OH OH S-adenosyl-L-methionine H CH3 H HN. i i i NH CN=N-C-C=N-N-C i methylglyoxal- HaN nh2 H bis(guanylhydrazone) ethidium bromide Br- HO NH h2 n NH, hydroxystilbamidine NH H2N C" H / / \ \ NHN=N C - NHC berenil NH II H2N " 7 V- O C H ^H 2CH20 — f t 7 c NH, propamidine NH NH ii h2n 7 0CH2(CH2)3CH20 - 7 7 c- NH, pentamidine Stuctures of compounds that interact with SAMDC. Gray-shaded portions indicate areas of similarity. 136 molecules, except ethidium bromide, have terminal guanido moieties. How these molecules act as competitive inhibitors is unclear, as none of these compounds are structurally similar enough to the enzyme substrate to be considered a true substrate analog. Heby, et a l, (1987), suggested that the diamidines may be analogs of polyamines. It has been determined for a number of SAMDCs that activity can be modulated by the addition of polyamines. Therefore, it is possible that these drugs effect SAMDC through its affinity for polyamines rather than by directly competing for the SAM binding site. The kinetic data, however, best indicated inhibition of a competitive nature. The higher Kj (of the two values determined from the data in table 13) for berenil may be in error because the results from this experiment fell very close to the limits of detection for the assay. It is also interesting to note that berenil is the only aromatic inhibitor with internal secondary amines. It is possible that these internal amines may make berenil a closer analog to the naturally occurring polyamines. The least effective inhibitor was ethidium bromide. This result does not come as a surprise asethidium bromide is the least structurally similar to polyamines of all off the inhibitors tested. EB does however, have a superficial resemblance to spermidine and can in some instances compete with this polyamine for binding (Sakai, et al., 1975). This inhibitor possibly can interact with SAMDC in this fashion. The exact mechanism of inhibition by MGBG and diamidines is poorly understood (Janne, et al., 1985; Kramer, et al., 1989). Though the inhibitors superficially resemble spermidine, and thus are not substrate analogs, inhibition kinetics appear to be strictly competitive (Fig. 35). One possible explanation for this phenomenon could be the presence of a spermidine binding site that, while the presence or absence of spermidine does not effect the kinetic response, an analog may have an 137 effect in this binding site. It is possible that the amoeba SAMDC has evolved sufficiently that, although it retains a spermidine binding site as do all other eucaryotic SAMDCs, this binding site has lost its major function—feedback regulation of the enzyme (Pegg, et a l, 1984). if this binding site was near, or in the active site, or if the binding of SAM affected the affinity of the site for the inhibitor, it is conceivable that SAM concentration could have a competitive effect on SAMDC response to inhibitors. If this is the case it could explain the altered response to the diamidines exhibited by this enzyme as well as the competitive nature of this response. 4.2. Microassay Applications It has been successfully demonstrated in this study that it is possible to reduce the quantity of both labeled and unlabeled substrate needed to quantitatively and qualitatively characterize SAMDC. The assays as developed generate less radioactive waste, and were more sensitive, at least qualitatively, and were more economical than the published procedure. The qualitative microtiter plate assay also allows one to screen many more samples. It is possible to assay at least ten times as many samples in a given time with the microplate assay as compared to the quantitative microassay. This performance is very useful for the development of chromatographic procedures, where quantitative recovery of the enzyme is important and thus every fraction needs to be assayed for activity. 4.2.1. In vivo response to SAMDC inhibitors In vivo response to MGBG in Acanthamoeba is variable. Kim, et al. (1987a) reported that MGBG inhibited multiplication of the organism but did not not induce encystment, however, enhancement of starvation-induced encystment by MGBG in Acanthamoeba culbertsonii was reported by Gupta, et a l, (1987a). The difference in the response of Acanthamoeba to this drug may be due to the different encystment assay 138 systems used as well as species-specific differences. Inhibition of multiplication caused by MGBG in A. castellanii can be prevented by the addition of polyamines to the medium. It is not reported in the study of Gupta, et al. (1986) whether exogenous polyamines had any effect on enhanced encystment. In vivo response (encystment) to these aromatic compounds occurred over a much narrower concentration range of drug and did not correlate well with the Ki values determined for these compounds. For example, whereas pentamidine induces encystment in the 1-5 mM range (Byers, et al., 1991) the Kj value for this inhibitor was 50 mM. The results for ethdium bromide were even more striking. Akins and Byers (1980) reported that ethidium bromide at a concentration of approximately 40 mM effectively induced encystment. The Kj for ethidium bromide is 1.3 mM. Only berenil exhibited a Kj of 4 mM, this value is similar to that reported by Akins and Byers (1980) as the concentration needed to induce encystment. I was unable to determine whether the difference between Kj and encystment response concentrations was due to a transport phenomenon, where some of the compound were taken up more readily than others, or whether these drugs had other targets than SAMDC within the cell. One way to have addressed these questions would have been through the use of radiolabeled drugs. Unfortunately none were available in this form. Kim, et al. (1987b) presented indirect evidence indicating that at least two targets were involved in the action of the these drugs in vivo. Exogenously added polyamines could block the encystment inducing activity of the aromatic diamidines but could not restore multiplication. This finding refuted the hypothesis presented in other studies that exogenous polyamines block the activity of various inhibitors simply by competing for transport into the cell (Madhubala, et al., 1988; Persson, eta l., 1989; Wright, et al., 1991). 139 The induction of encystment by diamidines is probably the most serious drawback to the use of these compounds as chemotherapeutic agents. Dibromopropamidine (Brolene) is currently the drug of choice in the treatment of Acanthamoeba keratitis. The induction of drug-resistant dormant cysts by this compound may be one reason why this and other agents test for treatment of the disease suffer from a high rate of reoccurrence. Since it is possible to show that growth inhibition and encystment induction are two discrete phenomena, then it may be possible to either modify the inhibitors so that only one of the processes is affected or modify the treatment conditions so that one process is favored. Should this be done and if inhibition of SAMDC is shown to be essential for either the induction of encystment or the inhibition of multiplication then the Kj data for these compounds might be useful in selecting an appropriate agent for chemotherapy. 4.3. Regulation of SAMDC 4.3.1. Regulation of SAMDC activity It has been demonstrated that SAMDC activity is rapidly eliminated from cells in which protein synthesis has been blocked. This result indicates that continuous synthesis of the enzyme is necessary to maintain cellular levels of activity. These experiments were not able to address the question of how the enzyme is regulated. Because only enzyme activity could be measured, it is unknown whether loss of activity during inhibition of protein synthesis was due to inactivation or degradation of the enzyme. SAMDC activity also rapidly dissappeared from differentiating trophozoites. This observation is consistant with the report of polyamine concentration decreases seen during the encystment process (Kim, et a l, 1987a). This is not a surprising observation as the enzyme has been associated with active growth in other systems ( Pracyk, et al., 1991; Scalabrino, et a l, 1987; Shirahata and Pegg, 1986) and the cyst stage of the 140 amoeba is an inactive stage (Neff and Neff, 1969). The rapidity with which the enzyme vanished from the cells correlates well with the short half-life determination. It would be of interest to see if the enzyme accumulates with the same rapidity in excysting cells. Attempts were made to examine this question but synchronous excystment required for this study was never achieved. 4.3.2. Intracellular SAM concentration S-adenosylmethionine concentrations decreased by 34% from late log-phase trophozoites to mature cysts (Tbl. 24), whereas, in an earlier study polyamine levels decreased by as much as 90% during encystment (Kim, e ta l., 1987a). The observed level of fluctuation in the intracellular concentration of SAM should not significantly effect SAMDC levels in itself because the Km determined for the enzyme was still within this range. I was unable to determine if these changes in SAM levels are significant because it is possible that, although the total amount of SAM did not decrease enough to significantly effect SAMDC activity, the compound could be sequestered in some intracellular compartment, thus further lowering the concentrations of SAM available for the enzyme. From these observations one may conclude that changes in overall levels of SAM are inconsequential with regards to the encystment process. However, major changes in SAM levels could play a role in deciding the developmental fate of the amoeba. It was beyond the scope of this study to determine if major variations in SAM do occur in cells that have been treated with compounds that influence SAM metabolism. 141 4.3.3. Attempts to quantitate SAMDC protein Attempts to develop a probe for the SAMDC protein were incomplete. Although it was possible to raise a SAMDC specific antibody in the rabbit (Fig. 15), the antiserum was not specific enough to be used to detect SAMDC in crude amoeba homogenates. This was in part due to the lack of sufficient antigen for immunization (only a single rabbit was immunized) and possibly the presence of undetected proteins in the antigen preparation. Given the fact that some amino acid sequence data is available, it should be possible to generate specific antiSAMDC antisera to a synthetic peptide based on the amino acid sequence. This was not attempted due to insufficient funds. 4.3.4. Attempts to clone the SAMDC cDNA Numerous attempts were made to clone the cDNA for the SAMDC gene. The cDNA, if obtained would have been used as a probe for examining rates of synthesis of SAMDC mRNA as well as in construction of an antisense RNA vector. With the first tool it would have been possible to examine the regulation of SAMDC gene expression. With the second tool it might have been possible to eliminate the gene product from the cell to determine if this enzyme actually plays a role in encystment. The antisense approach is the logical way to explore this problem as standard forms of mutagenesis have not been successful in generating likely nuclear mutations in Acanthamoeba and because the amoeba may be polyploid (Byers, 1986). Three million recombinant lambda phage plaques from a total of three different cDNA libraries were screened. Both a heterologous SAMDC probe (the cloned human SAMDC cDNA) as well as a degenerate oligonucleotide probe (20-mer, 64-fold degeneracy) based on the N-terminal amino acid sequence of the a-subunit were used to screen for the SAMDC cDNA. A total of twelve recombinant phages were isolated and 142 subcloned. Based on results from Sanger dideoxy sequencing of the inserted cDNAs, none were homologous to the putative SAMDC sequence. The polymerase chain reaction (PCR) was also used in an attempt to obtain the SAMDC cDNA, A nonhomologous primer to which twenty to twenty-five deoxythymidine residues had been added was used to prime cDNA synthesis from Acanthamoeba polyA+ RNA. PCR was performed on this cDNA using the previously described degenerate 20mer and the nonhomologous primer without dT addition. An 800 bp fragment was produced by this method but subsequent DNA sequencing could not identify any regions related to SAMDC. I cannot offer any definitive explaination for the failure to isolate a SAMDC cDNA sequence. In theory it should have been possible to isolate at least several sequences given the number of recombinants screened. It was not altogether surprising that the heterologous human probe was incapable of detecting a SAMDC cDNA, even though this probe did hybridize with sequences present on both genomic Southern blots and Northern blots of oligo dT selected RNA. In this case sequence divergence is most likely too great to yield stable specific hybrids. As to why the oligonucleotide probe did not detect the appropriate sequence I am at a loss for the explanation. Possibly the signal to noise ratio was too small given the degeneracy of the probe and the quantity of labeled oligonucleotide needed in the hybridizations. A variety of techniques were tried to improve the signal but none were effective. 4.4. Future directions Future directions for research into the relationship between SAMDC and encystment should begin with a thorough testing of new SAMDC inhibitors that have been recently described in the literature (Byers and Bush, et al., 1991). A variety of very specific SAMDC inhibitors (transition-state analogs) have been synthesized and are now 143 available to researchers. These compounds may offer information about the specific inhibition of SAMDC without secondary effects seen with the diamidines. The molecular cloning of the SAMDC cDNA should be possible if additional amino acid sequence can be obtained. Perhaps with additional protein sequence it would be possible to make several degenerate oligonucleotide probes and use these probes to screen the Acanthamoeba cDNA libraries that are available. It should also be possible to generate a more specific antiserum by using a synthetic peptide based on the a-subunit sequence. 144 4.5. Summary Members of the genus Acanthamoeba are known to cause several serious and debilitating human diseases. Studies with other protozoans have shown that inhibition of ornithine decarboxylase to be one of the most effective means to effect a cure in infections caused by these diseases (Bacchi and McCann, 1987). Previous studies in our lab have shown Acanthamoeba to be relatively insensitive to inhibitors of ornithine decarboxylase. Inhibitors of another enzyme in the polyamine biosynthetic pathway, SAMDC, have a much more pronounced effect on the amoeba (Kim, et al., 1987b). Inhibitors of SAMDC, such as the diamidines, have been used in the treatment of other protozoal diseases (Bachrach, e ta l, 1979). One of the major goals of this work was to investigate the efficacy of targeting polyamine metabolism for disruption in the treatment of Acanthamoeba infections. Because several antiprotozoal chemotheraputic agents, particularly the diamidines, were known to inhibit S-adenosylmethionine decarboxylase, a key enzyme in the biosynthesis of polyamines, it was chosen as the target for this study. By examining the biochemistry of this particular enzyme it was hoped that a better understanding of how diamidines acted as antiamoebic agents. Numerous studies with other organisms, including other protozoans, have irrefutably demonstrated that perturbation of normal polyamine metabolism can have major effects on the growth and differentiation of the organism (Bacchi and McCann, 1987). In earlier studies in the Byers' laboratory global polyamine metabolism was investigated. This study was begun by focusing in on the enzyme SAMDC. Before the enzyme itself was examined, attempts were made to analyse both the intracellular concentrations of both the substrate and the product of the SAMDC catalysed reaction. The product of the enzyme reaction was not detectable in the system used in this study, however, it was possible to identify and quantitate the substrate, SAM. From this data, it was determined that substrate levels, although they do fluctuate, do not 145 change radically during growth and differentiation of Acanthamoeba castellanii Neff when maintained axenically in defined growth medium. From this observation it would appear that SAM concentrations are not important in the regulation of polyamine metabolism in this organism. The second stage of this study was the complete biochemical characterization of the enzyme, SAMDC. A purification protocol specifically tailored for Acanthamoeba SAMDC was developed and used to obtain the purified enzyme. With the purified enzyme the following was done: 1. Two assays for SAMDC that optimized both the sensitivity and economy of reagents were developed. 2. The physical properties of the amoeba holoenzyme and a-subunit were determined including Mr and pi. 3. Biochemical properties of amoeba SAMDC including: Michaelis constant (Km), pH optima, and dissociation constants (Kj) for six known inhibitors of the enzyme were examined. 4. The N-terminal amino acid sequence for the first 19 residues of the catalytic subunit of SAMDC were determined. 5. The presence of a covalently-bound N-terminal pyruvoyl residue was confirmed for the alpha subunit. 146 6. The half-life of the enzyme in protein synthesis-inhibited amoebae was determined. 7. Variation of SAMDC activity during encystment was described. BIBLIOGRAPHY Akins, R. 1981. Differentiation of Acanthamoeba castellanii: evidence for adevelopmental morphogen and exploratory genetic approaches. Ph.D. diss. Columbus: The Ohio State University. Akins, R. and T. Byers. 1980. Differentiation promoting factors induced in Acanthamoeba by inhibitors of mitochondrial macromolecule synthesis. Dev. Biol. 78:126-40. Akins, R., S. Goz, and T. Byers. 1985. Factors influencing the encystment enhancing activity (EEA) of Acanthamoeba castellanii. J. Gen. Microbiol. 131:2609-2617. Anton, D. and R. Kutny. 1987a. Mechanism of substrate inactivation of Escherichia coli S-adenosylmethionine decarboxylase. Biochemistry. 26(20): 6444-7. Anton, D. and R. Kutny. 1987b. Escherichia coli S-adenosylmethionine decarboxylase. Subunit structure, reductive animation, and NH 2-terminal sequences. J. Biol. Chem. 262(6): 2817-22. Arroyo, G. and J. Quinn. 1989. Association of amoebae and actinomyces in an intrauterine contraceptive user. Acta. Cytol. 33:298-300. Atmar, V. and G. Kuehn. 1981. Phosphorylation of ornithine decarboxylase by a polyamine-dependent protein kinase. Proc. Natl. Acad. Sci. USA 78:5518-22. Bacchi, C. and P. McCann. 1987. Parasitic protozoa and polyamines, pages 317-44 In: Inhibition of polyamine metabolism: biological significance and basis for new therapies. P. McCann, A. Pegg, and A. Soerjdsma, eds. Academic Press: Orlando, FL. 371 pp. Bacchi, C., H. Nathan, S. Hutner, D. Duch, and C. Nichol. 1981. Prevention of polyamines in the curative effects of amicarbilide and imidocarb for Trypanosoma brucei infections in mice. Biochem, Pharmacol. 30:883-6. 147 148 Bachrach, U., S. Brem, S. Wertman, L. Schnur, and C. Greenblatt. 1979. Effects of diamidines and ethidium bromide on the growth of Leishmania spp. promastigotes. Exp. Parasitol 48:464-70. Balana-Fouce, R., D. Ordonez-Escudero, F. Sanz-Sanchez, and A. Garrido-Pertierra. 1986a. Properties of S-adenosylmethionine decarboxylase from rabbit liver: effects of some hydrazone compounds on its activity. Comp. Biochem. Physiol. B 83(1): 221-5. Balana-Fouce, R., D. Ordonez-Escudero, F. Sanz-Sanchez, and A. Garrido-Pertierra. 1986b. Inhibition of diamine oxidase and S-adenosylmethionine decarboxylase by diminacene aceturate (berenil). Biochem. Pharmacol. 35(9): 1597-600. Balasundaram, D., C. W. Tabor and H. Tabor. 1991. Spermidine or spermine is essential for the aerobic growth of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 88: 5872-5876. Behe, M. and G. Felsenfeld. 1981. Effects of methylation on a synthetic polynucleotide: The B-Z transition in poly(dG-m^dC). Proc. Natl. Acad. Sci. USA 78(3): 1619- 23. Beninati, S. and J. Folk. 1988. Covalent polyamine-protein conjugates: analysis and distribution. Adv. Exp. Med. Biol. 250:411-22. Bey, P. C. Danzin, V. Van Dorsselaer, P. Mamont, M. Jung, and C. Tardif. 1978. Analogues of ornithine as inhibitors of ornithine decarboxylase. New deductions concerning the the enzyme's active site. J. Med. Chem. 21:50-5. Bitonti, A., J. Dumont, and P. McCann. 1986. Characterization of Trypanosoma brucei brucei S-adenosyl-L-methionine decarboxylase and its inhibition by Berenil, pentamidine and methylglyoxal bis(guanylhydrazone). Biochem. J. 237:685-9. Boeker, E. and E. Fischer. 1983. Lysine decarboxylase ( Escherichia coli B). Methods Enzymol. 94:180-4. Borochovitz, M., A. Martinez, and G. Patterson. 1981. Osteomyelitis of a bone graft allograft of the mandible with Acanthamoeba castellanii infection. Human Path. 12:573-576. Bos, H., H. Volker-Dieben, and C. Kok-Van Alphen. 1981. Acanthamoeba keratitis in the Netherlands. Trans. Roy. Soc. Trop. Med. Hyg. 75:86-91. 149 Bowman, W., C. Tabor, and H. Tabor, 1973. Spermidine biosynthesis. Purification and properties of propylamine transferase from Escherichia coli. J. Biol. Chem. 248:2480-7. Brosnan, M. and Y. Hu. 1988. Ornithine decarboxylase in mammalian tissue. Adv. Exp. Med. Biol. 250:37-44. Butt, C.G. 1966. Primary amebic meningoencephalitis. N. Eng. J. Med. 274:1473- 1476. Byers, T. 1986. Molecular biology of DNA in Acanthamoeba, Amoeba, Entamoeba, and Naegleria. Int. Rev. Cytol., 99:311-41. Byers. T. 1979. Growth, reproduction, and differentiation in Acanthamoeba. Int. Rev. Cytol. 61:283-338. Byers, T.J., R. Akins, B. Maynard, R. Lefkin, and S. Martin. 1980. Rapid growth of Acanthamoeba in defined media: Induction of encystment by glucose-acetate starvation. J. Protozool. 27(2):216-9. Byers, T.J., T. Bush, P. Mccann, and A. Bitonti. 1991. Antitrypanosomal effects of polyamine biosynthesis inhibitors correlate with increases in Trypanosoma brucei brucei S-adenosyl-L-methionine. Biochem. J. 274: 527-533. Byers, T.J., B. Kim, L. King, and E. Hugo. 1991. Molecular aspects of the cell cycle and encystment of Acanthamoeba. Rev. Infect. Dis. 13: S373-S384. Byers, T.J., V. Rudick, and M. Rudick. 1969. Cell size, macromolecule composition, nuclear number, oxygen consumption, and cyst formation during two growth phases in unagitated cultures of Acanthamoeba castellanii. J. Protozool. 16:693- 699. Centers for Disease Control. 1986. Acanthamoeba keratitis assocoiated with contact lenses-United States. Morbidity and Mortality Weekly Report. 35:405-8. Cerva, L., C. Serbus, and B. Skocil. 1973. Isolation of limax amoebae from the nasal mucosa of man. Folia Parasitol. 20:97-103. Chang, R., I. Pan, and B. Rosenau. 1966. On the nature of the 'lipovirus'. J. Exp. Med. 124:1153-1160. Chang, P. and H. Soong. 1991. Acanthamoeba keratitis in non-contact lens wearers. Arch. Ophthal. 109:463-464. 150 Chi, L., J. Vogel, and A. Shelokov. 1959. Selective phagocytosis of nucleated erythrocytes by cytotoxic amebae in cell culture. Science 130:1763-1765. Choli, T., U. Kapp, and B. Wittmann-Liebold. 1989. Blotting of proteins onto Immobilon membranes—in situ characterization and comparison with high performance liquid chromatography. J. Chromatog. 476:59-72 C16, C., B. Tantini, P. Sacchi, and C. Caldarera. 1988. Spermine inhibition of basal and stimulated adenylate cyclase is mediated by the inhibitory GTP-binding protein (Gi). Adv. Exp. Med. Biol 250:535-46. Cohen, S. and J. Lichtenstein. 1960. Polyamines and ribosome structure. J. Biol. Chem. 235:2112-6. Cohn, M., C. Tabor, and H. Tabor. 1980. Regulatory mutations affecting ornithine decarboxylase activity in Saccharomyces cerevisiae. J. Bacteriol 142(3): 791-9. Culbertson, C. 1971. The pathogenecity of soil amebas. Ann. Rev. Microbiol. 25:231- 259. Culbertson, C. 1958. Acanthamoeba: observations on animal pathogenicity. Science 127:1506-1508. Culbertson, C., J. Smith, H. Cohen, J. Minner. 1959. Experimental infection of mice and monkeys by Acanthamoeba. Am. J. Path. 35:185-197. D'Alessio, J., G. Harris, P. Pema, and M. Paule. 1981. Ribosomal ribonucleic repeat unit of Acanthamoeba castellanii: cloning and restriction endonuclease map. Biochemistry 20:3822-27. DeJonckheere, J. 1991. Ecology of Acanthamoeba. Rev. Infect. 13:Dis. S385-S387. Edwards, M., D. Stemerick, A. Bitonti, J. Dumont, P. McCann, P. Bey and A. Sjoerdsma. 1991. Antimalarial polyamine analogues. J. Med. Chem. 34: 569- 574. Fairlamb, A. 1988. The role of glutathionylspermidine and trypanothione in regulation of intracellular spermidine levels during growth of C. fasciculata. Adv. Exp. Med. Biol. 250:667-76. Fairlamb, A., G. Henderson, and A. Cerami. 1986. Thee biosynthesis of trypanothione and N1 -glutathionylspermidine in Crithidia fasciculata. Mol. Biochem. Parasitol. 21:247-57. 151 Ferrante, A. 1991. Immunity to Acanthamoeba. Rev. Infect. 13:Dis. S403-S409. Feuerstein, B., L. Williams, H. Basu, and L. Marton. 1991. Implications and concepts of polyamine-nucleic acid interactions. J. Cell. Biochem. 46: 37-47. Rexner, S. 1892. Amoebae in an abcess of the jaw. John Hopkins Hosp. Bull.3:104. Fowler, M. and R. Carter. 1965. Acute pyogenic meningitis probably due to Acanthamoeba.: sppa preliminary report Br. Med. J. 2:740-745. Fredericq, E., R. Hacha, P. Colson and C. Houssier. 1991. Condensation and precipitation of chromatin by multivalent cations. J. Biomolec. Struc. Dyn. 8: 847-865. Fujita, K., Y. Murakami, and S. Hayashi. 1982. A macromolecular inhibitor of the antizyme to ornithine decarboxylase. Biochem. J. 204:647-53 Givens, R. and A. Underwood. 1986. Through a lens blearily. Newsweek, August 25, 1986. Gellert, M., K. Mizuuchi, M. O'Day, and H. Nash. 1976. Interaction of polyamines with DNA gyrase from Escherichia coli. Proc. Natl. Acad. Sci. USA 73:3872-6. Gosule, L. and J. Schellman. 1978. DNA condensation with polyamines. J. Mol. Biol. 121:311-27. Griffiths, A., L. Cumick, M. Unitt, and S. Wilcox. 1978. The use of physiological characterizations in the classification of five species of Acanthamoeba. J. Gen. Microbiol. 107:211-213. Gunderson, J. and M. Sogin. 1986. Length variation in eukaryotic rRNAs: small subunit rRNAs from the protists Acanthamoeba castellanii and Euglena gracilis. Gene 44:63-70. Gupta, S., P. Kishore, and O. Shukla. 1987a. Polyamine metabolism as a target for growth inhibition of Acanthamoeba culbertsoni. Am. J. Trop. Med. 37(3): Hyg. 550-5. Gupta S, O. Shukla, and R. Walter. 1987b. Putrescine-activated S-adenosylmethionine decarboxylase from Acanthamoeba culbertsoni. Mol. Biochem. Parasitol. 23(3): 247-52. 152 Hamana, K., M. Niitsu, K. Samejima and S. Matsuzaki. 1991. Polyamine distributions in thermophilic eubacteria belonging to Thermus and Acidothermus. J. Biochem. 109:444-449. Hammer, J., G. Jung, and E. Korn. 1986. Genetic evidence that Acanthamoeba myosin I is a true myosin. Proc. Natl. Acad. Sci. USA 83:4655-4659. Hampel, K., P. Crosson, and J. Lee. 1991. Polyamines favor DNA triplex formation at neutral pH. Biochemistry. 30: 4455-4459. Harlow, E. and D. Lane. 1988. Antibodies: a laboratory manual.Cold Spring Harbor Laboratories. New York. 726 pp. Hancock, W. and J. Sparrow. 1984. HPLC analysis of biological compound: a laboratory guide. DekkenNew York. 361 pp. Hayashi, S., Y. Murakami, S. Matsufuji, M. Nishiyama, R. Kanamoto, and T. Kameji. 1988. Studies on ornithine decarboxylase antizymc. Adv. Exp. Med. Biol. 250:25-36. Heby, O., G. Luk, and J. Schindler. 1987. Polyamine synthesis inhibitors act as both inducers and supressors of cell differentiation, pages 165-86 In: Inhibition of polyamine metabolism: biological significance and basis for new therapies. P. McCann, A. Pegg, and A. Soerjdsma, eds. Academic Press: Orlando, FL. 371 pp. Heller, J., W. Fong, and E. Canellakis. 1976. Induction of a protein inhibitor to ornithine decarboxylase by the end products of its reaction. Proc. Natl. Acad. Sci. USA 73:1858-63. Hemdndez-Yago, J., C. Gonzdlez-Bosch, V. Miralles, M. Marcote, and S. Grisolfa. 1988. Polyamines stimulate the in vitro transport of the precursor of ornithine transcarbamylase into rat liver mitochondria. Adv. Exp. Med Biol. 250:497-508. Hiatt, A., J. Mclndoo, and R. Malmberg. 1986. Regulation of polyamine biosynthesis in tobacco. Effects of inhibitors and exogenous polyamines on arginine decarboxylase, ornithine decarboxylase, and S, adenosylmethionine decarboxylase. J. Biol. Chem. 261(3): 1293-8. Hirst, L., W. Green, W. Merz, C. Kaufman, G. Visvesvara, A. Jensen, and M. Howard. 1984. Management of Acanthamoeba keratitis. Ophthalmology 91:1105-1111. 153 Hugo, E.R. and T.J. Byers. 1989. Purification and characterization of S-adcnosyl-L- mcthioninc decarboxylase from Acanthamoeba castellanii. J. Cell Biol. Abstract 1403 109:255a. Hugo, E., W. McLaughlin, K. Oh, and O. Tuovincn. 1991. Quantitative enumeration of Acanthamoeba for evaluation of cyst inactivation in contact lens care solutions. Invest. Ophthal. Vis. Sci. 32: 655-657. Hyvonin, T„ L. Alakuijala, L. Andcrsson, A. Khomutov, R. Khomutov, and T. Eloranta. 1988. l-aminooxy-3-aminopropane reversible prevents proliferation of cultured baby hamster kidney cells by interfering with polyamine synthesis. J. Biol. Chem. 263:11138-44. Jahnes, W., H. Fullmer, and C. Li. 1957. Free-living amoebas as contaminants in monkey tissue culture. Proc. Natl. Acad. Sci. USA 96:484-488. Janne, J., L. Alhonen-Hongisto, P. Nikula, and H. Elo. 1985. S-adenosylmethionine decarboxylase as target of chemotherapy. Adv. Enzyme Reg. 24: 125-39. Janne, O., A. Crozat, M. Julkunen, N. Hickok, L. Eisenberg, and E. Melanitou. 1988. Androgen regulation of ornithine decarboxylase and S-adenosylmethionine decarboxylase gene expression. Adv. Exp. Med. Biol. 250: 1-11. JSnne, J., H. Poso, and A. Raina. 1978. Polyamines in rapid growth and cancer. Bichetn. Biophys. Acta 473:241-93. Jones, D., G. Visvesvara, and M. Robinson. 1975. Acanthamoebapolyphaga keratitis and Acanthamoeba uveitis asociated with fatal meningoencephalitis. Trans. Ophthal. Soc. U.K. 95:221-232. Kahana, C. and D. Nathans. 1985. Translational regulation of mammalian ornithine decarboxylase by polyamines. J. Biol. Chem. 260:15390-8. Kameji T, and A. Pegg. 1987a. Effect of putrescine on the synthesis of S- adenosylmethionine decarboxylase. Biochem. J. 243(1): 285-8. Kameji T, and A. Pegg. 1987b. Inhibition of translation of mRNAs for ornithine decarboxylase and S-adenosylmethionine decarboxylase by polyamines.. J. Biol. Chem. 262(6): 2427-30. Kashiwagi, K., S. Taneja, Q. Xie, C. Tabor, and H. Tabor. 1988. S-adenosylmethionine decarboxylase from Escherichia coli and from Saccharomyces cerevisiae: cloning and overexpression of the genes.. Adv. Exp. Med. Biol. 250:73-9. 154 Key III, S., W. Green, E. Willaert, A. Stevens, and S. Key, Jr. 1980. Keratitis due to Acanthamoeba castellanii. A clinicopathological case report. Arch. Ophthal. 98:475-479. Kim, B., A. Sobota, A. Bitonti, P. McCann, and T. Byers. 1987a. Polyamine metabolism in Acanthamoeba: polyamine content and synthesis of ornithine, putrescine, and diaminopropane. J. Protozool. 34:278-84. Kim, B., P. McCann, and T. Byers. 1987b. Inhibition of multiplication in Acanthamoeba castellanii by specific inhibitors of ornithine decarboxylase. J. Protozool. 34:264—66. Kownin, P., E. Bateman, and M. Paule. 1987. Eukaryotic RNA polymerase I promotor binding is directed by protein contacts with transcription initiation factor and is DNA sequence-independent. Cell 50:693-9. Kownin, P., E. Bateman, and M. Paule. 1988. Effects of single base substitutions within the Acanthamoeba castellanii rRNA promotor on transcription and on binding transcription initiation factor and RNA polymerase I. Mol. Cell Biol. 8:747-53. Kramer, D., R. Khomutov, Y. Bukin, A. Khomutov, and C. Porter. 1989. Cellular characterization of a new irreversible inhibitor of S-adenosylmethionine decarboxylase and its use in determining the relative abilities of individual polyamines to sustain growth and viability of L1210 cells. Biochem. J. 259(2): 325-31. Ku6an, Z., T. Naranda, M. Plohl, V. NOthig-Laslo, and D. Weygand-Dura&evit. 1988. Effect of spermine on transfer RNA and transfer RNA-ribosome interactions. Adv. Exp. Med. Biol. 250:525-34. Laemmeli, U. 1970. Cleavage of structural proteins during the assembly of bacteriophage T4. Nature (London) 227:680-5. Lengy, J., R. Jakovlzevich, and B. Tolis. 1971. Recovery of hartmannelloid amoeba from a purulent ear discharge. Trop. Dis. Bull. 68:818. Loetscher, P., G. Pratt and M, Rechsteiner. 1991. The C-terminus of mouse ornithine decarboxylase confers rapid degradation on dihydrofolate reductase - support for the PEST hypothesis. J. Biol. Chem. 266:11213-18. LOvgren, T., A. Peterson, and R. Loftfield. 1978. The mechanism of aminoacylationof transfer ribonucleic acid. The role of magnesium and spermine in the synthesis of isoleucyl-tRNA. J. Biol. Chem. 253:6702-14. 155 Ma, P. 1987. Protozoa and the aquircd immunodeficiency syndrome (AIDS). ATCC Quarterly Newsletter 7:1-8 Mach, M., M. White, M. Ncubaucr, M. Dcgcn, and D. Morris. 1986. Isolation of a cDNA clone encoding S-adcnosylmcthioninc decarboxylase. Expression of the gene in mitogen-activated lymphocytes. J. Biol. Chem. 261:11697-706. Madhubala R, J. Sccrist, and A. Pegg. 1988. Effect of inhibitors of S- adcnosylmcthioninc decarboxylase on the contents of ornithine decarboxylase and S-adenosylmethionine decarboxylase in L1210 cells. Biochem. J. 254(1): 45-50. MacKay, R. and F. Doolittle. 1981. Nucleotide sequences of Acanthamoeba castellanii 5s and 5.8s ribosomal nucleic acids: phylogenetic and comparative structural analysis. Nucl. Acids Res. 9:3321-3334. Maniatis, T., E. Fritsch, and J. Sambrook. 1982. Molecular cloning. Cold Spring Harbor Laboratory. 545 pp. Martinez, A. J. 1991. Infection of the central nervous system due to Acanthamoeba. Rev. Infect. Dis. 13: S399-S402. Martinez, A.J. 1985. Free-living amebas: natural history, prevention, diagnosis, pathology, and treatment of disease. CRC Press, Boca Raton, Florida. 156 pp. Martinez, J., S. Markowitz, and R. Duma. 1975. Experimental pnemonitis and encephalitis caused by Acanthamoeba in mice: pathogenesis and ultrastructural features. J. Infect. Dis. 131:692-699. Marton, L. J., A. E. Pegg and D. R. Morris. 1991. Directions for polyamine research. J. Cell. Biochem. 45:7-8. Meggio, F., F. Flamigni, C. Caldarera, C. Guamiera, and L. Pinna. 1984. Phosphorylation of rat heart ornithine decarboxylase by type-2 casein kinase. Bichem. Biophys. Res. Comm. 122:997-1003. Meisler, D. M. and I. Rutherford. 1991. Acanthamoeba and disinfection of soft contact lenses. Rev. Infect. Dis. 13: S410-S412. Metcalf, B., P. Bey, C. Danzin, M. Jung, P. Casara, and J. Vevert. 1978. Catalytic irreversible inhibition of mammalian ornithine decarboxylase (E.C. 4.1.1.17) by substrate and product analogues. J. Am. Chem. Soc. 100(8):2551-3. Mitchell, J., M. Hicks, H. Chen, and J. Hoff. 1988. Modifications of ornithine decarboxylase induced by phosphatases. Adv. Exp. Med. Biol. 250:55-72. 156 Morris, D. 1991. A new perspective on ornithine decarboxylase regulation - prevention of polyaminc toxicity is the overriding theme. J. Cell. Biochem. 46:102-105. Morris, D. and M. White. 1988. Growth-rcgulation of the cellular levels and expression of the mRNA molecules coding for ornithine decarboxylase and S- adcnosylmcthioninc decarboxylase. Adv. Exp. Med. Biol. 250: 241-52. Moruzzi, M., G. Piccinini, B. Tadolini, M. Monti, B. Barbiroli, and G. Mczzctti. 1988. Effect of polyamincs on protein kinase C activation process. Adv. Exp. Med. Biol. 250:469-80. Mustelin, T., T. Pessa, S. Lapinjoki, J. Gynther, T. JSrvinen, T. Eloranta, and L. Andersson. 1988 Two phases of ornithine decarboxylase activation during lymphocyte mitogenesis. Adv. Exp. Med. Biol. 250:301-14. Nagington, J. 1975. Isolation of amoebae from eye infections in England. Trans. Ophthal. Soc. U.K. 95:207-209. Neff, R. 1957. Purification, axenic cultivation, and description of a soil ameba, Acanthamoeba sp. J. Protozool. 4:176-182. Neff, R. and R. Neff. 1972. Induction of differentiation in Acanthamoeba by inhibitors. Comptes Rendus des Travaux de Laboratoire Carlsberg39:111-68. Neff, R. and R. Neff. 1969. The biochemistry of amoebic encystment. Symp. Soc. Exp. Biol. 23:51-81. Neff, R., S. Ray, W. Benton, and M. Wilbom. 1964. Induction of synchronous encystment (differentiation) in Acanthamoeba sp. Meth. in Cell 1:55-83. Biol. Nellen, W. and D. Gallwitz. 1982. Actin genes and actin messenger RNA in Acanthamoeba castellanii. J. Mol. 159:1-18. Biol. Newton, B. 1975. Berenil: a trypanoside with selective activity against extranuclear DNA. pages 34-47 In: Antibiotics III: Mechanism of action of antimicrobial and anitumor agents., J. Concoran and F. Hahn, eds. Springer-Verlag: Berlin. Nikula, P., L. Alhonen-Hongisto, and J. Janne. 1985. Effects of bis(guanylhydrazones) on the activity and expression of ornithine decarboxylase. Biochem. J. 231(1): 213-6. 157 Oka, T., J. Perry, T. Takemota, T. Sukai, N. Terada, and H. Inoue. 1981. The multiple regulatory roles of polyamines in the hormonal induction of mammary gland development. Adv. Polyamine Res. 3:309-20. Oredsson, S., M. Kanje, P. Mamont, J. Wagner, and O. Heby. 1986. Polyamine depletion increases cellular ribonucleotide levels. Mol. Cell. Biochem. 70(1): 89- 96. Osata, M., N. Robinson, K. Wilhelmus, and D. Jones. 1991. In vitro evaluation of antimicrobial compounds for cysticidal activity against Acanthamoeba. Rev. Infect. Dis. 13: S431-S435. O'Shannessy, D. and R. Quarles. 1985. Specific conjugation reactions of the oligosaccharide moieties of immunoglobulins. J. Appl. Biochem. 7:347-55. Oshima, T., N. Hamasaki, and T. Uzawa. 1988. Biochemical properrties of unusual polyamines found in an extreme thermophile, Thermits thermophilus. Adv. Exp. Med. Biol. 250:633-42. Page, F. 1967. Redefinition of the genus Acanthamoeba withdescriptions of three species. J. Protozool. 14:704-723. Pajunen A., V. Isomaa, O. JSnne, and C. Bardin. 1982. Androgenic regulation of ornithine decarboxylase activity and its relationship to changes in cytosol and nuclear androgen receptors. J. Biol. Chem. 257:8190-9. Pajunen A., A. Crozat, O. JSnne, R. Ihalainen, P. Laitinen, B. Stanley, R. Madhubala, and A. Pegg. 1988. Structure and regulation of mammalian S- adenosylmethionine decarboxylase. J. Biol. Chem. 263(32): 17040-9. Panagiotidis, C., S. Huang, S. Tsirka, D. Kyriakidis, and E. Canellakis. 1988. Regulation of polyamine biosynthesis in Escherichia coli by the acidic antizyme and ribosomal proteins S20 and L34. Adv. Exp. Med. Biol. 250:13-24. Park, M., E.Wolff, Z. Smitmcbride, J. Hershey, and J. Folk. 1991. Comparison of the activities of variant forms of eIF-4D - the requirement for hypusine or deoxyhypusine. J. Biol. Chem. 266: 7988-7994. Paszkokolva, C., H. Yamamoto, M. Shahamat, T. Sawyer, G. Morris, and R. Colwell. 1991. Isolation of amoebae and Pseudomonas and Legionella spp. from eyewash stations. Appl. Environ. Microbiol. 57: 163-167. Paule, M., E. Bateman, L. Hoffman, C. Iida, M. Imboden, W. Kubaska, P. Kownin, L. Han, A. Lofquist, and P. Risi. 1991. Initiation and regulation mechanisms of 158 ribosomal RNA transcription in the eukaryote Acanthamoeba castellanii. Mol. Cell. Biochem. 104:119-126. Paulus, T., P. Kiyono, and R. Davis. 1982. Polyamine-deficient Neurospora crassa mutants and synthesis of cadaverine. J. Bacteriol. 152(l):291-7. Pegg, A. 1986. Recent advances in the biochemistry of polyamines in eucaryotes. Biochem. J. 234:249-75. Pegg, A. 1984. S-adenosylmethionine decarboxylase: a brief review. Cell Biochem. Fund. 2:11-26. Pegg, A., J. Coward, R. Talekar, and J. Secrist. 1986. Effects of certain 5-substituted adenosines on polyamine synthesis: selective inhibitors of spermine synthase. Biochemistry. 25(14): 4091-7. Pegg, A., H. Hibasami, I. Matsui, and D. Bethell. 1981. Formation and interconversion of putrescine and spermidine in mammalian cells. Adv. Enzyme Reg. 19:427-33. Pegg, A., D. Jones, and J. Secrist. 1988a. Effect of inhibitors of S-adenosylmethionine decarboxylase on polyamine content and growth of L1210 cells. Biochemistry. 27(5): 1408-15. Pegg, A., T. Kameji, A. Shirahata, B. Stanley, R. Madhubala, and A. Pajunen. 1988b. Regulation of mammalian S-adenosylmethionine decarboxylase. Adv. Enzyme. Regul. 27:43-55. Pegg, A., B. Stanley, A. Pajunen, A. Crozat, and O. Janne. 1988c. Properties of human and rodent S-adenosylmethionine decarboxylase. Adv. Exp. Med. Biol. 250:101- 9. Pegg, A., L. Wiest, and A. Pajunen. 1988d. Detection of proenzyme form of S- adenosylmethionine decarboxylase in extracts from rat prostate. Biochem. Biophys. Res. Commun. 150(2): 788-93. Persson, L., A. Khomutov, and R. Khomutov. 1989. Feedback regulation of S- adenosylmethionine decarboxylase synthesis. Biochem J. 257(3): 929-31. Pohjanpelto, P., I. Virtanen, and E. HOlttfi. 1981. Polyamine starvation causes disappearance of actin filaments and microtubules in polyamine-auxotrophic CHO cells. Nature 293:475-8. Pollard, T. and J. Cooper. 1986. Actin and actin-binding proteins. A critical evaluation of mechanisms and functions. Ann. Rev. Biochem. 55:987-1035. 159 Porter, C., J. McManis, D. Lee, and R. Bergeron. 1988. Selective regulation of S- adenosylmethionine decarboxylase activity by the spermine analogue 6- spermyne. Biochem. J. 254(2): 337-42. Porter, C., F. Mikles-Robertson, D. Kramer, and C. Dave. 1979. Correlation of ultrastructural and functional damage to mitochondria of ascites L1210 cells treated in vivo with methylglyoxal bis(guanylhydrazone) or ethidium bromide. Cancer res. 39:2414-21. Poulin, R., J. Larochelle, and P. Nadeau. 1984. Polyamines in Acanthamoeba castellanii: presence of an unususually high, osmotically sensitive pool of 1,3- diaminopropane. Biochem. Biophys. Res. Comm. 122:388-93. Pracyk, J., S. Lappi, and T. Slotkin. 1991. Role of thyroid hormone in the development of beta-adrenergic control of ornithine decarboxylase in rat heart and kidney. J. Pharm. Exp. Therap. 256: 757-766. Richards, C. 1968. Two new species of Hartmannella amoebae infecting freshwater shellfish. J. Protozool. 15:651-6. Rimm, D., T. Pollard, and P. Hieter. 1988. Resolution of Acanthamoeba castellanii chromosomes by pulsed field gel electrophoresis and construction of the initial linkage map. Chromosoma 97:219-23. Rubin, C. and O. Rosen. 1975. Protein phosphorylation. Ann. Rev. Biochem. 44:831- 73. Sakai, T., R. Torget, J. I, C. Freda, and S. Cohen. 1975. The binding of polyamines and of ethidium bromide to tRNA. Nucl. Acids. Res. 2(7): 1005-23. Samples, J., P. Binder, F. Luibel, R. Font, G. Visvesvara, C. Peter. 1984. Acanthamoeba keratitis possibly aquired from a hot tub. Arch. Ophthalmol. 102:707-710. Sawyer, T., G. Visvesvara, and B. Harke. 1977. Pathogenic amoebas from brackish and ocean sediments, with a description of Acanthamoeba hatchetti, n. sp. Science 196:1324-5. Scalabrino, G., E. Lorenzini, and M. Ferioli. 1991. Polyamines and mammalian hormones. 1. Biosynthesis, interconversion and hormone effects. Mol. Cell. Endocrin. 77: 1-35. 160 Schechter, P., J. Barlow, and A. Sjoerdsma. 1987. Clinical aspects of inhibition of ornithine decarboxylase with emphasis on theraputic trials with eflomithine (DFMO) in cancer and protozoan diseases, pages 345-64 in: Inhibition of polyamine metabolism: biological significance and basis for new therapies., P. McCann, A. Pegg, and A. Sjoerdsma, eds. Academic Press: Orlando, FL. Shirahata, A. and A. Pegg. 1986. Increased content of mRNA for a precursor of S- adenosylmethionine decarboxylase in rat prostate after treatment with 2- difluoromethylomithine. J. Biol. Chem. 261(29): 13833-7. Shirahata, A., K. Christman, A. Pegg. 1985. Quantitation of S-adenosylmethionine decarboxylase protein. Biochemistry 24(16): 4417-23. Shirahata, A. and A. Pegg. 1985. Regulation of S-adenosylmethionine decarboxylase activity in rat liver and prostate. J. Biol. Chem. 260(17): 9583-8. Snyder, R. D. and M. L. Edwards. 1991. Effects of polyamine analogs on the extent and fidelity of in vitro polypeptide synthesis. Biochem. Biophys. Res. Comm. 176: 1383-1392. Stehr-Green, J., T. Bailey, F. Brandt, J. Carr, W. Bond, and G. Visvesvara. 1987. Acanthamoeba keratitis in soft contact lens wearers. J. Am. Med. Assoc. 257:57- 60. Stevens, A., T. Kilpatrick, E. Willaert, and A. Capron. 1977. Serological analyses of the cell-surface antigens of Acanthamoeba species with plasma membrane antisera. J. Protozool. 24:316-320. StOhr, M., K. Bommert, I. Shulze, and H. Jantzen. 1987. The cell cycle and its relationship to development in Acanthamoeba castellanii. J. Cell 88:579- Sci. 589. Stratford, M. and A. Griffiths. 1978. Variations in the properties and and morphology of cysts of Acanthamoeba castellanii. J. Gen. Microbiol. 108:33-37. Tabor, C. 1960. The stabilizing effect of spermine and related amines on mitochondria and protoplasts. Biochem. Biophys. Res. Comm. 2:117-21. Tabor, H., E. Hafner, and C. Tabor. 1980. Construction of an Escherichia coli strain unable to synthesize putrescine, spermidine, or cadaverine: Characterization of two genes controlling lysine decarboxylase. J. Bacteriol. 144(3): 952-6. 161 Tabor, C. and H. Tabor. 1987. The speEspeD operon of Escherichia coli. Formation and processing of a proenzyme form of S-adenosylmethionine decarboxylase. J. Biol. Chem. 262(33): 16037-40. Tabor, C. and H. Tabor. 1984a. Polyamines. Ann. Rev. Biochem. 53:749-90. Tabor, C., and H. Tabor. 1984b. Methionine adenosyltransferase (S- adenosylmethionine synthetase) and S-adenosylmethionine decarboxylase. Adv. Enzymol. Rel. Areas Mol. Biol. 56:251-82. Tabor, H., C. Tabor, and E. Hafner. 1976. Convenient method for detecting 1^C02 in multiple samples: application to rapid screening for mutants. J. Bacteriol. 128:485-6. Tabor, H., C. Tabor, M. Cohn, and E. Hafner. 1981. Streptomycin resistance (rpsL) produces an absolute requirement for polyamines for growth of an Escherichia coli strain unable to synthesize putrescine or spermidine \D{speD-speB) DspecQ. J. Bacteriol. 147(2): 702-4. Tabor, C., H. Tabor, and Q. Xie. 1986. Spermidine synthase of Escherichia coli: localization of the speE gene. Proc. Natl. Acad. Sci. U. S. A. 83(16): 6040-4. Tadolini, B. and G. Hakim. 11988. Interaction of polyamines with phospholipids, spermine and Ca^+ competition for phosphatidylserine containing liposomes. Adv. Exp. Med. Biol. 250:481-90. Thomas, T., U. Gunnia, and T. Thomas. 1991. Polyamine-induced B-DNA to Z-DNA conformational transition of a plasmid DNA with (dG-dC)n insert. J. Biol. Chem. 266: 6137-6141. Toninello, A., D. Siliprandi, P. Castagnini, M. Novello, and N. Siliprandi. 1988. Bidirectonal transport of spermine across the inner membrane of liver mitochondria. Adv. Exp. Med. Biol. 250:491-6. Toninello, A., F. Di Lisa, D. Siliprandi, P., and N. Siliprandi. 1984. Protective and restorative effects of spermine on the membrane potential of rat liver mitochondria. In: Advances in polyamines in biomedical science, C. Caldarera and U. Bachrach, eds. CLUEB: Bologna. Tyagi, A., C. Tabor, and H. Tabor. 1983. Ornithine decarboxylase Saccharomyces cerevisiae). Methods Enzymol. 94:135-9. Visvesvara, G. 1991. Classification of Acanthamoeba. Rev. Infect. 13: Dis. S369-S372. 162 Visvesvara, G. and W. Balamuth. 1975. Comparative studies on related free-living and pathogenic amoebae with special reference to Acanthamoeba. J. Protozool. 22:245-256. Visvesvara, G., D. Jones, and N. Robinson. 1975. Isolation, identification, and biological characterization of Acanthamoeba polyphaga from a human eye. Am. J. Trop. Med. 24:784-790. Visvesvara, G., S. Mirra, F. Brandt, D. Moss, H. Mafhews, and A. Martinez. 1983. Isolation of two strains of Acanthamoeba castellanii from human tissues and their pathogenicity and isoenzyme profiles. J. Clin. Microbiol. 18(6):1405-1412. Visvesvara, G. and J. Stehr-Green. 1990. Epidemiology of free-living ameba infections. J. Protozool. 37(4):25S-33S. Volonski, M. 1931. Hartmannella castellanii Douglas et classification des Hartmannelles. Arch. Zool. Exp. 72:317-39.Gen. Wagner, J., C. Danzin, and P. Mamont 1982. Reversed-phase ion-pair liquid chromatographic procedure for the simultaneous analysis of S-adenosyl methionine, its metabolites, and natural polyamines. J. Chromatog. 227:349-68. Wang, S. and H. Feldman. 1961. Occurrence of Acanthamoeba in tissue cultures innoculated with human pharyngeal swabs. Antimicrob. Agents Chemother. 1:50- 54. Waring, M. 1975. Ethidium and propidium. pages 141-65 In: Antibiotics HI: Mechanism of action of antimicrobial and anitumor agents., J. Concoran and F. Hahn, eds. Springer-Verlag: Berlin. Willaert, E. and A. Stevens. 1976. Indirect immunofluorescent identification of Acanthamoeba causing meningoencephalitis. Path. Biol. 24:545-547. Willaert, E., A. Stevens, and G. Healy. 1978. Retrospective identification of Acanthamoeba culbertsoni in a case of amoebic meningoencephalitis. J. Clin. Path. 31:717-720. Williams-Ashman, H. and A. Pegg. 1981. Aminopropyl group transfers, In: Polyamines in biology and medicine. D. Morris and L. Marton, eds. Marcel Dekker:New York, 407 ppg. Williams-Ashman, H. and J. Seidenfeld. 1986. Aspects of the biochemical pharmacology of methylglyoxal bis(guanylhydrazone). Biochem. Pharmacol. 35:1217-25. Wright, P., T. Byers, D. Crossdoersen, P. McCann, and A. Bitonti. 1991. Irreversible inhibition of S-adenosylmethionine decarboxylase in Plasmodium falciparum- infected erythrocytes - growth inhibition in vitro. Biochem. Pharm. 41:1713- 1718. Xie, Q., C. Tabor, and H. Tabor. 1988. Spermidine biosynthesis in Escherichia coli: promoter and termination regions of the speED operon.. J. Bacteriol. 171(8): 4457-65. Zhu, C., A. Cumaraswamy, and H. Henney. 1989. Comparison of polyamine and S- adenosylmethionine content of growing and encysted Acanthamoeba isolates. Mol. Cell. Biochem. 90:145-53. Zubay, G. 1983. Biochemistry. Addison-Wesley: Reading, MA. 1268 pp. Zwick, M., M. Wiggs, and M. Paule. 1991. Sequence and organization of 5S RNA genes from the eukaryotic ptotist Acanthamoeba castellanii. 101:153-157. Gene. Appendix A OGMA z a DGM-11 2/1 Yeast Extract 35 glucose 15 Proteose peptone 75 magnesium sulfate 0.25 glucose 15 calcium chloride 0.0074 magnesium sulfate 0.25 potassium phosphate 0.27 calcium chloride 0.0074 ferric citrate 0.03 potassium phosphate 0.27 EDTA 0.01 ferric citrate 0.03 AMINO ACIDS arginine 1.0 pH 6.8 glycine 1.5 isoleucine 0.6 leucine 0.9 methionine 0.3 valine 0.7' histidine 0.2 lysine 1.0 phenylalanine 0.9 threonine 0.5 tryptophan 0.2 VITAMINS Biotin 0.2 b 12 0.001 thiamine 1 TRACE METALS (mg/ml) Zinc sulfate 1.0 Manganese chloride 2.3 molybdate 0.4 cobalt chloride 0.017 copper sulfate 0.0033 boric acid 0.1 EDTA 0.01 164 Appendix B Tryptic Fragments of Bovine serum albumin. Peak No. Retention Time Area Percent Area 1 2.481 96045 .593 2 2.901 117519 .725 3 3.282 107141 .661 4 4.826 260969 1.610 5 13.970 1144148 7.059 6 17.300 542551 3.347 7 19.927 16601 .102 8 21.844 206160 1.271 9 22.083 97910 .604 10 22.405 152306 .939 11 23.814 323553 1.996 12 24.514 314518 1.940 13 26.165 254542 1.570 14 27.500 488736 3.015 15 29.048 256575 1.583 16 29.365 512850 3.164 17 30.265 417174 2.574 18 31.210 206761 1.276 19 31.598 523228 3.228 20 32.519 210366 1.298 21 32.925 188070 1.160 22 33.190 381798 2.356 23 33.810 316969 1.956 24 34.405 358691 2.213 25 35.740 310804 1.918 26 36.853 250804 1.547 27 37.800 263037 1.623 28 38.414 313730 1.936 29 38.959 1024191 6.319 30 39.901 599419 3.698 31 40.624 596044 3.677 32 41.463 317521 1.959 33 42.050 352504 2.174 34 42.675 519168 3.203 165 166 Tryptic fragments of cytochrome C Peak Number Retention time ______Area______Percent Area 1 2.325 8027 2.698 2 2.967 4981 1.674 3 3.515 4206 1.414 4 4.780 2445 0.822 5 8.112 1029 0.345 6 10.900 1144 0.385 7 13.827 15879 5.338 8 17.039 164879 55.424 9 19.525 1525 0.513 10 20.610 1737 0.584 11 22.286 3932 1.322 12 22.678 3857 1.300 13 38.880 8058 2.709 14 40.369 18499 6.218 15 41.640 7451 2.505 16 47.425 21760 7.315 17 50.017 7740 2.602 18 52.425 701 0.236 19 55.690 5360 1.802 20 57.139 6184 2.079 21 58.446 4703 1.581 22 60.236 3388 1.139