CHARACTERIZATION OF NOVEL POST-TRANSLATIONAL MODIFICATIONS OF MICROTUBULE ASSOCIATED PROTEINS AND IMMUNOLOGICALLY RELATED MODIFICATIONS OF DNA TOPOISOMERASE II
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate School of
The Ohio State University
By
Min Ding, M.S.
The Ohio State University 1996
Dissertation Committee: Approved by
Dale D. Vandre, Ph.D., Adviser
Robert M. DePhilip, Ph.D. Adviser Amanda A. Simcox, Ph.D. Molecular, Cellular and Developmental Biology Kenneth H. Jones, Ph.D. Graduate Program UMI Number: 9639227
UMI Microform 9639227 Copyright 1996, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 ABSTRACT
As a component of the cytoskeleton, microtubules (MTs) play indispensable
roles during the cell cycle and neuronal development. MT dynamics and functions
are regulated by a group of proteins called microtubule associated proteins (MAPs),
which in turn are regulated by post-translational modifications. It has been
established that MAPs are phosphorylated in a cell cycle dependent and/or
developmentally regulated fashion. One phosphorylation site, recognized by the
monoclonal antibody MPM-2, is found on both MAP4 in mitotic cells and MAP1B in
differentiating PC12 cells. MPM-2 antibody also recognizes mitotic DNA topoisomerase II, an enzyme required during mitosis. This study documents work on post-translational modification of MAPs as well as the phosphorylation site shared by MAPs and topoisomerase II. It is shown here that MAPI, MAP2, and
MAP4 are glycosylated. In particular, MAP2 and MAP4 are modified by a single O- linked N-Acetylglucosamine (O-GlcNAc), adding MAPs to the family of intracellular
O-GlcNAc modified proteins. At least 10% of brain MAP2 is modified by O-GlcNAc.
Multiple O-GlcNAc modification sites exist on MAP2, and these are all located in the
MAP2 projection domain. The phosphorylation of MAPs was examined using a series of synthetic phosphopeptides based upon the MPM-2 phosphoepitope shared between certain MAPs and topoisomerase II. The most essential element
ii of the MPM-2 epitope identified here consists of an aromatic amino acid -2 residues
N-terminal to the phosphorylated threonine. An aromatic amino acid at the +2 position is also important for the antibody recognition. This sequence is indeed the in vivo MPM-2 epitope as substantiated by a rabbit polyclonal antibody generated against the topoisomerase II synthetic phosphopeptide. This affinity purified antibody, PTE1, recognized a band that co-migrates with DNA topoisomerase II on immunoblots of mitotic cells. Preincubation with the phosphopeptide, but not the dephosphopeptide, abolished PTETs ability to recognize this band. PTE1 also recognizes a distinctive group of proteins in mitotic HeLa cell lysates.
Immunocytochemical localization has shown that PTE1 specifically stains mitotic chromosomes, spindle poles and midbodies. These results indicate that proteins recognized by the PTE1 antibody share phosphoepitopes recognized by the MPM-2 antibody, including those present on several MAPs. To My Parents
Boqi Ding and Mindi Shen,
Whose Unconditional Support Is the Foundation of All My
Achievements
and To My Wife
Hong Lu,
Whose Boundless Love Makes My Life Complete ACKNOWLEDGMENTS
Looking back on the past twenty-nine years, joining Dr. Dale D. Vandre’s lab in the summer of 1993 was one of the wisest decisions I have ever made.
Under his guidance, I have matured from a rookie in scientific research to a semi independent investigator. I am grateful for his confidence in me, and his willingness to let me explore new ideas and experiment with new approaches. I admire his style of scientific investigation and his philosophy of graduate training. What I have learned from him is invaluable now, and will be invaluable throughout my career.
Dr. Vandre has been both a mentor and a friend, sharing with me his experience and insights not only in biological research but also in career development and personal life. His advice and support have helped make some of my personal aspirations come true. I also want to thank him for his patience and painstaking effort in correcting the linguistic errors in my manuscripts. Without his support, this dissertation would not be possible.
I wish to thank Professors Robert M. DePhilip, Kenneth H. Jones, John M.
Robinson, and Amanda A. Simcox for their constant interest in my research and valuable suggestions. I want to especially thank Professor Amanda A. Simcox, who had acted kindly as my temporary adviser during my first year at OSU. I wish to thank past and current graduate students in Dr. Vandre's lab, Dr.
Yunhi Choi, Mrs. Yang Feng, Mrs. Christine Kondratick, and Mr. Colin Lowry, for their companionship during my endeavor here. Especially I want to thank Yunhi for the MAP4 samples she generously provided, and Yang for allowing me to steal her reagents from time to time. I enjoyed Christine’s sincere friendship, and my conversations with Colin that made some monotonous intervals between experiments interesting. I also want to thank my friends who helped to make
Columbus our home away from home, Dr. Timothy Cain, Mr. Chong Fu, Mr. Yiping
Jia, Mr. Qiang Tong, and Mr. Guoshan Tsen. Among the many memorable moments we have shared together are our passionate discussions of science and our five-thousand-mile cross-country trip, and everything in between.
I am deeply indebted to my parents, Boqi Ding and Mindi Shen, for their unconditional love and tremendous sacrifice. They have always put their sons' needs before theirs, and have done everything in their power to make sure we grow up physically, intellectually, and morally sound. They have exemplified to us the meaning of love without reservation, pride in endeavors disregarding results, and perseverance before the most daunting obstacles. I am forever grateful for the solid foundation they molded inside me, upon which, I build everything I have and will have. I am also grateful for my grandfathers. One started the family tradition of love, self-sacrification, and respect for knowledge. The story of the other, whom I have never met, from an illiterate blacksmith’s son to a self-made factory manager
vi more than half century ago, has always inspired me for the best.
Finally, I want to thank my wife, Hong Lu, for your unwavering love and support. I thank you for taking care of most of our mundane chores in your busy schedule to allow me more time for my aspirations. You help me to unwind when the going gets tough, and to focus when my imagination deviated too far from the reality. I am particularly grateful for your understanding and/or tolerance of my many unconventional thoughts, hobbies, and aspirations. I am lucky to have you as my wife, and I am looking forward to share life’s ups and downs with you for the many years to come. VITA
March 13, 1967 ...... Born - Shanghai, China
1989 ...... Bachelor of Science, Genetics and Genetic Engineering, Fudan University, Shanghai, China
1989-1991 ...... Assistant Engineer (Biotechnology), Jiangxi National Pharmaceutical Company, Nanchang, PRC
1991 -1992 ...... Graduate Teaching Assistant, Department of Biology, Northeast Louisiana University, Monroe, Louisiana
1995...... Master of Science, Program in Molecular, Cellular and Developmental Biology, The Ohio State University Columbus, Ohio
1992-present ...... Graduate Teaching and Research Associate Program in Molecular, Cellular and Developmental Biology, The Ohio State University Columbus, Ohio
PUBLICATIONS
1. Ding, M., Robinson, J.M., Burry, R.B., and Vandre, D.D. Human Phagocytic Leukocytes Have an Extremely Dynamic Microtubule Array. (1993) Molecular Biology of the Cell 4, 451 a. [abstract]
viii 2. Ding, M., and Vandre, D.D. Evidence For the Glycoprotein Nature of High Molecular Weight MAPs. (1994) Molecular Biology of the Cell 5, 167a. [abstract]
3. Ding, M., Vandre, D.D., Behrens, B.C., and Robinson, J.M. The Microtubule Cytoskeleton in Human Leukocytes is a Highly Dynamic Structure. (1995) European Journal of Cell Biology 66, 234-245.
4. Ding, M. and Vandre, D.D. High Molecular Weight Microtubule Associated Proteins Contain O-linked N-Acetylglucosamine. (1996) Journal of Biological Chemistry 271(21), 12555-12561.
FIELDS OF STUDY
Major Field: Molecular, Cellular and Developmental Biology TABLE OF CONTENTS
Page
ABSTRACT...... ii
DEDICATION...... iv
ACKNOWLEDGMENTS...... v
VITA...... viii
LIST OF DIAGRAM...... xii
LIST OF TABLES...... xiii
LIST OF FIGURES...... xiv
LIST OF ABBREVIATIONS...... xvi
INTRODUCTION...... 1
CHAPTERS:
1. HIGH MOLECULAR WEIGHT MICROTUBULE-ASSOCIATED PROTEINS CONTAIN O-LINKED N-ACETYLGLUCOSAMINE.
INTRODUCTION...... 11 MATERIALS AND METHODS...... 15 RESULTS...... 21 DISCUSSION...... 30 DIAGRAM AND FIGURES...... 39
2. PARTIAL CHARACTERIZATION OF THE MPM-2 PHOSPHOEPITOPE.
INTRODUCTION...... 52 MATERIALS AND METHODS...... 56 RESULTS...... 60 DISCUSSION...... 67 TABLE AND FIGURES...... 72
3. CHARACTERIZATION OF A NOVEL MITOSIS SPECIFIC ANTIBODY
INTRODUCTION...... 87 MATERIALS AND METHODS...... 92 RESULTS...... 97 DISCUSSION...... 105 TABLE AND FIGURES...... 112
SUMMARY...... 130
REFERENCES...... 136
APPENDIX A:
THE MICROTUBULE CYTOSKELETON IN HUMAN PHAGOCYTIC LEUKOCYTES IS A HIGHLY DYNAMIC STRUCTURE
ABSTRACT...... 144 INTRODUCTION...... 145 MATERIALS AND METHODS...... 149 RESULTS...... 153 DISCUSSION...... 163 ACKNOWLEDGMENTS...... 171 REFERENCES...... 172 TABLES AND FIGURES...... 175 LIST OF DIAGRAM
DIAGRAM PAGE
1. Detection of O-GlcNAc Modification ...... 39
xii LIST OF TABLES
TABLES PAGE
1. Peptide constructs ...... 72
2. Comparison Between MPM-2 and PTE1 reactive proteins ...... 112
3. Fixation protocols ...... 175
4. Microtubule numbers in human leukocytes ...... 176 LIST OF FIGURES
FIGURES PAGE
1. Glycoprotein detection using biotin-hydrazide/streptavidin following metaperiodate oxidation ...... 40
2. Galactosyltrqansferase labeling and glycosidic linkage analysis of MAP2 and MAP4 ...... 42
3. Chromatographic analysis of the p-elimination products from galactosyltransferase labeled MAP2 and ovalbumin ...... 44
4. Stoichiometry of O-GlcNAc modified MAP2 ...... 46
5. Localization of the carbohydrates on MAP2 ...... 48
6. Tryptic mapping of rat MAP2 ...... 50
7. Recognition of the P2-KLH conjugate by the MPM-2 antibody ...... 73
8. Competition of synthetic peptides for MPM-2 antibody binding to mitotic MAP4...... 75
9. Competition of synthetic peptides for MPM-2 antibody binding to the P2-KLH conjugate ...... 77
10. MPM-2 antibody binding to modified synthetic peptide-KLH conjugates..79
11. Competition of modified phosphopeptides for MPM-2 binding to the P2-KLH conjugate ...... 81
12. Binding of the MPM-2 antibody to synthetic peptide constructs ...... 83
13. Model of the MPM-2 phosphoepitope ...... 85
14. Immunoblot analysis of polyclonal topoisomerase II peptide antibodies ...... 113
xiv 15. PTE1 recognizes DNA topoisomerase II ...... 115
16. PTE1 specifically recognizes the phosphorylated topoisomerase II peptide on immunoblot ...... 117
17. The phosphorylated topoisomerase II peptide could inhibit binding of the PTE1 IgG...... 119
18. The PTE1 antigens could be partially removed by alkaline phosphatase treatment ...... 121
19. PTE1 recognizes rat brain M API ...... 123
20. Immunocytochemical localization of the PTE1 antibody and MPM-2.. 125
21. Localization of the PTE1 antigens during various mitotic stages ...... 128
22. Comparison of microtubule staining patterns obtained in human leukocytes using different fixation protocols ...... 177
23. Microtubule staining obtained in different human leukocytes ...... 179
24. Representative field of microtubule staining in a mixed population of leukocytes ...... 181
25. Microtubule turnover in monocytes and neutrophils ...... 183
26. Regrowth of microtubules in monocytes and neutrophils following release from nocodazole ...... 185
27. Reorganization of microtubules in neutrophils following stimulation with the chemotactic peptide fMLP ...... 187
28. Taxol induced microtubule bundles are present in the leukocytes of patients undergoing taxol chemotherapy ...... 189
xv LIST OF ABBREVIATIONS
DMEM Dulbecco’s Modification of Eagle’s Medium
DNA DeoxyriboNucleic Acid
ECL Enhanced ChemiLuminescence
FITC Fluorescein IsoThio Cyanate
FPLC Fast Protein Liquid Chromatography
HeLa Human epithelial carcinoma cell line
HMW MAP High Molecular Weight Microtubule Associated Protein
HPLC High Performance Liquid Chromatography
JAR Human choriocarcinoma cell line kDa,kD Kilo Daltons
KLH Keyhole Limpet Hemocyanin
MAP Microtubule Associated Protein
MAPK, p42mapk Mitogen-Activated Protein Kinase
MPM-2 Mitotic Protein Monoclonal antibody 2
MT MicroTubule
MTOC MicroTubule Organization Center
O-GlcNAc single O-linked N-Acetylglucosamine
ON Overnight
xvi PBS Phosphate Buffered Saline
PC12 Rat Pheochromocytoma cell line
PTE1 Phosphorylated Topoisomerase II Epitope 1
PVDF PoiyVinylidine DiFluoride
RT Room Temperature
SDS-PAGE Sodium Dodecyl Sulfate-PolyAcrylamide Gel Electrophoresis
TBS Tris Buffered Saline
xvii INTRODUCTION
One of the most magnificent features of life is its ability to reproduce itself.
Without reproduction, life as we know it would not exist on earth. The reproduction of a complicated multicellular organism begins as a fertilized egg. The remarkable transformation from a single fertilized egg to an individual is completed through a series of selected cell divisions and controlled cellular differentiation. Cell division is not only the foundation of reproduction, it also plays a crucial role during the life of an individual. For example, our bodies constantly produce new blood cells through cell division of stem cells in the bone marrow, which is necessary to replenish the population of cells circulating in the blood stream. In response to injury, the skin will initiate a repair process which involves the reproduction of various cell types. Each cell division must also be carried out properly to ensure the correct distribution of genetic material to each progeny cell. Thus, cell division is a centerpiece of life on earth.
Cell division can be described in terms of the cell cycle. Each cell cycle is composed of two distinctive stages, mitosis and interphase. Interphase can be further subdivided into G1 phase, S phase, and G2 phase. G1 phase starts after a
1 cell exits from the previous mitosis. A distinct point in the G1 phase, designated as
GO, functions in a fashion similar to that of a rest area on highways, where a cell
could temporarily exit the cell cycle loop. The next stage, S phase, is the period
when DNA is replicated. After S phase, cells enter G2 phase to finish other
preparative metabolic activity before mitosis. The ensuing mitosis can be further
subdivided into four specific stages. The first stage is called prophase. Major events
observed during prophase include chromosome condensation and the nuclear
envelope breakdown. Following nuclear envelope breakdown, the microtubule
network is reorganized to form the mitotic spindle. During the next stage of mitosis,
metaphase, chromosomes are attached to the spindle poles via the kinetochore
microtubules and are moved to a position centrally located between the two spindle
poles. Anaphase follows metaphase and starts with the abrupt splitting and
separation of sister chromatids, which then move to the opposite spindle poles. The final stage of mitosis, telophase, is marked by reformation of the nuclear envelope, the decondensation of chromosomes, and the disassembly of the mitotic spindle.
Finally, the cell completes cytokinesis and splits into two identical daughter cells.
The fundamental purpose of mitosis is to separate the sister chromatids into two daughter cells. At the beginning of mitosis, the diffuse interphase chromatin becomes gradually condensed and forms the distinctive chromosomes. Each chromosome contains two sister chromatids replicated during S phase. The sister chromatids pair with each other along their long axis and are linked together
2 primarily near a special sequence of their DNA called centromeres. Each centromere is associated with a pair of kinetochores, which function to attach each chromatid to one of the two spindle poles through an interaction with spindle microtubules. During metaphase, when chromosomes are lined up along the central plate, one kinetochore of each chromosome faces opposite spindle poles. At the onset of anaphase, sister chromatids are abruptly separated from each other, and are moved towards the opposite spindle poles. Finally, the nuclear envelopes are re-formed around the two groups of separated chromosomes, and chromosome decondensation returns them to their dispersed interphase conformations. This complex process of chromosome condensation, decondensation, and sister chromatid splitting entails the coordination and effort of many cellular components, one of which is DNA topoisomerase II.
DNA topoisomerase II belongs to a group of enzymes called DNA topoisomerases. DNA topoisomerases have evolved to resolve the topological dilemmas arising during various stages of DNA metabolism, and they are classified as type I or II according to their biological functions. DNA topoisomerase I can transiently break a single strand of DNA, pass the intact single strand through the break point, and reseal the break. On the other hand, DNA topoisomerase II will introduce a double-strand break and pass an intact double strand through the breakage before resealing it (Roca, 1995).The ability of topoisomerase II to knot/unknot and catenate/decatenate a closed circular double-strand DNA molecule
3 makes it indispensible during mitosis when extensive chromosome conformational changes occur. It has been shown that DNA topoisomerase II is required for the separation of sister chromatids during anaphase (DiNardo et al., 1984; Holm et al.,
1985; Rose and Holm, 1993). It is also believed that topoisomerase II plays an important role in chromosome condensation during prophase, and possibly chromosome decondensation during telophase {Uemura et al., 1987; Hirano and
Mitchison, 1993). DNA topoisomerase II is known to be phosphorylated in vivo in a cell cycle regulated manner {Wells and Hickson, 1995), and phosphorylation on some sites is shown to enhance its enzymatic activities to different extents (DeVore et al., 1992; Corbett et al., 1992, 1993a, 1993b). This correlates with the observation that the greatly elevated topoisomerase II activity during mitosis could not be explained simply by increased topoisomerase II synthesis. In most cell types examined, topoisomerase II protein levels vary only two to three fold during the cell cycle, peaking in G2 and M phase (Gasser et al., 1986; Cardenas et al., 1992;
Whalen et al., 1991).
The formidable task of separating sister chromatids into two daughter cells also requires the highly coordinated spatial and temporal formation and function of the mitotic spindle. The principal components of the mitotic spindle are the microtubules (Vandre and Borisy, 1985; Kirschner and Mitchison, 1986). During interphase, microtubules (MTs) usually have their minus end embedded in the centrosome, and undergo growth and shrinkage at their plus end. The MTs radiate
4 from the centrosome to the periphery of the cell in all directions forming a cytoplasmic array of fibers. The centrosome, also known as the microtubule organization center (MTOC) in animal cells, is composed of a pair of centrioles surrounded by the amorphous pericentriolar material (PCM) (Kimble and Kuriyama,
1992). Though centrioles undergo replication during S phase, the two pairs of centrioles remain within the same PCM and function as a single MTOC through out interphase. At the beginning of prophase, the centrosome splits into two and separates, with each daughter centrosome composed of a pair of centrioles and half of the original PCM. Shortly after separation, the MT nucleating capacity of the centrosomes increases tremendously (Kuriyama and Borisy, 1981). The dynamic properties of MTs also undergo significant change after the cell enters prophase.
The average half-life of MTs has decreased from around 5 minutes to 15 seconds
(Salmon et al., 1984; Saxton et al., 1984). These changes in nucleation capacity and dynamics result in the reorganization of the cytoplasmic MT array and formation of the spindle characterized by a larger number of MTs that are shorter in overall length (Kirschner and Mitchison, 1986). Numerous MTs grow from each centrosomes and form two astral structures, and the MTs from these asters are able to interact with the chromosomes after nuclear envelope breakdown. The entire mitotic MT array and associated proteins form the bipolar mitotic spindle. The two centrosomes are located at opposite ends of the spindle and are referred to as the spindle poles. The kinetochores, an aggregate of various proteins attached to the centromeres, provide the anchor sites where MTs are captured and stabilized. The
5 two kinetochores of the same chromosome will bind to the MTs originating from different spindle poles (Rieder, 1982). As a result of balancing forces exerted via
MTs and MT associated motor proteins, the chromosomes congress to a point centrally located between the two spindle poles at metaphase, with each of the kinetochores facing the opposite spindle poles. During anaphase, the separated sister chromatids are transported along the MTs towards the corresponding spindle poles. The nuclei of the two daughter cells form around the separated chromosome masses following the completion of anaphase.
The drastic decrease in MT half-life during mitosis, among other properties of MTs, is regulated by a group of proteins called microtubule associated proteins
(MAPs) (Hyams and Lloyd, 1994). MAPs are generally divided into two categories based upon their functional properties, the motor and non-motor proteins. The motor proteins include the kinesin superfamily and the dynein family, the non-motor proteins category includes MAPI A, MAP1B, MAP2, MAP3, MAP4 and tau proteins.
Both motor and non-motor MAPs are believed to be involved in the organization of the spindle structure and the separation of sister chromatids. Similar to DNA topoisomerase II, MAPs are regulated through post-translational modification
(Olmsted, 1986). Various lines of evidence have shown that phosphorylation of
MAPs can alter their affinity for MTs and thus affect MT dynamics (Murthy and
Flavin, 1983; Ookata et al., 1995). It is also believed that phosphorylation of MAPs may regulate their interaction with other proteins. The MT nucleating capacity of the
6 centrosome is also regulated by the phosphorylation state of associated proteins
{Centonze and Borisy, 1990), which includes several MAPs.
Most interestingly, topoisomerase II and MAPs (MAPI and MAP4) share a common phosphoepitope recognized by a mitosis specific antibody called MPM-2
(Vandre et al., 1986, 1991; Taagepera et al., 1993). The MPM-2 mouse monoclonal antibody was raised against mitotic HeLa cell extracts and was selected for its preferential staining of mitotic cells (Davis et al., 1983). It has been shown that the
MPM-2 epitope involves a phosphate. On Western blots, the MPM-2 antibody recognizes an array of proteins that are only reactive during mitosis. Using indirect immunofluorescent staining techniques, MPM-2 antigens could be localized to many important components of the mitotic apparatus, including chromosomes, kinetochores, centrosomes and midbodies (Vandre et al., 1984; Hirano and
Mitchison, 1991). Microinjection of the MPM-2 antibody into fertilized Xenopus eggs and HeLa cells resulted in the blockage of the onset and the completion of M- phase, indicating that the MPM-2 reactive proteins play an indispensable role during mitosis (Davis et al., 1989; Kuang et al., 1989). Despite intense research interest in MPM-2 phosphorylation, little is known about the essential amino acids that constitute the MPM-2 epitope. Efforts have also failed to reveal the in vivo MPM-2 phosphorylation site on most MPM-2 reactive proteins, including MAPI, MAP4, and
DNA topoisomerase II.
7 The overall goals of this study were to investigate novel post-translational
modifications of MAPs, as well as the characterization of the mitotic-phosphorylation
site shared by MAPs and topoisomerase II, namely the MPM-2 phosphoepitope.
Specific aims include 1), to investigate the potential glycosylation of various MAPs;
2), to characterize the essential components of the MPM-2 phosphoepitope in vitro\
and 3), to generate antibodies against the phosphopeptide derived from the
tentative topoisomerase II MPM-2 site in a effort to verify that the site is
phosphorylated in vivo; and 4), to characterize the topoisomerase phospho-peptide
specific antibody and compare it to the MPM-2 antibody.
Chapter 1 describes the work related to the investigation of potential
glycosylation of MAPs. Evidence of glycosylation on rat brain M A P I, MAP2, and
HeLa cell MAP4 are presented. Although glycoproteins are usually found on the cell
surface or within lumenal compartments, a unique form of glycosylation exists
almost exclusively on cytoplasmic and nuclear proteins. This so-called O-GlcNAc
modification involves a single N-Acetylglucosamine O-linked to either serine or threonine residues (Hart et al., 1989; Haltiwanger et al., 1992). Almost all known O-
GlcNAc modified proteins share two important characteristics. First, they are also
phosphorylated in a regulated manner; secondly, they form reversible multimeric
complexes (Haltiwanger et al., 1992). It is believed that O-GlcNAc modification
plays an important role in regulating protein functions, similar to phosphorylation.
Chapter 1 describes the identification of O-GlcNAc modification on rat brain MAP2
8 and HeLa cell MAP4, and the stoichiometric analysis and chromatographic mapping of O-GlcNAc modification on rat brain MAP2. The idenfication of these saccharide moieties on HMW MAPs should provide a new perspective in understanding the regulatory mechanisms controling the functional properties of MAPs and how they interact with MTs.
Chapter 2 describes efforts used in characterizing the essential elements of the MPM-2 phosphoepitope. A series of synthetic phosphopeptides derived from a tentatively identified MPM-2 site on DNA topoisomerase II were used to test a model of the MPM-2 epitope formulated from known MPM-2 sequences. It is shown here that while the aromatic amino acids at either the -2 or +2 positions {relative to the phosphorylated threonine) are important for its MPM-2 reactivity, substitution of the +1 position amino acid to alanine does not affect the phosphopeptide’s affinity for MPM-2 antibody. The results presented here clarify ambiguities around the consensus definition of an MPM-2 recognition site. This information should facilitate the identification of other MPM-2 reactive proteins in vivo and their corresponding
MPM-2 reactive sequences, which in turn will shed further light on mitotic regulation in general.
Although the topoisomerase II phosphopeptide was shown to be MPM-2 reactive in vitro, the question remains as to whether this sequence is phosphorylated in vivo. To answer this question, rabbit polyclonal antibodies were
9 raised against the synthetic peptides prepared from the topoisomerase II sequence in both their dephosphorylated or phosphorylated forms. Chapter 3 describes the work of generation, purification, and characterization of these antibodies. The antibody raised against the phosphorylated peptide, PTE1, recognizes a band in immunoblots of a mitotic HeLa lysate that co-migrates with topoisomerase II. The antibody lost its ability to recognize this band if preincubated with the phosphopeptide, but not the dephosphorylated peptide. Limited dephosphorylation by alkaline phosphatase also indicated the presence of a phosphate within the
PTE1 epitope. Most significantly, PTE1 only recognized the phosphorylated form of the peptide on the immunoblots. Based on this evidence, it is highly likely that this topoisomerase II sequence is phosphorylated in vivo and is the site responsible for the MPM-2 recognition of endogenous topoisomerase II. Furthermore, PTE1 specifically recognizes a group of proteins in mitotic lysates, which is similar to the pattern of MPM-2 reactive proteins in these samples. Using indirect immunofluorescent staining, it is also shown that PTE1 stains mitotic chromosomes as well as other important mitotic structures including spindle poles and midbodies.
These observations indicate that PTE1 is a unique marker of mitotic phosphorylation , similar but not identical to the MPM-2 antibody. It is expected that
PTE1 will serve as a useful tool for analyzing mitosis specific phosphorylation on
PTE1 reactive proteins, helping to elucidate 1), which proteins are mitotically regulated; 2), what the regulatory/phosphorylation sites on these proteins are; and
3), the identity of the corresponding kinases.
10 CHAPTER 1
INTRODUCTION
Microtubules (MTs) are one of the major components of the cytoskeleton, and play an important role in the organization of the cytoplasm. As an example, disruption of MTs by various MT destabilizing agents results in major changes in cytoplasmic organization, including collapse of intermediate filaments, and redistribution of the Golgi apparatus and the endoplasmic reticulum (Soltys and
Gupta, 1992). MTs are also involved in such diverse cellular functions as maintenance of cell shape, movement of eukaryotic cilia and flagella, formation of the mitotic spindle, and regulation of organelle distribution and vesicle movements
(Hyams and Lloyd, 1994). Regulation of the dynamic properties of MTs is thought to play a role in many of these processes (Kirschner and Mitchison, 1986).
A group of proteins that bind to MTs in vivo and copurify with MTs, collectively defined as MT associated proteins (MAPs), modulate MT dynamics and function. MAPs can be categorized into two major classes according to their primary function, i.e., motor proteins, which include the kinesin superfamily and the dynein family, and non-motor proteins, which are traditionally further divided into the
11 high molecular weight MAPs (HMW MAPs, includes MAPI A, MAP1B, MAP2,
MAP4) and the lower molecular weight tau proteins. Among the identified in vivo functions of MAPs in neuronal cells, kinesin and cytoplasmic dynein are responsible for anterograde and retrograde axonal transport respectively, while tau and MAP2 appear to be required for initial neurite growth as shown using antisense mRNA
(Caceres and Kosik, 1990; Dinsmore and Solomon, 1991). It is well-known that the functions of MAPs can be regulated through phosphorylation. For example, phosphorylation of MAPs can alter their affinity for MTs in vitro and affect their ability to stabilize MTs (Olmsted, 1986). Further, the phosphorylation of MAP4 has been coupled to the onset of mitosis in vivo, and this phosphorylation may play a functional role in regulating MT dynamics during the interphase to mitosis transition
(Ookata et al., 1995). Post-translational modification of MAPs other than phosphorylation may also play a role in regulating MAP function. However, little is known about other potential forms of MAP modification.
During the past decade a unique glycosylation of cytoplasmic and nuclear proteins has been characterized and identified as an O-GlcNAc modification. This modification is comprised of a single N-acetylglucosamine O-linked to either serine or threonine residues (Hart et al., 1989; Haltiwanger et al., 1992a). Virtually all of the O-GlcNAc-modified proteins identified to date are also phosphorylated, and they form reversible multimeric complexes in a regulated manner. Analogous to reversible phosphorylation reactions catalyzed by specific kinases and
12 phosphatases, the corresponding O-GlcNAc transferases and glycosidases have also been identified (Haltiwanger et al, 1992b; Dong and Hart, 1994). It has been proposed that O-GlcNAc modifications may play a regulatory role similar to that of phosphorylation. A recent report has shown that O-GlcNAcs on the associated 67 kDa protein (p67) of eukaryotic initiation factor 2 (elF-2) were indispensable for its ability to protect elF-2 from inactivation following phosphorylation by elF-2 kinase
(Chakraborty et al., 1994). Interestingly, this rapidly growing list of O-GlcNAc- modified proteins include several cytoskeletal components, namely, cytokeratin 13
(King and Hounsell, 1989), cytokeratin 8 and 18 (Chou et al., 1992) and neurofilaments (Dong et al., 1993). This prompted us to ask if glycosylation, O-
GlcNAc modification in particular, could be another general means of post- translational modification of MAPs.
We report here that the HMW MAPs are glycosylated as indicated by their labeling with biotin-hydrazide following periodate oxidation. The biotin-hydrazide labeling method is a common technique used to detect saccharide moieties on proteins (Wilchek and Bayer, 1987). More importantly, both MAP2 and MAP4 contain O-linked N-acetylglucosamine residues in the monosaccharide form, adding these MAPs to the growing list of O-GlcNAc-modified proteins. We have further shown that the biotin-hydrazide-labeled carbohydrate and the O-GlcNAc modification are both localized to the MAP2 projection domain. Analysis of MAP2 tryptic digests indicates that there may be several O-GlcNAc modification sites on
13 MAP2. Our findings have revealed a previously unknown form of post-translational
modification on HMW MAPs, which may prove to be critical in understanding their
intracellular functions, especially their interactions with MTs and other cellular components.
14 MATERIALS AND METHODS
Materials
The [6-3H]gtucosamine (27 Ci/mmol) was obtained from ICN Biomedicals,
Inc. The ECL Glycoprotein Detection System was obtained from Amersham, and the O-GlcNAc Detection Kit was purchased from Oxford GlycoSystems, Inc. Bovine milk galactosyltransferase, UDP-galactose, UDP-[6-3H]galactose (15.3 Ci/mmol) were obtained from Sigma. ENTENSIFY™ Universal Autoradiography Enhancer used in fluorography was from Du Pont NEN Research Products. The GlcNAc(31-
4GlcNAc standard was obtained from Seikagaku America, Inc. BCA protein assay reagents were purchased from Pierce, Human plasma thrombin was obtained from
Calbiochem. The PepRPC HR 5/5 column and FPLC system were obtained from
Pharmacia.
Preparation of microtubule associated proteins
Fresh rat brains were homogenized in 1.5 volumes of cold PEM buffer (0.1
M PIPES-NaOH, 1 mM EGTA, 1 mM MgS04, pH6.6) followed by centrifugation at
30,000 g in a JA-20 rotor (Beckman J2-21 Centrifuge) for 15 min at 4°C. The supernatant was centrifuged again at 180,000 g in a Ti-60 rotor (Beckman L8-M
Ultracentrifuge) for 90 min at 4°C. The supernatant from the second centrifugation
15 was recovered, and taxol and GTP were added to a final concentration of 20 /j.M and 1 mM, respectively. The sample was warmed to 37°C for 15 min followed by centrifugation at 37°C at 30,000 g in a JA-20 rotor for 30 min. The rat MT pellet was washed in warm PEM buffer containing 20 jj,M taxol and 1 mM GTP and resuspended in PEM.
Rat MAP2 and tubulin were purified from the rat MT pellet by adding NaCI to a final concentration of 0.35 M. After incubation in a boiling water bath for 8 min, the sample was centrifuged at 15,800 g {Microcentrifuge 5402, Brinkmann
Instruments, Inc.) at 4°C for 15 min. Heat stable MAP2 was recovered in the supernatant. Alternatively, after addition of NaCI to 0.35 M, the taxol-stabilized rat
MTs were incubated at 37°C for 15 min followed by centrifugation at 15,800 g at
37°C for 15 min. The salt extracted pellet was resuspended in PEM and used as taxol-stabilized tubulin.
HeLa MAP4 was kindly provided by Yunhi Choi in this lab, and was purified essentially as described by Vallee and Collins (Vallee and Collins, 1986).
Glycoprotein Detection
Carbohydrates on MAPs were labeled using the ECL Glycoprotein Detection
System following the manufacturer’s protocol with some modifications. Briefly,
16 samples were separated on SDS-PAGE and transferred to nitrocellulose membranes, followed by reduction with 1 mg/ml sodium borohydrate in TBS (154 mM NaCI, 10 mM Tris-Base, pH 7.4) to remove endogenous reacting groups (6 times, 15 min each time). Sodium metaperiodate was used to oxidize the vicinal hydroxyls of carbohydrate moieties, resulting in aldehyde generation. The samples were subsequently treated with biotin-hydrazide to incorporate biotin onto the oxidized carbohydrate. Finally, the biotin was detected by horseradish peroxidase conjugated streptavidin using enhanced chemiluminescence (ECL).
Galactosyltransferase Labeling
Bovine milk galactosyltransferase was autogalactosylated prior to use
(Roquemore et al., 1994). MAP samples were usually incubated at 0°C for 3 hr in labeling buffer (10 mM HEPES-NaOH, 10 mM galactose, 5 mM MnCI2, pH 7.3) containing 1 ptC\ UDP-[3H]galactose and 30 mU p-galactosyltransferase at a final volume of 45 ul. After the reactions were stopped by addition of ethylenediaminetetraacetic acid (EDTA) to 10 mM, a portion of the samples was separated on SDS-PAGE and visualized by fluorography. Alternatively, the remaining samples were incubated at 37°C overnight either with p-elimination buffer
(1 M NaBH4, 0.1 M NaOH), p-galactosidase, or distilled water, and then analyzed by adsorption to polyvinylidine difluoride (PVDF) membrane and assayed according to manufacturer’s instructions using the O-GlcNAc Detection Kit.
17 Carbohydrate Composition Analysis
Rat MAP2 and ovalbumin were labeled with galactosyltransferase as described above. Samples were then dialyzed against 0.2 M NaCI to remove free
UDP-[3H]galactose and lyophilized. Samples were either incubated with 300 ^1 (3- elimination buffer at 37°C for 18 hr followed by neutralization with ice-cold 4 M acetic acid or with distilled water as a control. The reaction mixtures were lyophilized and washed twice with 1 M acetic acid/methanol in the presence of excess galactose to remove residual borate. Finally, samples were applied to Bio-
Gel P-4 column (-250 mesh, 0.9x50 cm) at 16 ml/h in 0.2 M ammonium acetate.
The column’s Vo and V, volumes were established by using cytochrome C and galactose, respectively. [3HjGlucosamine and GlcNAcpi-4GlcNAc were also used to calibrate the column. Galactose was detected by phenol sulfuric acid assay
(Dubois et al., 1956). After separation on the P-4 column, an equal volume was removed from each eluted fraction and [3H] was measured in a scintillation counter.
The mole concentration of rat MAP2 was calculated from its absorbance at 562 nm using the BCA protein assay. After the galactosyltransferase labeling reaction, the incorporated radioactivity was detected by PVDF membrane adsorption assay, and the total moles of conjugated [3H]galactose was deduced using the specific activity of the UDP-[3H]galactose.
18 Thrombin Digestion and Microtubule Binding Assay
Rat MAP2 was digested by thrombin according to Joly et al. (Joly et al.,
1989) at a final concentration of 32 U/ml, and reactions were stopped by addition of phenylmethylsulfonyl fluoride to 1 mM. Part of the reaction mixture was separated by SDS-PAGE and analyzed by the Glycoprotein Detection System; the remaining part was tested for microtubule binding ability as described below.
Thrombin digested MAP2 was mixed with taxol stabilized rat tubulin. After incubating at 37°C for 15 min in the presence of 1 mM GTP and 20 pM taxol, the samples were centrifuged at 10,000 g for 10 min. The pellets were resuspended in SDS sample buffer after the supernatant had been collected. The whole mixture, supernatant and pellet, were then analyzed by SDS-PAGE and visualized by coomassie brilliant blue staining. Some rat MAP2 was labeled by [3H]-galactose, as described earlier, prior to thrombin digestion under the same conditions. These samples were then separated by SDS-PAGE, stained with Coomassie Blue, and analyzed by fluorography.
Tryptic Mapping of MAP2
Rat brain MAP2 was labeled with [3H]-galactose as described above, and was dialyzed against 0.2 M NaCI followed by distilled water. The sample was then
19 lyophilized and processed for trypsin digestion at an enzyme:substrate ratio of 1:40
(w/w) at 37°c for 24 hr. The peptides were then separated on a PepRPC HR 5/5 reverse phase column using the Pharmacia FPLC (Fast Protein Liquid
Chromatography) system. The peptides were bound to the column in 0.1% trifluoroacetic acid in FPLC-grade water and eluted with an increasing gradient of acetonitrile. A constant level of 0.1% trifluoroacetic acid was present in all solvent mixtures. Elution of peptides was monitored by A215, and the radioactivity in each fractions was then analyzed by liquid scintillation counting.
SDS-PAGE and Immunoblot Analysis
Samples were analyzed by SDS-PAGE and immunoblot as described earlier
(Vandre et al., 1991) with the following modifications. Gradient gels (4-12% or 7-
17% acrylamide) were used for gel electrophoresis, and proteins were transferred to nitrocellulose paper with a 0.2 um pore size at 250 mA for 1 hr in transfer buffer containing SDS (25 mM TRIS-base, 192 mM glycine, 20% methanol, 0.05% SDS); a second transfer was at 150 mA for 15 min using the transfer buffer without SDS.
Nitrocellulose paper was incubated in 10% (v/v) heat-inactivated horse serum in
TBS for 1 hr to block non-specific binding sites, rinsed with TBS, and incubated with various primary antibodies for 1 hr. After washing in TBS, the transfer was incubated with peroxidase-conjugated goat anti-mouse antibody, washed, and visualized by ECL.
20 RESULTS
HMW MAPs are glycosylated.
In recent reports, tau, a low molecular weight MAP, was shown to be non-
enzymatically glycated in the brain tissue of Alzheimer’s disease patients (Ledesma
et al, 1994; Yan et al., 1994). In addition it has been reported that claustrin, a
chicken brain homologue of MAP1B, is a keratan sulfate-containing protein (Burg
and Cole, 1994). Further, it was shown that rat brain MAP1B was also sensitive to
keratanase digestion, which hydrolyzes the p-galactosidic linkages present in
keratan sulfate. These reports suggested that post-translational modifications other than phosphorylation might also occur on MAPs.
In order to determine whether glycosylation might be another common post-
translational modification of HMW MAPs, we employed a well-established
glycoprotein detection method (Wilchek and Bayer, 1987). In essence, this
approach is based upon the oxidation of the vicinal hydroxyls on carbohydrate
moieties by periodate, and the subsequent formation of aldehyde groups. The
newly generated aldehyde groups are available to react with various hydrazide
probes, the binding of which can be detected following completion of the reaction.
Using a biotin-hydrazide/streptavidin detection system, we observed that MAPI A,
MAP1B and MAP2 present in rat brain MT samples were all labeled following
21 periodate oxidation (Fig. 1, MT). The specificity of this reaction was confirmed by the negative results obtained in the experiments omitting either periodate oxidation
(Fig. 1. MT Per -) or biotin-hydrazide conjugation (Fig. 1, MT Hyd -). Isolation of heat stable MAP2, following boiling of the rat brain MT preparation, confirmed the reaction of this MAP with the biotin-hydrazide carbohydrate detection system (Fig. 1,
MAP2). These results indicate that brain HMW MAPs contain some form of carbohydrate modification. Further, MAP4 isolated from HeLa MT preparations was also strongly labeled by the biotin-hydrazide (Fig. 1, MAP4). Thus, the presence of carbohydrate residues on HMW MAPs was not restricted to brain samples, but also included the major MAP found in non-neuronal cells.
MAP2 and MAP4 contain O-linked non-reducing terminal N- acetylglucosamine.
Although cytoplasmic proteins rarely contain complex carbohydrate modifications, recent evidence shows that there is a unique O-GlcNAc modification found almost exclusively on cytoplasmic and nuclear proteins, including some cytoskeletal components (Haltiwanger et al., 1992b). In an effort to determine which carbohydrate components existed on MAPs, we examined HMW MAPs following a standard approach used for the detection of O-GlcNAc (Diagram 1).
22 The isolated MAPs were labeled using a UDP-[3H]galactose- galactosyltransferase system which specifically transfers [3H]galactose to a terminal non-reducing N-acetylglucosamine. Labeled samples were then separated by SDS-
PAGE, stained with coomassie, and the stained samples were subjected to fluorography. Ovalbumin containing N-linked terminal non-reducing N- acetylglucosamine was used as a control (Fig.2A, lanes t and 4). Our results showed that both mitotic and interphase MAP4 isolated from HeLa cells were labeled {Fig. 2A, lanes 2 and 5, 3 and 6), and that rat brain MAP2 was also heavily labeled by [3H]galactose (Fig. 2A, lanes 7 and 8). Rat brain MAPI was only weakly labeled by [3H]galactose (data not shown). In addition to the MAP4 band, another heavily-labeled band was present in both of the MAP4 samples. These additional labeled bands were minor coomassie staining bands, suggesting that this band could be a breakdown product of MAP4, or alternatively, it could be another heat stable HeLa MAP that contains a high amount of N-acetylglucosamine.
Interestingly, both MAP4 and the lower molecular weight protein showed a higher degree of labeling in the mitotic sample, suggesting a potential cell cycle dependent regulation of this glycosylation.
In order to prove that the label was covalently linked to the protein via O- glycosidic bonds, equal amounts of the labeled samples were incubated with either p-galactosidase or (3-elimination buffer overnight. The radioactivity retained by the samples following this treatment was analyzed after binding of the protein to PVDF,
23 and were compared to control samples that were incubated with distilled water only
(Fig. 2B). p-galactosidase, an enzyme that specifically cleaves terminal non reducing galactose from carbohydrate, removed at least 70% of the incorporated radioactivity from MAP2, mitotic MAP4 and interphase MAP4. Furthermore, p- elimination, which cleaves O-linked carbohydrate from protein but leaves N-linked carbohydrate intact, removed 90% of the radioactivity from MAP2, and 70% from both mitotic MAP4 and interphase MAP4. These data indicated that MAP2, mitotic
MAP4 and interphase MAP4 all contain O-linked terminal non-reducing N- acetylglucosamines. On the other hand, ovalbumin showed high sensitivity to p- galactosidase but resistance to p-elimination, as expected for a protein containing
N-linked terminal N-acetylglucosamine (Fig. 2B).
MAP2 contains GlcNAc directly O-linked to the protein.
Since the rat brain MAP2 sample was relatively pure, and the labeling was highly specific (Fig. 2A, lanes 7 and 8), further characterization of the carbohydrate associated with MAPs focused on MAP2. In order to ascertain that MAP2 contained
O-GlcNAc directly linked to the protein, we needed to provide evidence that the labeled carbohydrate was a monosaccharide instead of an O-linked polysaccharide containing terminal GlcNAc residues. Briefly, MAP2 was labeled with [3H]galactose using galactosyltransferase, dialyzed, and incubated either with water as control or with p-elimination buffer to remove covalently-bound carbohydrate. The samples
24 were then analyzed by high resolution Bio-Gel P-4 chromatography. If the original carbohydrate residue was a monosaccharide, the expected p-elimination product would be [3H]Gaipi-4GlcNAcitol. This product should migrate slower on the Bio-Gel
P-4 than the disaccharide control GlcNAcpi-4GlcNAc, since two simple monosaccharides behave as one GlcNAc on the Bio-Gel P-4 column (Kobata,
1994). On the other hand, if the original carbohydrate contains a terminal GlcNAc in addition to one or more monosaccharide residues, the [3H]galactose-containing
p-elimination product would migrate at least at the same position as GlcNAcpt-
4GlcNAc, if not faster. A labeled complex carbohydrate containing terminal GlcNAc
residue would be excluded from the column resin and migrate with the void volume.
Ovalbumin was used as a control, and as expected, P-elimination did not significantly change the elution position of the [3H]galactose-containing peak (Fig.
3A). As shown in Fig. 3B, the radioactivity associated with the MAP2 control sample also eluted at the Vo position. However, the [3H]galactose-labeled fraction obtained after p-elimination of the MAP2 sample migrated as a single peak between
GlcNAcpi-4GlcNAc and glucosamine. These results exclude the possibility that the labeled terminal GlcNAc on MAP2 was associated with a complex polysaccharide moiety. We thus conclude that the original carbohydrate moiety on MAP2 is an O- linked N-acetylglucosamine monosaccharide. Similar results were obtained with
[3Hjgalactose-labeled MAP4 (data not presented).
25 In order to determine the extent of O-GlcNAc modification, identical amounts of MAP2 were galactosyltransferase labeled for different time periods at 0°C, and the total incorporated radioactivity was measured (Fig. 4A). The reaction proceeded rapidly, and by 180 min all of the available GlcNAcs were labeled. Higher temperature (37°C) or longer incubation time (up to 5 hr) did not generate a significant difference in the total incorporated radioactivity (data not shown). The
180-min-samples were used to deduce the maximum amount of [3H]galactose incorporated. The average ratio obtained from eight separate experiments was one
O-GlcNAc in every eleven MAP2 molecules (Fig. 4B). While it is possible that this number reflects the in vivo stoichiometry, some O-GlcNAcs may have been cleaved by glycosidase activity during the purification process or may have been inaccessible to the galactosyltransferase. Therefore, this result probably represents a minimal estimate of the O-GlcNAc modification of rat brain MAP2.
MAP2 is glycosylated on its projection domain.
MAP2 is a rod-like molecule composed of a C-terminal binding domain and a 200 kDa N-terminal projection domain that protrudes away from MT surface. A protease sensitive region (PSR) exists between these two domains, and chymotrypsin (Vallee, 1980) or thrombin (Joly et al., 1989) digestion of MAP2 will generate small binding fragments between 28-39 kDa and large projection fragments between 140-240 kDa (Fig. 5A). It has been shown that phosphorylation
26 occurs within both the projection and binding domains (Vallee, 1980), and extensive phosphorylation on MAP2 reduces its ability to promote microtubule assembly or bind to preformed MT (Ookata et al., 1995).
In order to elucidate if glycosylation may affect MAP2 binding to MT in a similar manner, we examined the location of the glycosylation sites on MAP2.
Briefly, rat MAP2 was digested with thrombin and the various proteolytic fragments were examined either for their MT binding ability or the presence of carbohydrate.
Comparison of the results allowed us to determine whether binding or projection fragments of MAP2 contained carbohydrate. Galactosyltransferase and the biotin- hydrazide/streptavidin system were both used to track the carbohydrate containing fragments. As expected after limited proteolysis, several high molecular weight projection fragments (PF) of MAP2 were generated that lost their MT binding ability and remained in the supernatant, while several low molecular mass binding fragments (BF) around 30 kDa cosedimented with MT after incubation with taxol- stabilized tubulin (Fig. 5B). Most of the large projection fragments of MAP2 contained both the [3H]galactose-labeled O-GlcNAc residues and the biotin- hydrazide reactive carbohydrate, while none of the small MT binding fragments were labeled with either reagent (compare Fig. 5 B and C). The largest proteolytic fragment retained MT binding ability and was also glycosylated, presumably containing both the binding domain and a partial projection domain due to incomplete proteolysis. Since carbohydrate was absent from the binding domain.
27 It is unlikely that they directly regulate the MT binding ability of MAP2; instead, the carbohydrate modification might play a role in maintaining MAP2 conformation or mediating its interaction with other cellular components.
MAP2 contains as many as three sites of O-GlcNAc modification.
Although the average ratio of O-GlcNAc modification on the isolated heat stable rat brain MAP2 samples was less than 1 GlcNAc per protein molecule, it was possible that multiple sites on the protein could be O-GlcNAc modified. In order to address this possibility, we digested [3H]-galactose-labeled MAP2 with trypsin and separated the proteolytic fragments using a reverse-phase column by FPLC.
After extensive digestion of MAP2 by trypsin, the resulting peptides were separated on the reverse phase column using a linear acetonitrile gradient (See Materials and
Methods). The amount of tritium recovered in each fraction was analyzed by scintillation counting. Two independent experiments yielded identical results, and the distribution of radioactivity from one experiment is shown in Fig.6. Three labeled peaks were observed eluting at 25%, 33%, and 38% of acetonitrile respectively. The third peak was the most heavily-labeled, suggesting several possibilities: first, this peptide may have contained an O-GlcNAc modification site that was more accessible to the galactosyltransferase; secondly, this site was more commonly modified in vivo; or thirdly, more than one O-GlcNAc modification site existed on this peptide. The tryptic mapping results suggest that at least three sites on rat brain
28 MAP2 are capable of being modified by O-GlcNAc. Since the rat brain MAP2 was obtained from the whole brain homogenate, it is possible that different populations of MAP2 molecules exist in vivo, some of which are glycosylated while others are not, analogous to phosphorylation.
29 DISCUSSION
The results presented here demonstrated that HMW MAPs can be considered as glycoproteins as indicated by both their reactivity to biotin-hydrazide following periodate oxidation and labeling with [3H]galactose following galactosyltransferase incubation. Using galactosyltransferase we have shown that brain MAP2 and HeLa MAP4 incorporate [3H]galactose demonstrating the presence of terminal non-reducing GlcNAc residues. Further, we have shown that the
GlcNAc residues on MAP2 are directly O-linked to the protein as a monosaccharide.
MAP4 is also modified by single GlcNAc residues as the final p-elimination product of MAP4 eluted as a single peak at the same position as the p-elimination product of MAP2 (data not presented). Our results also suggested that there was an increase in the level of the O-GlcNAc modification in the mitotic MAP4 sample as compared to the interphase sample. This observation is similar to a previous report which showed that keratin 8 and 18 both have an elevated O-GlcNAc modification level in mitotic arrested human epithelial cells (Chou and Omary, 1993), and may indicate a cell cycle dependent regulation of the O-GlcNAc modification on MAP4.
In an effort to ascertain the location of the saccharide moiety associated with the MAPs, we generated fragments of MAP2 following thrombin digestion. The brief thrombin induced digestion of MAP2 generated a limited number of high molecular weight bands that did not bind to taxol-stabilized MTs. The complexity of the bands
30 was probably due to incomplete digestion of the MAP2 sample. All the high
molecular weight projection fragments could be labeled by both
galactosyltransferase and biotin-hydrazide. However, the comparative labeling
intensity of the individual bands was obviously different. This may simply reflect the
inability of galactosyltransferase to access all of the available O-GlcNAc residues,
or it could indicate that in addition to the simple O-linked GlcNAc residues identified
by the galactosyltransferase, there are other carbohydrate modifications of the MAP
detected by the biotin-hydrazide reaction. Tryptic mapping of the [3H]-galactose
labeled rat brain MAP2 using reverse phase FPLC showed that three distinct peaks
were present. Therefore, at least three different sites on the projection domain of
MAP2 may be O-GlcNAc modified. Confirmation of this result will require
identification of the O-GlcNAc-modified peptides and sequence analysis.
Despite the presence of multiple modification sites, we obtained results
suggesting a low overall stoichiometry of O-GlcNAc on the MAP2. We believe that this labeling density may be an underestimate of the in vivo amount of O-GlcNAc on MAP2 for several reasons. First, our detection methods using galactosyltransferase might only label a portion of the total O-GlcNAcs present on the MAP, due to inaccessibility of the galactosyltransferase; secondly, no precautions were taken to preserve the glycosylation state of the protein during purification, and some of the O-GlcNAcs could have been lost due to the exposure of the protein to glycosidases in the brain extract; and thirdly, there might be
31 different sub-populations of glycosylated MAPs in vivo. Whether such sub populations of MAP2 exist remains to be determined, however. It is also possible that a single MAP2 molecule could contain multiple O-GlcNAc modification sites in vivo as well.
Evidence for glycosylation, both in the form of O-GlcNAc and more complex carbohydrate, have also been demonstrated in other proteins associated with the cytoskeleton. Cytokeratins 13 (King and Hounsell, 1989), 8 and 18 (Chou et al.,
1992) have all been shown to contain the O-GlcNAc modification. In addition, four keratin isoforms present in human keratinocytes have been demonstrated to be reactive to several monoclonal antibodies raised against keratan sulfate, and immunocytochemical analysis using these keratan sulfate antibodies confirms that keratin filaments are one source of cellular immunoreactivity in these keratinocytes
(Schafer and Sorrell, 1993). Neurofilament subunits NF-L and NF-M have also been shown to contain O-GlcNAc, with stoichiometries similar to those reported here for MAP2 of 0.1 and 0.15 mole GlcNAc/mole protein for NF-L and NF-M respectively (Dong et al., 1993). It should be noted, however, that the level of modification determined for the neurofilament proteins was only obtained after analysis on Dionex CarboPAc PA1 of the GlcNAcs released following acid hydrolysis. When measuring the ratio of incorporated galactose to proteins, as we have reported here for MAP2, levels of only 0.011 and 0.028 mole Gal/mole protein for NF-L and NF-M respectively were obtained (Dong et al., 1993). Thus, the ratio
32 of modification we obtained for MAP2 should also be considered as the minimal degree of MAP2 modification.
Various studies with O-GlcNAc-modified proteins have revealed that the O-
GlcNAc residues provide different functions in different proteins (Haltiwanger et al.,
1992a). These functional properties could be dependent on the particular glycosylation site, requiring site specific transferases analogous to phosphorylation consensus sequences. One of the putative functions of O-GlcNAc is to regulate protein phosphorylation. For example, O-GlcNAc modification may modulate phosphorylation by occupying or blocking the phosphorylation site. Mutually exclusive modifications of phosphorylation and glycosylation have been shown in the C-terminal of RNA Polymerase II, this fragment was either phosphorylated or
O-GlcNAc modified, but not both (Kelly et al., 1993). This mutually exclusive model is supported by the notion that the identified O-GlcNAc modification sites on many proteins resemble a number of proline-directed kinases sites. Interestingly, the
HMW MAPs contain several consensus proline-directed kinase phosphorylation sites, and MAP4 has been shown to be a substrate for p34cdc2 (Ookata et al., 1995).
Alternatively, the O-GlcNAc modification of one protein could regulate phosphorylation on a second protein, as in the case of O-GlcNAc modified p67.
When deglycosylated, p67 loses its ability to protect the elF-2 a-subunit from phosphorylation (Chakraborty et al., 1994). O-GlcNAc modification does not necessarily inhibit phosphorylation, however, as both phosphorylation and O-
33 GlcNAc modification levels on keratin 8 and 18 in mitotic-arrested human colonic cell line HT29 were higher compared to control cells (Chou and Omary, 1993). In addition to phosphorylation, it has been hypothesized that O-GlcNAc may be crucial for intracellular protein-protein recognition in a manner similar to lectins, or in regulation of the susceptibility of a protein to protease (Haltiwanger et al., 1992a).
Some or even all of these putative roles for O-GlcNAc could be potentially important for MAP2 and MAP4 functions. Both MAP2 and MAP4 have been shown to interact with many cytoplasmic components in addition to MT. With regards to
MAP4, a recent report has shown that cyclin B binds to MAP4 (Ookata et al., 1995).
Thus, bound cyclin B localizes p34cdc2 kinase, a potential regulator of M-phase MT dynamics, to the MT. Furthermore, other kinase activities have been found to associate with other MAPs (Theurkauf and Vallee, 1982). How the binding of these cellular components to MAPs is regulated remains largely undefined. It is known that MAP2 and MAP4 are phosphorylated at multiple sites and in a regulated manner. The MAP phosphorylation state might in turn control the interaction between MAPs and other cellular components. The phosphorylation state of MAPs could in part be regulated by O-GlcNAc modification as observed for other 0-
GlcNAc modified proteins. Alternatively, the O-GlcNAc on MAP2 and MAP4 may directly regulate the binding and/or phosphorylation of MAP-associated-proteins, such as kinases, that may be crucial for the biological functions of these associated proteins.
34 From the results that we have obtained, it would appear that glycosylation of the HMW MAPs is a general post-translational modification of these proteins.
Previous studies have provided evidence suggesting that MAPs might be glycosylated in vivo. It has been reported that tau, a low molecular weight MAP, is present in a glycated form in paired helical filaments that are components of neurofibrillary tangles in brain tissue of Alzheimer's disease patients (Ledesma et al., 1994; Yan et al., 1994). It was also reported that glycation of tau decreased the binding affinity of tau proteins for tubulin, which is another characteristic of tau obtained from Alzheimer’s tissue. Claustrin, a chicken brain homologue of MAP1B, has been shown to be sensitive to keratanase treatment, and the protein was reported to be reactive with monoclonal antibodies raised against cartilage keratan sulfate (Burg and Cole, 1994). In addition, claustrin was shown to incorporate
[3H]glucosamine when added to the culture medium of the chick glial cultures, further indicating its glycoprotein nature. Briones and Wiche (Briones and Wiche,
1985), demonstrated that certain antibodies specific for MAPI and MAP2 identified a sulfoglycoprotein component of the extracellular matrix secreted by 3T3 cells.
While these results were interpreted as cross-reaction of these MAP antibodies with a distinct extracellular glycoprotein, it is also possible that the sulfoglycoprotein antigen detected in the extracellular matrix was a secreted form of MAP. This would correlate with the hypothesis that the keratin sulfate containing proteoglycan claustrin might represent a secreted form of MAP1B (Burg and Cole, 1994). It has also been suggested that MAP1B may exist as a transmembrane cell surface protein in rat cortical cell cultures and may be involved in synaptogenesis
35 (Muramoto et al., 1994). All these reports indicate the possibility that HMW MAPs
may exist in different populations which may differ in their amino acid composition,
post-translational modification (phosphorylation and glycosylation) and/or
distribution.
We have also found that HMW MAPs, such as MAP2 and MAP4, exhibit a
high degree of sensitivity to keratanase digestion (Ding and Vandre, unpublished
observations). However, our observations indicate that keratanase digestion did not
generate a stable core protein for each of the HMW MAPs. We also observed that the addition of keratan sulfate to the reaction mixture failed to inhibit the keratanase
digestion of MAPs. Further, the chemical deglycosylation of MAPs with
trifluoromethanesulfonic acid did not induce an observable molecular weight shift
in the protein following SDS-PAGE analysis. We were also unable to stain rat brain
MAPs with the keratan sulfate specific antibody 5D4 (Ding and Vandre, unpublished
results). Thus, it appears unlikely that MAP2 or MAP4 contain complex
carbohydrate modifications, such as has been reported for claustrin. Indeed, if
other HMW MAPs contain complex carbohydrates like keratan sulfate, it would be
against prevailing models of glycosylation. Complex glycoproteins generally exist
either at the cell surface or within lumenal compartments. Nevertheless, several
reports have indicated that some glycoproteins are present in the nucleus and
cytosol. For example, it has been shown that while chondroitin sulfate
proteoglycans are exclusively extracellular in 7 d postnatal brains, they become
predominantly cytoplasmic in the adult brains (Aquino et al., 1984a; Aquino et al.,
36 1984b). The question of how various complex carbohydrate structures have reached the nucleus and/or cytoplasm has been answered in at least one case. A unique type of nuclear heparan sulfate has been shown to be originally synthesized by conventional glycosylation pathways onto heparan sulfate proteoglycans which are subsequently secreted. After being specifically endocytosed, the free heparan sulfate is released from the proteoglycans, further modified, and transported into the nucleus (Aquino et al., 1984a). Tentative complex carbohydrate moieties on HMW
MAPs such as MAP1B could be added via similar pathways. Alternatively, HMW
MAPs could be glycosylated by an unknown cascade of giycosyltransferases in the cytosol. In fact, several glycosyltransferase activities have been demonstrated in the cytosol and nucleus (Hart et al., 1989).
The biological functions of complex carbohydrates on cytoplasmic and nuclear proteins are poorly understood. However, it has been suggested that they might serve to stabilize proteins in an otherwise unfavorable conformation. HMW
MAPs have long projection domains that are thought to extend away from the surface of the MT. It is possible that carbohydrate moieties associated with the projection domain help maintain this extended conformation. Carbohydrate moieties located within the projection domain could also prevent untimely proteolysis of this extended region of the protein within the cytoplasm.
37 In summary, this report has shown for the first time that glycosylation is a common post-translational modification shared by HMW MAPs. In addition, we have added MAP2 and MAP4 to the growing list of O-GlcNAc-modified proteins.
We are currently in the process identifying the sites modified by O-linked GlcNAc on MAP2. Further work will be necessary to investigate the keratanase sensitive feature of the HMW MAPs, and to determine the functional consequences of MAP
O-glycosylation.
38 |i -Galaclosyiliansleiase k t UL>f-(JH J-Galactose
—Q-^^Polypeptnte^)
* G lc N A c 1“ G lcN A c
t
Chromatographic Analysis
Diagram 1 Detection of O-filcNAc Modification
39 Fig.1. Glycoprotein detection using biotin-hydrazide/streptavidin following metaperiodate oxidation.
Rat brain MTs {MT) were separated by 4% SDS-UREA gel and transferred to nitrocellulose. Nitrocellulose transfers were then analyzed using the Glycoprotein
Detection Kit as described in Materials and Methods. Control experiments were done by omitting either the sodium metaperiodate oxidation step (Per -) or the biotin-hydrazide conjugation step (Hyd -). The position of MAPI A, MAP1B, and
MAP2 as determined by coomassie staining and immunoblot analysis (data not presented), are indicated as 1A, 1B, and 2 respectively. Boiled MAP2 and boiled
MAP4 were separated by 4% SDS-UREA gel and 7.5% SDS-PAGE respectively.
They were either stained with coomassie brilliant blue (C) or analyzed by the
Glycoprotein Detection Kit (G). This highly specific reaction indicated that MAPI,
MAP2 and MAP4 contained carbohydrate moieties.
40 MT Hyd + - + MAP2 MAP4 Per — + + C G C G
. , ^ 1A — m m - 1B “ ^ 2
FTGURF: 1
41 Fig.2. Galactosyltransferase labeling and glycosidic linkage analysis of MAP2 and MAP4.
Samples were galactosyltransferase labeled as described in Materials and
Methods. A, labeled samples were separated by either 7-17% (lanes 1-6) or 7%
(lanes 7-8) SDS-PAGE, and stained by coomassie brilliant blue (C) followed by fluorography analysis (FI) of the same gels. Lanes 1 and 4: ovalbumin; lanes 2 and
5: mitotic MAP4; lanes 3 and 6: interphase MAP4; lanes 7 and 8: rat brain MAP2.
B, labeled ovalbumin (1), MAP2 (2), mitotic MAP4 (3) and interphase M A P 4
(4) were incubated at 37°C for 18 h either with p-galactosidase ( I ), with (3-elimination buffer ( H ), or with distilled water ( ■ ) as control.
Radioactivity retained after various treatments was plotted as a percentage of the total radioactivity in control samples. Note the high sensitivity of MAP2 and MAP4 to 3-elimination buffer as compared to ovalbumin.
42 FI C FI B
100
£ 80
8 0
4 0
u 20
I TC.IJRI- 2
43 Fig.3. Chromatographic analysis of the p-elimination products from galactosyltransferase-labeled MAP2 and ovalbumin.
Galactosyltransferase-labeled ovalbumin (A) and MAP2 (B) were analyzed by Bio-Gel P-4 chromatography following incubation with (□--□) or without ( *-■
) p-elimination buffer. The column’s VD and V; volumes were established by using cytochrome C and galactose, respectively. Glucosamine (2) and GlcNAcpi-
4GlcNAc (1) were also used to calibrate the column. The volume of each fraction is 0.63 ml. The p-elimination product of galactosyltransferase-labeled MAP2 migrated as a single peak between glucosamine and GlcNAcpi-4GlcNAc (B), as expected for an O-GlcNAc-modified protein.
44 */' *u_ lY , ' DTmt-Cn-rn ; 20 30 40 50 60
Fraction Number
Vo 1 2Vi Y .. /'S
rffj.s -tlji TnWi 20 30 40 50 60
Fraction Number
FIGURI-; 3 Fig.4. Stoichiometry of O-GlcNAc modified MAP2.
An equal amount of rat brain MAP2 was galactosyltransferase-labeled for 1,
3, 6, 15, 90 or 180 min, and the total incorporated radioactivity was detected by the
PVDF adsorption assay. The 180-min-samples were used to estimate the mole ratio of O-GlcNAc to protein. A, a representative graph of galactosyltransferase labeling reactions on MAP2. B, the average mole ratio of incorporated [3H]galactose to
MAP2, calculated from 8 experiments.
46 Incubation Time(mlnute)
rat MAPZ ratio (mol gal/mol protein] 0.091 ± 0.01 G
f - ' I G U R F . 4 Fig.5. Localization of the carbohydrates on MAP2.
Rat MAP2 was incubated with either distilled water or thrombin at 37°C for
1 h. The samples were subsequently analyzed for their MT binding ability (B), or tested for the presence of carbohydrates (C). The projection fragments (PF) and binding fragments (BF) are indicated. A, schematic drawing of MAP2. PSR: protease sensitive region. B, taxol stabilized tubulin was used to test the MT binding ability of undigested MAP2 (lanes 3-5) or thrombin digested MAP2 (lanes 6-8). Lane
1: molecular weight marker (97, 66, 55, 42, 40 and 30 kD); lane 2: taxol stabilized tubulin. The whole mixture (T), pellet (P) and supernatant (S) were separated on 4-
12% SDS-PAGE and stained with coomassie. C, samples were galactosyltransferase labeled, separated on 4-12% SDS-PAGE, and analyzed by fluorography (FI) following coomassie staining (C). Alternatively, samples were separated by SDS-PAGE, transferred to nitrocellulose paper, and analyzed by the
Glycoprotein Detection Kit (Gl). Lanes 1, 3 and 5: undigested MAP2; lanes 2, 4 and
6: thrombin digested MAP2. The results indicate that the projection domain of MAP2 contains the carbohydrate moieties. Also, note a slight difference in labeling patterns between the [3H]galactose labeled fragments and the biotin-hydrazide labeled fragments suggesting that multiple carbohydrate moieties may be present.
48 A
MAP2 PSR ______X NH* —| ■ COOH / \ M T binding domain
B 12 3 4 5 6 7 8 — - —MAP2
; lPF
^ * i i lBF T P S T P S C 12 34 5 6
► » -MAP2 «iJPF ?)PF
^ )BF |BF
C FI Gl
F I C U R T - : 5
49 Fig.6. Tryptic mapping of rat MAP2.
Boiled rat brain MAP2 was labeled by [3H]-galactose, lyophilized and digested with trypsin as specified in the Materials and Methods. The peptides were then applied to a PepRPC HR 5/5 FPLC column and eluted at a flow rate of 0.5 ml/min. The volume of each fraction is 0.5 ml. 30 ul solution from each fraction were analyzed by scintillation counting and the CPM of each fraction were plotted against the elution position and was superimposed with the concentration of acetonitrile.
Three [3H] containing peaks eluting at approximately 25%, 33%, and 38% of acetonitrile were obtained.
50 o % of Acetonitrife o o o o o co (0 CM O
o © o o © o o o i2 oi *} mo CHAPTER 2
INTRODUCTION
MPM-2 is a monoclonal antibody that was raised against mitotic HeLa cells and is characterized by its preferential staining of mitotic cells following indirect immunofluorescence localization (Davis et al., 1983). A dramatic increase in the levels of MPM-2 reactive proteins is observed during the G2/M transition in all eukaryotic species examined. Some of the MPM-2 antigens are present on important components of the mitotic apparatus, including chromosomes, kinetochores, centrosomes and midbodies (Hirano and Mitchison, 1991; Vandre et al., 1984). Among the identified cell cycle dependent MPM-2 reactive proteins are microtubule associated proteins 1 and 4 (MAPI, MAP4) (Vandre et al., 1986;
Vandre et al., 1991), DNA topoisomerase II (Taagepera et al., 1993a), p42mapk
(Taagepera et al., 1994), cdc25 (Kuang et al., 1994), m ytl (Mueller et al., 1995a), and weel (Mueller et al., 1995b). The importance of these MPM-2 antigens was confirmed following microinjection of the MPM-2 antibody into fertilized Xenopus eggs and HeLa cells, which blocked both the onset and the completion of M-phase
(Davis et al., 1989; Kuang et al., 1989). It was proposed that a single MPM-2 kinase was responsible for phosphorylating these cell cycle regulated MPM-2 reactive proteins. It was also hypothesized that this kinase, together with these MPM-2 reactive proteins, constitute an integral part of the M-phase regulatory mechanism.
Despite significant effort, the MPM-2 epitope and the putative MPM-2 kinase have not been clearly identified.
It has been established that the MPM-2 antibody recognizes a phosphoepitope. However, the other essential components of the MPM-2 epitope remain to be defined. A recent report indicated that the MPM-2 epitope contains a proline residue located on the C-terminal side of the phosphorylated threonine or serine residue (Westendorf et al., 1994). On the other hand, the MPM-2 site on
p 4 2 mapk was Shown to contain a phosphorylated threonine residue in the sequence
TEY {Taagepera et al., 1994). These results suggest that the proline residue C- terminal to the phosphorylated residues is not required for MPM-2 recognition. The tyrosine residue in this sequence of p42mapk is also phosphorylated in vivo, and limited mutagenesis of this site showed that the Y->F mutation maintained MPM-2 reactivity, while the Y->E mutation showed weakened reactivity. Thus, it was demonstrated that the phosphorylation of the tyrosine residue or the negative charge at that position was not required for MPM-2 reactivity. In addition, the downstream F/Y appeared to promote MPM-2 binding while E hindered binding
(Taagepera et al., 1994). Based upon these two reports, it is difficult to identify common elements that would be required to define an epitope recognized by the same monoclonal antibody.
53 We have carried out a comparative analysis of all the reported MPM-2 reactive sequences (Taagepera et al., 1994; Westendorf et al., 1994), and based upon this analysis we propose that the MPM-2 epitope is likely composed of three elements. Our model would predict that maximal MPM-2 antibody binding would be achieved if 1), the phosphorylated residue was a threonine; 2), that the phosphorylated amino acid was located near an aromatic amino acid to its N- terminal side; and 3), that either an aromatic or positively charged amino acid was located to its C-terminal side. To begin to test this model, we designed synthetic peptides based upon a potential MPM-2 reactive site associated with human topoisomerase II (Taagepera et al, 1993b). We have demonstrated that the phosphorylated form of a 13 amino acid synthetic peptide corresponding to this topoisomerase II site is MPM-2 reactive, while the corresponding dephosphopeptide is not recognized by the antibody. We have also shown that this synthetic phosphopeptide is sufficient to compete for antibody binding to a native MPM-2 antigen MAP4. We designed and synthesized three more phosphopeptides each containing a single amino acid change from the native sequence. These altered peptides allowed us to examine the contribution to the epitope of aromatic residues both N-terminal and C-terminal to the phosphothreonine residue. We show that the alteration of the aromatic residue to the N-terminal side nearly eliminated MPM-2 recognition. While change to the C-terminal aromatic residue also reduced antibody binding, this reduction was not to the same extent as the alteration of the N-terminal
54 aromatic residue. These results strongly support a role for these flanking aromatic residues in defining the MPM-2 epitope. Further, location of an aromatic amino acid to the N-terminal side of the phosphorylated residue may be critical for antibody recognition. Our findings provide the basis for a more complete analysis of the
MPM-2 epitope, and may be useful in identifying potential MPM-2 target sequences.
55 MATERIALS AND METHODS
Materials
Peptides P1 and P2 were synthesized by AnaSpec Incorporated (San Jose,
CA). Peptides P3, P4 and P5 were synthesized by Princeton BioMolecules
Corporation (Columbus, OH). The HPLC purity of these peptides was greater than
95%. MPM-2 mouse monoclonal antibody was a generous gift from Dr. Potu Rao
(Department of Chemotherapy Research, The University of Texas M.D. Anderson
Hospital and Tumor Institute, Houston, TX), or purchased from Upstate
Biotechnology Incorporated (Lake Placid, NY). Peroxidase-conjugated goat anti mouse antibody was purchased from Kirkegaard & Perry Laboratories
(Gaithersburg, Maryland). Keyhole limpet hemocyanin was obtained from
CalBiochem (La Jolla, CA), and m-maleimidobenzoyl-W-hydroxysuccinimide ester from Pierce (Rockford, Illinois). Prepared PD-10 Sephadex columns were obtained from Pharmacia (Uppsala, Sweden). Mitotic HeLa MAP4 was kindly provided by
Yunhi Choi (This lab) and was purified essentially as described (Vallee and Collins,
1986).
56 Conjugating peptides to KLH
Synthetic peptides were conjugated to keyhole limpet hemocyanin (KLH) using a modification of a described procedure (Coligan et al., 1994). Briefly, KLH
(5 mg) was dissolved in 0.5 ml Buffer A (0.01 M phosphate buffer, 0.9 M NaCI, 0.01
M EDTA, pH 7.0) and dialyzed against Buffer A overnight at 4°C. M- maleimidobenzoyl-/V-hydroxysuccinimide ester (MBS) was dissolved in dimethylformamide (15 mg/ml) immediately before use, and 70 pi was added to the dialyzed KLH. The mixture was gently stirred for 30 min at room temperature, at which time it was applied to a PD-10 column and eluted with Buffer B (0.05 M phosphate buffer, 0.9 M NaCI, 0.01 EDTA, pH 6.0). The first peak eluted (KLH-
MBS) was pooled and mixed with 5 mg of synthetic peptide that was dissolved in
1 ml of Buffer B. The pH was adjusted to 7.3 and sample was gently stirred for 3 h at room temperature. The reaction mixture was dialyzed twice against distilled water at 4°C, lyophilized, and stored at -20°C.
Dot Blot and Competition Experiments
A Bio-Dot™ apparatus (BioRad Laboratory, Hercules, CA) was used for applying samples to nitrocellulose paper. Samples were first diluted to a total volume of 50 pi in TBS (154 mM NaCI, 10 mM Tris-Base, pH 7.4), and added to each well. The samples were applied to the nitrocellulose under vacuum, and the
57 wells were then washed several times with a large volume of TBS. The
nitrocellulose was removed from the Bio-Dot™ apparatus. In cases where the
samples were peptides rather than peptide conjugates, the nitrocellulose was
incubated in 0.2% glutaraldehyde in PBS (140 mM NaCI, 1.5 mM KH3P0 4, 2.7 mM
KCI, 6.5 mM Na3HP04, pH 7.4) for 45 min after application of the peptide samples.
The unreacted aldehyde groups were then reduced by treatment with two changes
of NaBH4(1 mg/ml in TBS). The nitrocellulose was then rinsed in TBS and blocked
in 10% heat inactivated horse serum in TBS for 1 hr. Dot blots were incubated with
various dilutions of the MPM-2 antibody for 1 h, washed in TBS and then incubated
with the secondary antibody for 1 h. The blots were developed with 4-chloro-1-
naphthol. In the competition experiments, the procedures were the same as above
except that the MPM-2 antibody was preincubated with various amounts of synthetic
peptides for 1 h prior to incubation with the dot blot.
Image Analysis
Images of the developed dot blots were captured using an MTI-CCD72
camera equipped with a Nikon lens (AF Micro Nikkor 60 mm 1:2.8 D). All of the dot
blots quantified within each figure of the text were captured in the same image to
insure that the acquisition of data was internally consistent. The captured images were analyzed using Optimas™ Image Analysis Software (Optimas Corporation).
The integrated optical densities were extracted from the images and exported to
58 Microsoft Excel for Windows™ (Microsoft Corporation), and used to plot the corresponding charts.
59 RESULTS
The MPM-2 antibody recognizes a phosphopeptide conserved in human
DNA topoisomerase lla and 11(3.
We examined the sequences of the previously reported MPM-2 reactive
peptides and protein epitopes (Taagepera et al., 1994; Westendorf et al., 1994), in
an effort to identify similarities in their structures that may be related to antibody
recognition. The presence or absence of structurally related groups of amino acids, were determined for each residue of the MPM-2 reactive peptides on both the N- terminal and C-terminal sides of the phosphorylated amino acid. A group of amino acids was considered irrelevant to the MPM-2 epitope if none of the amino acids within the group appeared consistently. As a result of this analysis, we found that
90% of the sequences examined contained an aromatic amino acid (most commonly F or W and to a lesser extent Y) on the N-terminal side of the
phosphorylated T/S residue. Most frequently, this aromatic residue was located at the -2 position. The MPM-2 reactive sequences also contained either an aromatic amino acid or one or more positively charged amino acids (most commonly R, K) on the C-terminal side of the phosphorylated T/S residue. Threonine was clearly the most frequent amino acid located at the phosphorylation site. In addition, while the majority of the MPM-2 reactive peptides contained a proline residue adjacent to the phosphorylated residue on the C-terminal side, the presence of a proline was
60 not required at this position since the MPM-2 epitope site reported for p42mapk lacked this proline residue (Taagepera et al., 1994). Our analysis led us to hypothesize that the MPM-2 epitope would typically contain a phosphothreonine flanked by an aromatic amino acid located within 2 to 3 residues on both the N- and C-terminal sides.
In order to test the validity of this hypothesis, we designed a series of synthetic peptides that could be used to examine the potential contribution of specific amino acids to the MPM-2 epitope. We first needed to establish that a small synthetic peptide obtained from a defined MPM-2 reactive protein was capable of binding to the MPM-2 antibody. Further, this phosphopeptide needed to have an affinity for the MPM-2 antibody that was comparable to a native in vivo MPM-2 antigen. It has been reported that topoisomerase Ilex and lip were MPM-2 reactive in a cell cycle dependent fashion (Taagepera et al., 1993a). Thus, we prepared a synthetic peptide containing 13 amino acids based upon a putative MPM-2 epitope site in topoisomerase II (Taagepera et al., 1993b). An amide-cystine residue was added to the C-terminal end of the peptide to allow for conjugation to carrier protein
(Table 1, P1). The phosphorylated form of this peptide was also synthesized (Table
1, P2). This phosphopeptide met the three characteristics we proposed for an MPM-
2 epitope, it contained a phosphothreonine residue, an aromatic residue at the -2 position relative to the phosphothreonine, and an aromatic residue at the +2 position. These peptides were individually conjugated to KLH as described in
61 Materials and Methods and tested for their ability to react with the MPM-2 antibody.
While an unconjugated KLH control along with the P1-KLH conjugate showed no affinity for the MPM-2 antibody (see below, Fig. 10A, KLH and P1), the P2-KLH conjugate exhibited strong affinity for the MPM-2 antibody similar to that for a human MPM-2 reactive protein, mitotic HeLa cell MAP4 (Fig. 7A). We observed a titration of the MPM-2 antibody against a fixed amount of the P2-KLH conjugate
(Fig. 7C), similar to that observed using mitotic MAP4 (Fig. 7B).
Since the conjugation efficiency of each synthetic peptide with KLH may not be identical, it was not possible to use the peptide conjugate alone to quantitatively assess the level of MPM-2 reactivity. In addition, there was a slight possibility that certain amino acid/motifs on KLH may contribute to the MPM-2 reactivity of the peptide after conjugation. In order to obtain an unambiguous result, we tested the affinity of the unconjugated peptides for the MPM-2 antibody. Varying amounts of each peptide were preincubated with a constant amount of the MPM-2 antibody.
This peptide-antibody mixture was then incubated with nitrocellulose dot blots containing mitotic HeLa cell MAP4, a native MPM-2 reactive protein. If the unconjugated peptide could effectively bind to the MPM-2 antibody, it would compete for reaction with the immobilized MAP4. We demonstrated that the P2 phosphopeptide could effectively block the binding of the MPM-2 antibody to mitotic
MAP4 (Fig. 8A, P2). This inhibition was clearly dependent upon the amount of peptide preincubated with the MPM-2 antibody (Fig. 8B). On the other hand, the
62 nonphosphorylated P1 peptide failed to exhibit any inhibition of the binding between the MPM-2 antibody and MAP4 (Fig. 8A, P1). Similar results were observed when the P2 phosphopeptide-KLH conjugate was used in place of MAP4 (Fig. 9, A and
B). These results showed that the unconjugated 14 amino acid phosphopeptide could not only bind to the MPM-2 antibody, but that its affinity for MPM-2 antibody is comparable to that of a native MPM-2 reactive antigen. This suggested that the
P2 phosphopeptide, derived from human topoisomerase II, contains all the elements of the MPM-2 epitope necessary for MPM-2 binding.
Critical components of the MPM-2 phosphoepitope include aromatic amino acids located both to the N- and C-terminal of the phosphorylated residue.
Having shown that the P2 phosphopeptide derived from topoisomerase II was indeed sufficient to serve as an MPM-2 epitope, we further examined the characteristics of this epitope by changing selected amino acids to alanine (Table
1, peptides P3, P4, and P5). These residues were selected to examine the importance of the aromatic amino acids on either side of the phosphothreonine residue. Each of these phosphopeptides were also conjugated to KLH and tested on dot-blots for their MPM-2 reactivity as described above. We compared the staining of the KLH conjugates containing the modified peptides with the original
P2-KLH conjugate (Fig. 10A, P3). Changing from N at the +1 position to A did not generate any observable difference in the affinity of this peptide conjugate for the
63 MPM-2 antibody (Fig. 10). This result was not unexpected, since this modification did not affect either of the flanking aromatic residues. In addition, the minimal effect of this alteration further indicated that proline was not required at the +1 position in the MPM-2 epitope.
In contrast, changing the aromatic amino acid W to A at the -2 position greatly reduced this peptide’s ability to bind to MPM-2 antibody. Binding of MPM-2 antibody to the P4-KLH conjugate was observed only at extremely high substrate amounts, and also higher antibody concentrations (Fig. 10B, P4). The P4-KLH conjugate was not detected using lower antibody concentrations that still reacted with the P2-KLH conjugate (Fig. 10A, P4). Interestingly, changing the aromatic acid at the +2 position from F to A produced little difference in the ability of this peptide conjugate, P5-KLH, to react with the MPM-2 antibody (Fig. 10A, 10 and 2 ug).
Decreased staining was only observed when the amount of the P5-KLH conjugate present on the dot blot was greatly reduced (Fig. 10A, 0.4 ug).
These results showed that the aromatic amino acid on the N-terminal side of the phosphothreonine is critical for the binding of MPM-2 antibody. However, we had also proposed an important role in the binding of MPM-2 for the aromatic amino acid at the +2 position, but the dot-blots suggested that this amino acid has only a limited effect on the antibody-antigen binding.
64 In order to obtain a more precise indication as to the extent with which these amino acids might contribute to the MPM-2 epitope, we again examined each peptide using the competition experiments. As expected, preincubation with peptide P3 inhibited the ability of the MPM-2 antibody to bind to the P2-KLH conjugate (Fig. 11). In contrast, peptide P4 did not inhibit binding of the MPM-2 antibody, except at relatively high phosphopeptide concentrations (Fig. 11). Both the P3 and P4 competition experiments were consistent with the previous dot blot results (Fig. 10). The P5 phosphopeptide showed an intermediate ability to compete against the P2-KLH conjugate (Fig. 10), despite the ability of the P5-KLH conjugate to bind the MPM-2 antibody when blotted on nitrocellulose (Fig. 10). In part, the high local concentration of antigens on the dot blot may facilitate the binding of the antibody, while the intrinsic affinity of the antibody-antigen complex is the predominant factor influencing the competition experiments with the unconjugated peptides. Thus, the competition experiment provides a more accurate quantitative assessment of peptide-antibody affinity. When the peptide-KLH conjugates were tested on the nitrocellulose paper, the difference in the affinity could have been partially compensated for by the high local concentration of antigens, and thus little variation of MPM-2 staining between the P5-KLH and P2-
KLH was observed.
Most recently, we developed a protocol to retain peptides on nitrocellulose paper based upon a report by Eldik and Wolchok (Eldik and Wolchok, 1984). As
65 expected, the MPM-2 antibody recognized the phosphorylated topoisomerase II peptide P2 but not the dephosphorylated peptide P1 (Fig.12, P1 and P2). In addition, the MPM-2 antibody showed little difference in its affinity for P2 and P3.
However, unlike the results obtained using peptide-KLH conjugates, peptide P5 showed a greatly decreased affinity for the MPM-2 antibody (Fig. 12, -KLH, P2, P3, and P5). This result is consistent with the result obtained from the competition experiments and supports the notion that the aromatic amino acid at the +2 position is another important element in the MPM-2 epitope. I have also been able to separate the peptides from cellular proteins using a high percentage SDS-PAGE and transfer them to nitrocellulose paper for immunoblot using similar glutaraldehyde treatment. This approach was then used to test if the dephosphorylated topoisomerase II peptide could be phosphorylated by various kinase sources. Unfortunately, I have not been able to rephosphorylate this peptide to date (data not shown).
In addition to the phosphorylated threonine, our results suggested that for
MPM-2 recognition of the P2 phosphopeptide, the aromatic residue at position -2 was essential. The aromatic residue at the +2 position was also important, but contributes less to the antibody binding than that at the -2 position. A diagram that depicts our model of the MPM-2 epitope, which is based upon these results, is presented in Figure 13.
66 DISCUSSION
In this report, we have shown that a synthetic phosphopeptide composed of
14 amino acids has the capacity to serve as an epitope for the MPM-2 monoclonal antibody. This phosphopeptide can compete against both native and synthetic
MPM-2 antigens for the binding of the MPM-2 antibody. This peptide was derived from the sequence of topoisomerase II, which has been reported to be an MPM-2 reactive protein that is phosphorylated in a cell cycle dependent fashion {Taagepera et al., 1993a). We have demonstrated the importance of an aromatic amino acid on the N-terminal side of the phosphothreonine for the antibody-antigen interaction.
Our data on the +2 position indicates that an aromatic amino acid at the C-terminal side will also enhance the binding between the MPM-2 antibody and antigen, confirming results reported for the MPM-2 epitope of p42mapk (Taagepera et al.,
1994). In contrast to the previously reported requirement for a proline at the +1 position of the MPM-2 epitope (Westendorf et al., 1994), we have shown that proline is not required at this position for antibody recognition.
It should be noted that the proline directed MPM-2 epitope sites are associated with mitosis-specific phosphorylation, whereas the p42mapk site is not mitosis-specific. Identification of the proline-directed epitopes, however, required the prior phosphorylation of peptide substrates with kinases present in a mitotic extract (Westendorf et al., 1994). Thus, these MPM-2 reactive phosphopeptides
67 were preselected by their ability to be phosphorylated in vitro, and it is possible that other MPM-2 reactive substrates were not detected due to the conditions used to phosphorylate the peptide samples. Our results suggest that the +1 position has little or no influence on the recognition by MPM-2. Similarly, a proline was not present in the +1 position of the MPM-2 epitope reported on p42mapk (Taagepera et al., 1994). Therefore, mitosis-specific substrates for the MPM-2 antibody may contain a proline residue at the +1 position, but this is not necessarily a required element of the epitope.
Based upon the analysis of the MPM-2 reactive sequences reported here, and those previously described (Taagepera et al., 1994; Westendorf et al., 1994), we propose a model for the interaction between the MPM-2 antibody and the MPM-
2 epitope (Fig. 13). Most importantly, the epitope requires a phosphorylated amino acid, preferably phosphothreonine, although phosphoserine may be able to substitute. However, the phosphothreonine itself is not sufficient for the binding of antibody. An aromatic amino acid on the N-terminal side of the phosphothreonine appears to have a major positive influence on forming a stable complex between the
MPM-2 antibody and the antigen. The stability of the complex will be further enhanced if there is an aromatic amino acid or positively charged amino acid on the
C-terminal side of the phosphothreonine. Thus, an MPM-2 phosphoepitope will exhibit maximal antibody affinity when both the flanking N- and C-terminal elements are present. We speculate that native MPM-2 reactive proteins will contain these
68 epitope determinants. All of the factors that define the MPM-2 epitope have not been fully delineated, however, and we would expect that other elements having either positive and negative effects remain to be defined. For example, a proline residue located at the +1 position may enhance antibody binding.
Several protein kinases have been implicated in the phosphorylation of MPM-
2 epitope, including MEK (Taagepera et al., 1994), MAPK (Kuang and Ashorn,
1993), and cdc2 ( Westendorf et al., 1994). In addition, a novel and as yet unidentified kinase that phosphorylates the MPM-2 epitope site of topoisomerase
II corresponding to peptide P2 defined here may also exist. Although it is possible that a single MPM-2 kinase phosphorylates most MPM-2 sites in vivo, our data on the MPM-2 epitope, as well as the MPM-2 kinase characterization to date, indicates that the MPM-2 epitope may co-exist or partially overlap with a number of kinase consensus sequences. Each of these kinases may be responsible for the phosphorylation of one or more MPM-2 reactive proteins. It is likely that only some of the substrate proteins for these kinases contain all of the essential elements of the MPM-2 epitope, and only these selected substrates will become MPM-2 reactive following phosphorylation. This would explain how the cdc2 kinase might restore
MPM-2 reactivity on some proteins, whereas most cdc2 sites would not be recognized by the MPM-2 antibody.
69 Several MPM-2 reactive proteins have been shown to be present during
interphase, and they only become MPM-2 reactive during M-phase (Vandre et al.,
1991; Taagepera et al., 1993a; Tombes et al., 1991). Thus, the phosphorylation of the MPM-2 site may serve to regulate the functions of the various MPM-2 reactive
proteins in a cell cycle dependent fashion (Vandre and Borisy, 1985). To date,
however, the functional significance of the MPM-2 epitope site has only been
established in p42mapk (Taagepera et al., 1994). The phosphorylated threonine
residue required for MPM-2 reactivity located in the TEY site is involved in activating the kinase activity of p42mapk (Anderson et al., 1990; Payne et al., 1991). The
activity of topoisomerase II is regulated in a cell cycle dependent fashion and this activity is thought to be regulated by phosphorylation (Cardenas et al., 1992;
Vassetzky et al., 1994; Dang et al., 1994; Poljak and Kas, 1995; Watt and Hickson,
1994). It is known that mammalian cells require the active topoisomerase II to pass a G2 checkpoint (Downes et al., 1994), and a recent report has identified several sites on HeLa topoisomerase I la proteins that are phosphorylated predominantly or exclusively during the G2 and M phases (Wells and Hickson., 1995). It is possible that the MPM-2 site on topoisomerase II may also be important in
regulating topoisomerase II activity during mitosis, however, this remains to be determined.
Thus, the MPM-2 reactive topoisomerase II peptide sequence we have
identified here could be used as a unique substrate to identify/isolate a novel in vivo
70 topoisomerase II kinase, which may be important in the regulation of topoisomerase activity. Coupled with our knowledge of the MPM-2 epitope, this approach could be applied in general to purify or identify the corresponding in vivo kinases of other
MPM-2 reactive proteins. This approach would clearly establish whether there is a single in vivo MPM-2 epitope kinase, or whether there are multiple kinases involved in establishing MPM-2 reactivity of different proteins during M-phase.
Preliminary evidence indicates that antibodies generated against the phosphorylated topoisomerase II peptide P2 used in these studies specifically recognize a number of proteins of mitotic cell lysates by immunoblot analysis.
These proteins are not detected in similar interphase cell lysates. These results suggest that the topoisomerase II phosphorylation site reported here is modified in vivo in a cell cycle dependent fashion, and may represent a common phosphoepitope present in selected mitotic proteins. It appears likely that a single
MPM-2 epitope kinase does not exist, but that the MPM-2 epitope will prove important for the identification of a number of unique mitosis-specific phosphoproteins and their respective kinases.
71 Peptide Constructs
PI*: [NH2] -RKEWLT NFMEDRRC- [CONH2] P2 : [ NH2 ] - RKEWLTPNFMEDRRC - [ CONHz ] P3 : [NH2 ] - RKEWLTpAFMEDRRC - [ CONH2 ] P4 : [ NH2 ] - RKEALTpNFMEDRRC - [ CONH2 ] P5 : [NH2 ] -RKEWLTPNAMEDRRC- [CONH2]
* The first 13 amino acids in PI correspond to aa660--aa672 in human DNA topoisomerase Ila and aa682~-aa694 in human DNA topoisomerase 11(3,
TABLE 1
72 Fig.7. Recognition of the P2-KLH conjugate by the MPM-2 antibody.
A. Dot-b)ots of mitotic MAP4 and the P2-KLH conjugate at different MPM-2 antibody concentrations {1: 1:1000000; 2: 1:100000; 3: 1:20000; 4:10000; 5:
1:1000). Mitotic MAP4 (1.5 ug) and P2-KLH conjugate (1.25 ug) were applied to each dot. B. Corresponding plot of the amount of bound MPM-2 antibody to mitotic
MAP4 (Integrated Optical Density) against the antibody concentrations used.
Based upon the data from panel A. C. Corresponding plot of the amount of bound
MPM-2 antibody to the P2-KLH conjugate (Integrated Optical Density) against the antibody concentrations used. Based upon the data from panel A.
73 A 1 2 3 4 5
MAP4 $ # • #
P2-KLH 9 9 9 0
B
2 >- cyi
1
0 6 5 4 3
Log MPM2 Dilution c
4 >* ’K 3
2
1
0 6 5 4
Log MPM2 Dilution
I'IGURH 7
74 Fig.8. Competition of synthetic peptides for MPM-2 antibody binding to mitotic
MAP4.
A. Peptides P1 or P2 were used to compete for binding of MPM-2 antibody against mitotic MAP4 blotted onto nitrocellulose paper. Mitotic MAP4 (1.5 ug) was applied to each dot. Different amounts (ug) of the P1 or P2 synthetic peptides were added to 3 ml of MPM-2 antibody solution diluted to 1:20000 and preincubated for
1 h prior to application to the dot-blot. B. The amount of bound MPM-2 antibody was determined using image analysis software and plotted against the amounts of synthetic peptides P1 {□) or P2 (■) preincubated with the MPM-2 antibody.
75 Integrated Optical Density etd (ug) Peptide <] & Zc Fig.9. Competition of synthetic peptides for MPM-2 antibody binding to the P2-
KLH conjugate.
A. Peptides P1 or P2 were used to compete for binding of MPM-2 antibody against P2-KLH conjugate blotted onto nitrocellulose paper. P2-KLH conjugate
(1.25 ug) was applied to each dot. Different amounts (ug) of the P1 or P2 synthetic peptides were added to 3 ml of MPM-2 antibody solution diluted to 1:20000 and preincubated for 1 h prior to application to the dot-blot. B. The amount of bound
MPM-2 antibody was determined using image analysis software and plotted against the amounts of synthetic peptides Pt (□) or P2 (■) preincubated with the MPM-2 antibody.
77 Integtated Optical Density co o O O NJ & 03 "0 a ro • • atn
etd (ug) Peptide • • '■-J 00 oa VC • •
to o o ro o Fig. 10. MPM-2 antibody binding to modified synthetic peptide-KLH
conjugates.
Different amounts (10, 2, and 0.4 ug) of KLH (KLH) or peptide-KLH conjugates (P I, P2, P3, P4, and P5) were applied to the nitrocellulose paper and tested for their MPM-2 binding ability at different antibody dilutions (A: 1:10000; B:
1:1000). Antibody binding was compared between the control topoisomerase
peptide P1 and P2, and these modified phosphopeptides P3, P4, and P5 (see Table
1).
79 FIGURH 10
80 Fig.11. Competition of modified phosphopeptides for MPM-2 binding to the
P2-KLH conjugate.
Synthetic phosphopeptides P3, P4 or P5 were used to compete for binding of the MPM-2 antibody against the P2-KLH conjugate blotted onto nitrocellulose paper. The bound MPM-2 antibody was visualized, and the density of the reaction product determined by image analysis, and plotted against the amounts of the P3
{■), P4 (□) or P5 (a ) synthetic peptides preincubated with the MPM-2 antibody. P2-
KLH conjugate (1.25 ug) was applied per dot. Different amounts of each synthetic peptide was added to 3 ml of the MPM-2 antibody diluted to 1:20000, and preincubated for 1 h prior to application to the dot-blot.
81 3 £ em am 0 2 u 1 TJ 9 1 □
Ic 0 0 50 100 150 200 Peptide (ug)
■’I GURU 11
82 Fig. 12. Binding of the MPM-2 antibody to synthetic peptide constructs.
4 ug of either peptides (-KLH: P1, P2, P3, P4, P5) or peptide-KLH conjugates
(+KLH: P1, P2, P3, P4, P5) was applied to each well of the dot-blot apparatus. The nitrocellulose was then specifically treated with glutaraldehyde to retain peptides as described in Materials and Methods. The MPM-2 antibody used here was diluted
1:1000.
83 P1 P2 P3
— KLH # •
+ KLH • #
FTCURF. 12
84 Fig. 13. Model of the MPM-2 phosphoepitope.
The MPM-2 epitope is proposed to consist of three elements, each of which contributes to maximal affinity for the antibody as indicated by the arrows. Wider arrows indicate a larger contribution for the binding. The elements include the phosphorylated amino acid, flanking aromatic residue to the N-terminal side, and flanking aromatic or positively charged amino acid to the C-terminal side as described in the text.
85 MPM-2 Antibody I NH,J? MPM-2 Epitope L COOh
FTtlURF; 13
86 CHAPTER 3
INTRODUCTION
DNA undergoes a variety of metabolic transformations during a single cell cycle. DNA must be replicated during the S phase, chromatin needs to be organized and condensed during prophase, sister chromatids need to be separated during metaphase, and chromosomes need to be decondensed during telophase. AH of these events impose topological dilemmas upon the extremely extended and tangled linear DNA molecules. In part, this formidable task is accomplished due to the activity of a group of enzymes termed DNA topoisomerases. DNA topoisomerases can be divided into two types according to their functions.
Topoisomerase I can pass a single strand of DNA through another single strand by temporarily opening up the latter. Topoisomerase II functions similarly to topoisomerase I, except it passes a double-stranded DNA through another double strand DNA (Roca, 1995).
Human cells express two types of DNA topoisomerase II enzymes, namely, topoisomerase lla and topoisomerase lip. These two enzymes are produced by two
87 separate genes (Drake et al., 1987, 1989; Chung et al., 1989; Austin and Fisher,
1990; Jenkins et al., 1992; Austin et al., 1993) differentially during cell cycle transit
and following oncogenic transformation of cells (Woessner et al., 1990,1991). Each
enzyme is composed of three discrete domains. The N-terminal domain contains the ATP-binding site and has ATPase activity (Ali et al., 1993). The central domain
binds DNA and is responsible for DNA breakage and reunion. The C-terminal
domain is less conserved among the topoisomerase II enzymes and is not directly
invovled in forming the catalytic structure, instead, this domain is believed to play
potential roles in nuclear localization, dimerization, and regulation of enzymatic
activity (Watt and Hickson, 1994).
Topoisomerase II is a nuclear enzyme and is shown to be a major structural component of the mitotic chromosome scaffolds (Eamshaw et al., 1985). Antibodies against topoisomerase II show specific axial staining in both human and the Indian
muntjac chromosomes (Eamshaw and Mackay, 1994; Saitoh and Laemmli, 1994).
However, immunolocalization of topoisomerase II using Xenopus extract (Hirano and Mitchison, 1993) and live Drosophila embryos (Swedlow et al., 1993) reveals that the enzyme is distributed uniformly throughout chromosomes. The varied observations are possibly due to the existance of multiple topoisomerase II forms, varying at either the primary amino acid sequence (a or (3) or the phosphorylation state. Each of these forms may distribute differently and possibly serving different
roles. Swedlow et al. (1993) has reported that three distinct topoisomerase II
88 populations exist in Drosophila embryos. It has also been shown that the topoisomerase lla, but not topoisomerase lip, is a component of the chromosome scaffold (Zini et al., 1992).
Topoisomerase II is essential for cell viability. Yeast strains containing temperature sensitive topoisomerase II mutations fail to grow at the non-permissive temperature as they could not separate the replicated sister chromatids, which remained extensively knotted and tangled (DiNardo et al., 1984; Holm et al., 1985;
Uemura and Yanagida, 1986). Correlating with its importance during mitosis, topoisomerase II activity is regulated in a cell cycle dependent fashion and its activity is greatly enhanced during mitosis (Wells and Hickson, 1995). It is known that this enhancement is largely due to selective phosphorylation of certain topoisomerase II sites (DeVore et al., 1992; Corbett et al., 1992, 1993a, 1993b).
Interestingly, topoisomerase becomes MPM-2 reactive after the cell enters mitosis, indicating it contains an MPM-2 site that is phosphorylated during mitosis
(Taagepera et al., 1993). Since the MPM-2 epitope itself is believed to be an important mitotic regulatory site shared by many proteins phosphorylated during mitosis (Davis et al., 1983; Vandre et al., 1986), it is proposed that this phosphorylation event might contribute to the elevation of topoisomerase II activity during mitosis.
89 In Chapter 2, we described efforts that have been made to characterize the essential elements of the MPM-2 epitope. It was shown that the phosphorylated peptide derived from the topoisomerase II sequence is MPM-2 reactive on dot-blots and is capable of competing for MPM-2 binding against native MPM-2 antigens.
However, it remained to be shown that this was not strictly an in vitro phenomenon related solely to the synthetic phosphopeptides.
In order to prove that this topoisomerase II sequence is indeed the in vivo
MPM-2 site present on topoisomerase II, we needed to show that this site was phosphorylated in vivo. Therefore, we have generated rabbit polyclonal antibodies against either the dephosphorylated or phosphorylated DNA topoisomerase II peptides. The immune serum against the phosphopeptide (Ab2) was further affinity purified to obtain the IgG that is specific for the phosphopeptide.The assertion that this antibody recognizes a phosphoepitope was confirmed using three different lines of evidence. First, this antibody recognized the phosphorylated topoisomerase II peptide transferred onto nitrocellulose, but did not recognize the dephosphorylated form of the peptide; secondly, competition experiments showed that only the phosphorylated peptide was able to inhibit antibody binding to the native antigens present in mitotic HeLa cell lysates; and thirdly, limited treatment with alkaline phosphatase was able to eliminate a large portion of the PTE1 reactive species in mitotic HeLa cell lysate. We have also demonstrated that the affinity purified antibody, PTE1, recognized a band comigrating with topoisomerase II on 4% UREA
90 gels. Taken together, these results indicate that topoisomerase II is likely to be phosphorylated on this particular site in vivo, and that this site is responsible for the affinity of the MPM-2 antibody.
In addition to DNA topoisomerase II, PTE1 also recognized a large number of proteins in the mitotic HeLa cell lysate. When compared to the MPM-2 reactive proteins present in the same cell lysate, it was observed that PTE1 and MPM-2 each recognized a distinct subset of proteins. However, the majority of the immunoreactive proteins seemed to be both MPM-2 and PTE1 reactive. Testing of various MAPs showed that PTE1 recognized rat brain M API, which is also MPM-2 reactive. Indirect immunofluorescence localization in JAR cells showed that PTE1 antigens were present on several important components of the mitotic apparatus, including the chromosomes, spindle poles, and midbody. The appearance of PTE1 antigens was highly regulated, both spacially and temporally. Similar to the immunoblot results, PTE1 showed similar but not identical cellular staining when compared to MPM-2. The results presented here not only verified the in vivo phosphorylation of the tentative MPM-2 sequence on topoisomerase II, it also established PTE1 as a unique marker for certain mitotic phosphoproteins. PTE1 should provide researchers with a new unique probe for studying mitotic regulation and phosphorylation.
91 MATERIALS AND METHODS
Materials
Peptides P1 and P2 were synthesized by AnaSpec Incorporated to greater than 95% purity by HPLC. MPM-2 mouse monoclonal antibody was a generous gift from Dr. Potu Rao {Department of Chemotherapy Research, The University of
Texas M.D. Anderson Hospital and Tumor Institute, Houston, TX), or purchased from Upstate Biotechnology Incorporated. Mouse monoclonal Ab-1 against the C- terminal domain of topoisomerase lla was obtained from Oncogene Science, Inc.
Peroxidase-conjugated goat anti-mouse antibody was purchased from Kirkegaard
& Perry Laboratories. Peroxidase-conjugated goat anti-rabbit antibody was from
American Qualex. FITC conjugated goat anti-mouse and goat anti-rabbit antibodies were from Cappell and Durham. Keyhole limpet hemocyanin was obtained from
CalBiochem, and m-maleimidobenzoyl-A/-hydroxysuccinimide ester from Pierce.
HM-2 mouse monoclonal antibody, nocodazole, and hydroxyurea were from Sigma
Chemical Company; Calf intestinal alkaline phosphatase was from Boehringer
Mannheim Biochemicals. SeeBlue™ molecular weight marker was from Novel
Experimental Technology. Prepared PD-10 Sephadex columns, HiTrap™ affinity columns, and CNBr-activated Sepharose 4B were from Pharmacia Biotech. Rat brain MT was purified as described in the Materials and Methods section in Chapter
1.
92 Cells and Cell synchronization
HeLa human epitheloid carcinoma cells were maintained in DMEM supplemented with 10% fetal bovine serum and antibiotics (100 units/ml penicillin,
0.1 mg/ml streptomycin). Nocodazole was added to the culture medium to a final concentration of 0.04 ug/ml to arrest cells in mitosis and the synchronized mitotic cells were usually harvested 16-24 hr later. For the preparation of interphase HeLa cells, a final concentration of 2.5 mM hydroxyurea was used in place of nocodazole.
Rabbit polyclonal antibody generation
Peptides were conjugated to KLH and purified as described in the Materials and Methods section of Chapter 2. 1 mg of conjugated peptide was dissolved in 1 ml PBS and mixed thoroughly with 1 ml of complete Freund’s Adjuvant. The rabbits were anesthetized using ketamine and xylazine (45 mg and 9 mg /kg, respectively) and the antigen/adjuvant mixture was injected subcutaneously at multiple sites along the rabbits abdominal wall. The rabbits were boosted 20 days later with 0.5 mg of the corresponding conjugated peptide mixed with an equal volume of incomplete Freund’s Adjuvant. The rabbits were bled 11 days after the boost. The blood was allowed to clot at room temperature and then sit overnight at 4°C. The immune serum was recovered from the sample containing the retracted clot.
93 Affinity Purification of Polyclonal Antibody
Purified IgG was obtained by running the rabbit immune serum through the
HiTrap™ Protein G affinity columns according to the manufacturer’s protocol. The
rabbit serum was applied to the column pre-equilibrated with three column volumes
of start buffer (20 mM Na-phosphate, pH 7.0), followed by a continuous wash in the start buffer until no material appeared in the effluent. The elution buffer (0.1 M
glycine-HCI, pH 2.7) was then applied to the column to elute the IgG.
5 mg of either the dephosphorylated or phosphorylated topoisomerase II
peptides was dissolved in the coupling buffer (0.1 M NaHC03, 0.5 M NaCI, pH 8.3)
and mixed with 1 ml CNBr-activated Sepharose 4B gel. The mixture was gently
stirred at 4°C overnight. The excess peptide was washed away using a large volume of coupling buffer. The remaining active groups were then blocked by
incubating the gel with the blocking buffer (0.1 M Tris-HCI, pH 8.0) at RT for 2 hr.
Finally, the beads were washed with three cycles in alternative Buffer A (0.1 M
acetate, 0.5 M NaCI, pH 4.0) and Buffer B (0.1 M Tris-HCI, 0.5 M NaCI, pH 8.0). The
beads were then packed into a 3 ml syringe and equilibrated in the start buffer for future use. The purified IgG from the HiTrap™ affinity column was first passed
through the affinity column coupled with the dephosphorylated peptide. The flow
through was then applied to the affinity column coupled with the phosphorylated
94 peptide. The bound IgG was released from this column by the same elution buffer employed for HiTrap™ Protein G column and used in the subsequent experiments as the phosphopeptide-specific IgG.
Dephosphorylation by alkaline phosphatase
Mitotic HeLa cell lysates were incubated with various amounts of calf intestinal alkaline phosphatase in the dephosphorylation buffer (0.05M Tris-HCI, 0.1 mM EDTA, pH 8.5) in the presence of a protease inhibitor cocktail (Chapter 1,
Materials and Methods). The reactions were stopped by the addition of SDS sample buffer.
Polyacrylamide gel electrophoresis, western blot, and competition experiment
Standard SDS-PAGE, UREA gel, and western blots were performed as described in Chapter 1 Materials and Methods, except where SDS-PAGE and western blots were used to test peptide samples. In these experiments, the nitrocellulose paper was incubated in 0.2% glutaraldehyde in PBS (140 mM NaCI,
1.5 mM KH2PO<, 2.7 mM KCI, 6.5 mM Na HPO 4 pH 7.4) for 45 min immediately after transfer, and the unreacted aldehyde groups were then reduced by treatment with two changes of NaBH4(1 mg/ml in TBS). The nitrocellulose paper was subsequently
95 rinsed in TBS (0.9% NaCI, 10 mM Tris-base, pH 7.4) and processed for immunological detection. In the competition experiments, the primary antibody
PTE1 was preincubated with various amounts of synthetic peptides for 1 h prior to incubation with the nitrocellulose paper.
Indirect Immunofluorescent Staining
JAR cells were subcultured onto glass coverslips for indirect immunofluorescent staining with the antibodies. After a brief rinse (30 sec) in PBSa
(140 mM NaCI, 1.5 mM KH2P041 2.7 mM KCI, 6.5 mM Na2HP04, 0.002% NaN3, pH
7.4), JAR cells were fixed/lysed with 0.7% glutaraldehyde and 0.5% Triton X-100 in PBSa for 15 min at 22°C. The unreacted aldehyde groups were reduced by treatment with 2 changes of NaBH4(1-2 mg/ml in TBS) over 20-30 minutes. The cells were washed again with three changes of PBSa and processed for immunofluorescence staining as described in the Materials and Methods section of the Appendix.
96 RESULTS
The rabbit polyclonal antibody generated against DNA topoisomerase II phosphopeptide recognizes a large number of mitotic proteins including topoisomerase II.
It was shown in Chapter 2 that the phosphorylated DNA topoisomerase II peptide was MPM-2 reactive. However, the question remained as to whether this site was phosphorylated in vivo. To address this, both the phosphorylated and dephosphorylated DNA topoisomerase II peptides were used to immunize rabbits in hope of generating rabbit polyclonal antibodies specific for either the dephosphorylated or the phosphorylated form of the topoisomerase II sequence.
The immune serum (Ab1) generated against the dephosphorylated peptide contained limited immunological activity, and was similar to its preimmune serum when probed against either interphase or mitotic HeLa cell lysates (Fig.14, lanes 4,
5,12,13). On the other hand, while the preimmune serum of the antibody generated against the phosphorylated peptide showed no immunological affinity for either interphase or mitotic HeLa lysates (Fig. 14, lanes 6 and 14), the immune serum
(Ab2) selectively recognized a large number of bands in the mitotic HeLa cell lysate
(Fig.14, compare lanes 7 and 15). As shown in Fig.14, lanes 8 and 16, the protein amounts in the interphase and mitotic HeLa cell lysates were comparable.
97 To remove any nonspecific immunological reactivity present in Ab2, the immune serum was subjected to three different affinity purification steps. The immune serum was first passed through the HiTrap™ Protein G affinity column to obtain a purified IgG fraction. The IgG was then passed through an affinity column constructed using the dephosphorylated peptide coupled to CNBr-activated
Sepharose 4B beads. The flowthrough from this purification step presumably contained two types of IgG. The first type were those IgGs that had no affinity for the topoisomerase peptide, including the endogenous IgG and the IgG generated against the carrier protein KLH. The second type included those IgGs that specifically recognized the phosphorylated form of the topoisomerase II sequence.
This group of antibodies would require the presence of the phosphate group for peptide binding, thus they would not bind to the dephosphopeptide affinity column.
This flowthrough fraction was then subjected to a third affinity column containing the phosphorylated topoisomerase II peptide as its ligand. The bound antibody was eluted and contained only the IgG fraction that was specific for the phosphorylated topoisomerase II sequence. This affinity purified IgG fraction was designated PTE1 for Phosphorylated Topoisomerase II Epitope 1 and was used for most of the further analysis of Ab2. PTE1 IgG refers to the HiTrap™ Protein G affinity purified IgG fraction of Ab2.
As shown in Fig.14, lanes 2 and 10, PTE1 specifically recognized a group of mitotic proteins in the mitotic HeLa cell lysate. There is a minor difference in the
98 staining pattern between the PTE1 and the original immune serum, due to the
elimination of non-specific IgG from Ab2. Interestingly, most of the PTE1 reactive
proteins in the mitotic HeLa cell lysate overlapped with MPM-2 reactive proteins on
immunoblot (Fig.14, compare lanes 10 and 11). On the other hand, both PTE1 and
MPM-2 each recognized an unique subset of proteins that were not detected by the
other antibody (Table 2).
To test if DNA topoisomerase II is indeed phosphorylated in vivo at this
particular site, the phosphoepitope specific antibody, PTE1, was used to blot a
mitotic HeLa cell lysate separated on a 4% UREA gel. As expected, PTE1
recognized a band comigrating with topoisomerase II (Fig.15, lanes 2 and 4). This
band was also recognized by MPM-2 (Fig. 15, lane 3). Since PTE1 recognizes the
phosphopeptide derived from the topoisomerase II sequence, and since the
possibility of overlapping bands on the 4% UREA gel is very small, it was concluded that this comigrating band recognized by PTE1 was topoisomerase II. Further,
PTE1 only recognizes the topoisomerase II sequence in its phosphorylated form,
thus it is highly likely that topoisomerase II is phosphorylated at this site in vivo,
which also contributes to its observed MPM-2 reactivity.
99 The rabbit polyclonal antibody generated against the DNA topoisomerase II phosphopeptide is phosphoepitope specific.
To confirm that the immune serum generated against the phosphorylated topoisomerase peptide indeed only recognizes the phosphoepitope, peptides were separated by SDS-PAGE and transferred to nitrocellulose for immunoblotting. As expected, Ab1 recognized both the dephosphorylated and phosphorylated forms of the peptides {Fig.16, lanes 1 and 2). On the other hand, Ab2 only recognized the phosphorylated peptide (Fig. 16, lane 4). This result confirmed that Ab2 required a phosphorylated residue within its epitope.
In an effort to further substantiate the conclusion that this antibody recognizes a phosphoepitope, both peptide competition experiments and antigen dephosphorylation experiments were carried out. In the competition experiments,
PTE1 IgG was preincubated with various amounts of either the dephosphorylated peptide or the phosphorylated peptide. The antibody/peptide mixtures were then used to probe mitotic HeLa lysates by immunoblot. As shown in Fig. 17, preincubation of PTE1 IgG with amounts of the dephosphorylated peptide up to 100 ug showed no observable difference in affinity for its antigens {Fig. 17, lanes 2 and
3) as compared to the control {Fig.17, lane 1). However, the presence of 10 ug of phosphopeptide in a solution containing 60 ug of the PTE1 IgG almost completely eliminated its ability to bind to the antigens. This result strongly indicated that the
100 specificity of the antibody required a phosphorylated epitope. In addition, incubation with the phosphopeptide, but not the dephosphorylated peptide, inhibited antibody binding to the endogenous topoisomerase II band (Fig.17, arrow), further indicating that topoisomerase II is phosphorylated in vivo on this site. Finally, the limited dephosphorylation of the mitotic HeLa cell lysate by calf intestinal alkaline phosphatase was also able to eliminate the binding of PTE1 to a large number of the mitotic proteins as compared to the control (Fig. 18, lanes 3 and 4). Taken together, we conclude that PTE1 indeed recognizes a phosphoepitope that is shared by many mitotic proteins.
The PTE1 epitope is also present on rat brain M API.
Since PTE1 recognizes some proteins that are also MPM-2 reactive, it was of interest to see if PTE1 also recognized MPM-2 reactive MAPs, namely MAPI and
MAP4. Rat brain MT was employed to test this possibility. The major proteins present in the rat brain MT are tubulins, MAPI, and MAP2. MPM-2 antibody was used to show the position of MAPI (Fig. 19, lane 2), and HM-2 mouse monoclonal antibody was used to show the position of MAP2 (Fig. 19, lane 3). On both 4%
UREA gel and 7% SDS-PAGE, PTE1 was shown to specifically recognize the
MAPI but not the MAP2 (Fig.19, lane 1 and data not shown).
101 PTE1 antigens are localized to the centrosomes and chromosomes of mitotic cells.
Since PTE1 recognizes a group of proteins that are presumably phosphorylated during mitosis, it is likely that these proteins play important roles in the process of cell division. This is clearly the case for topoisomerase II. Therefore it was of interest to localize the PTE1 reactive proteins in cells. In particular, it was important to determine if they were associated with any components of the mitotic apparatus. The human JAR cell line was chosen for the indirect immunofluorescence localization of the PTE1 antigens, because they remain flat during mitosis, making mitotic structures easy to observe. As shown in Fig.20, the preimmune serum of Ab2 showed a weak generalized background staining of both interphase and mitotic JAR cells (Fig.20, panels B and D). Ab2, on the other hand, brightly stained the spindle poles and the chromosomes of mitotic cells (Fig.20, panel H). Ab2 also showed some generalized cytoplasmic staining in a few interphase cells, but showed minimal staining in most other interphase cells (Fig.20, panel F). Affinity purified PTE1 showed much more specific staining pattern in both interphase and mitotic JAR cells (Fig.20, panels J and L).
As expected from the immunoblot results, immunofluorescent staining of both the PTE1 and MPM-2 antibodies showed some similarities. However, obvious differences were also observed. The MPM-2 antibody stained some patches within
102 the interphase nuclei, while the PTE1 showed no discrete localization within the interphase nuclei. In the mitotic cells, both MPM-2 and PTE1 recognized the chromosomes and spindle poles, however, MPM-2 also recognized some fibrous structures that radiated from the spindle poles to the chromosomes that were not detected by the PTE1 (Fig.20, panel P).
The PTE1 antibody was tested in more detail to examine the localization of its antigens through out the cell cycle. It appeared that chromosomes and centrosomes become PTE1 reactive prior to the nuclear envelope breakdown (data not presented), and by prometaphase they could readily be stained by the PTE1 antibody (Fig.21, panel A). Intense staining by PTE1 of centrosomes, later to become the spindle poles, and the chromosomes could be observed throughout metaphase (Fig.21, panel B) into early anaphase (Fig.21, panel C). By late anaphase (Fig.21, panel D) and early telophase (Fig.21, panel E), the staining of chromosomes and spindle poles decreased in comparison to metaphase cells.
Staining was eventually lost sometime during telophase. However, PTE1 antigens could be observed during cytokinesis at the midbody (Fig.21, panel F). The highly regulated presence of PTE1 antigens on these important components of mitotic apparatus throughout mitosis was highly suggestive that these PTE1 -reactive proteins are important mitotic regulators. They are specifically phosphorylated and dephosphorylated during different stages of mitosis, possibly becoming functionally
103 activated or inactivated, and are likely involved in the regulation of the structures with which they are localized.
104 DISCUSSION
In this Chapter, it was shown that a rabbit polyclonal antibody generated against a topoisomerase II phosphopeptide recognized a protein in the mitotic HeLa cell lysate that comigrated with topoisomerase II on immunoblots. It was also shown that this antibody, PTE1, recognizes a phosphoepitope by both the peptide competition experiments and the dephosphorylation experiment. The obligatory presence of a phosphate within the PTE1 epitope was confirmed by the observation that PTE1 failed to recognize the dephosphorylated form of the topoisomerase II peptide on immunoblot. These results lead to the conclusion that DNA topoisomerase II is indeed phosphorylated in vivo at this site, which contributes to the affinity of topoisomerase II for the MPM-2 antibody. In addition to topoisomerase
II, PTE1 also recognized a group of proteins present in the mitotic HeLa cell lysate.
The pattern of PTE1 reactive proteins on the blot is similar to, but not identical to, that of the MPM-2 reactive proteins. As shown in the indirect immunofluorescent study, PTE1 stained some important components of the mitotic apparatus, including chromosomes, spindle poles, and the midbody. This result is again similar to MPM-2 staining. However, MPM-2, but not PTE1, also stains some fibrous structures in the mitotic spindle and some foci within the interphase nuclei. The data presented here defines PTE1 as a unique marker for selected mitotic phosphoproteins, and should aid in the future study of mitotic regulation and phosphorylation.
105 As mentioned above, in addition to recognition of a common subset of mitotic proteins, both the MPM-2 and the PTE1 antibodies also recognize a distinct subset of bands on the nitrocellulose that are unique to each antibody. The most probable explanation is that the PTE1 epitope partially overlaps with the MPM-2 epitope and that neither of the epitopes are identical, with each having a selective epitope requirement. Thus, any proteins containing the amino acids required in both epitopes will be recognized by both antibodies. On the other hand, proteins containing the essential amino acids for only one epitope will only be recognized by the corresponding antibody. In addition, there is the possibility that the polyclonal antibody contains some antibodies whose epitopes, besides the phosphothreonine, do not overlap with the MPM-2 epitope. These other antibodies could contribute to the additional bands that were observed selectively in the blot probed by PTE1.
A topoisomerase II amino acid sequence specific antibody was used to mark the position of topoisomerase II, and it was observed that a PTE1-reactive band always comigrated with topoisomerase II on the various SDS-PAGE gels used. To reduce the chance that this was an unrelated protein, mitotic HeLa cell lysates were separated on a 4% UREA gel and nitrocellulose transfers were probed with the
PTE1, MPM-2, and topoisomerase II specific antibodies. It was obvious that both
MPM-2 and PTE1 recognized a band that ran at the same exact position on the gel.
This band was equivalent to the top portion of the band recognized by the topoisomerase II antibody. It is believed that this top portion of the topoisomerase
106 II band is in its hyperphosphorylated form. There remains a small possibility that this
PTE1 reactive band is unrelated to topoisomerase II, and that the molecular weights of these proteins are just strikingly similar. One way to ascertain the identity of this
PTE1 reactive band is to immunoprecipitate the mitotic HeLa cell iysate with topoisomerase II antibody and see if this band could still be recognized by the PTE1 antibody. Another possible but unlikely scenario, is that PTE1 recognized a different phosphorylation site on topoisomerase II. This would require that the sequence used to generate the antibody was not phosphorylated in vivo. To eliminate this remote possibility, it will be necessary to carry out limited proteolysis of topoisomerase II and sequence the amino acids of the smallest PTE1 reactive band.
In the competition experiments, preincubation with the phosphorylated peptide prevented the purified PTE1 IgG from recognizing nearly all of its antigens.
There were, however, still a few bands that remained immunoreactive. This was probably due to the presence of non-specific antibodies in the PTE1 IgG preparation used. The ideal experiment would be carried out using the affinity purified antibody PTE1 instead of the IgG. The amounts of PTE1 currently available were not sufficient for these competition experiments. Alkaline phosphatase treatment of the mitotic HeLa cell lysate also successfully abolished the affinity of many protein species for PTE1. Prolonged treatment at 37°C resulted in a fewer
107 number of PTE1 reactive proteins, but this treatment may also have resulted in some protein degradation as shown by coomassie staining (data not shown).
The crystal structure of the yeast DNA topoisomerase II central domain, responsible for DNA binding and DNA breakage/reunion, was recently reported
(Berger et al., 1996). According to the report, this domain has a shape of a flattened crescent, and forms a heart-shaped structure with a large central hole when dimerized. The polypeptide chain of this domain folds into two distinct subfragments linked together by a relaxed linker sequence of 48 amino acids. Each subfragment forms a stable three-dimensional structure, but the relative position of the two subfragments could be changed by bending the linker sequence during various catalytic stages of the topoisomerase II. This conformational change is believed to be essential for both DNA binding and DNA breakage/reunion. Interestingly, the active-site tyrosine residue, responsible for breaking DNA molecules and forming a covalent bond with the new 5' end, is localized close to the linker sequence in the crystal structure. After comparing the yeast and human topoisomerase II sequences, we find that the MPM-2 site identified here falls within the linker sequence between the two subfragments of topoisomerase II central domain. This finding provides structural evidence for the potential regulatory roles of MPM-2 phosphorylation. Addition of a phosphate within the linker sequence could significantly modify the elasticity of the sequence. Alternatively, addition of a phosphate could affect the activity of the neighboring tyrosine residue involved in
DNA breakage/reunion. Since the linker sequence is located in the vicinity where
108 DNA fragments binds or passes, a negatively charged phosphate would be likely to affect these processes.
Many kinases have been reported that are capable of phosphorylating topoisomerase II in vitro, and these are potential candidates for the in vivo kinases responsible for the phosphorylation of topoisomerase II. For example, the specific activity of the topoisomerase II from Drosophila cells is stimulated 2-3 fold after phosphorylation in vitro by casein kinase II or protein kinase C (DeVore et al., 1992;
Corbett et al., 1992, 1993a, 1993b). It also has been shown that casein kinase II could reactivate topoisomerase II that had previously been inactivated by treatment with alkaline phosphatase (Cardenas and Gasser, 1993; Cardenas et al., 1993).
In addition, many of the in vivo phosphorylation sites can be phosphorylated in vitro by either p34cdc2 or mitogen-activated protein (MAP) kinase (Kuang and Ashorn,
1993; Westendorf et al., 1994). It is known that the phosphorylation of topoisomerase II is maximal during the G2/M phases of the cell cycle in many cell lines examined (Heck et al., 1989; Saijo et al., 1992; Burden et al., 1993). Results obtained from two dimensional tryptic phosphopeptide mapping further indicated that several sites on HeLa topoisomerase II were phosphorylated predominantly or exclusively during the G2 and M phase (Wells and Hickson, 1995). It is likely one of these sites is the MPM-2 site shown here.
Previous reports have indicated that both MAPK and p34cdc2 are capable of restoring MPM-2 reactivity on some proteins. However, the topoisomerase II MPM-2
109 sequence shown here does not fit the consensus motifs of either of these kinases.
This MPM-2 sequence also does not fit the consensus motif for protein kinase C, which requires a basic amino acid context. On the other hand, it is very close to a typical casein kinase II site which requires an acidic amino acid at the +3 position and does not contain any basic amino acids at the -1/+1 position nor proline at the
-1 position (Pinna, 1990). In the sequence identified here, the threonine residue is surrounded by two glutamic acid residues at the -3 and +4 positions and contains neither basic amino acids nor proline between these glutamic acid residues. This leads to the possibility that this site is phosphorylated in vivo by a casein kinase II like kinase.
The alternative scenario is that this site is phosphorylated by still another kinase. For example, one likely candidate is human polo-like kinase (Plk 1), a novel protein kinase family implicated in cell cycle regulation (Golsteyn et al., 1995). Plk
1 is a homolog of the Drosophila polo kinase, which is believed to regulate spindle function and chromosome segregation (Llamazares et al., 1991). It’s possible that
Plk 1 regulates chromosome segregation by phosphorylating the DNA topoisomerase II. Another possible candidate is NIMA kinase. NIMA is a cell cycle- regulated kinase required for the G2/M transition in Aspergillus nidulans. When expressed in other eukaryotic systems, it induced germinal vesicle breakdown in
Xenopus oocytes and premature mitotic events in HeLa cells (Lu and Hunter, 1995).
It is reported that NIMA optimally phosphorylates a FRXS/T site (Lu et al., 1994).
Subsitution of the F residue at the -3 position with A completely abolished NIMA
110 phosphorylation. NIMA is the first kinase reported that requires an aromatic amino acid within three residues of the S/T. Since our proposed model of MPM-2 epitope also requires an aromatic amino acid within three residues at the N-terminal side of the T/S, it’s likely that a NIMA-like kinase(s) is responsible for phosphorylation of the
MPM-2 site on topoisomerase II, and possibly other MPM-2 proteins.
111 MPM-2 Reractivity PTE1 Reactivity Number of Protein Bands + + 10 ++ + 12 + ++ 4 + - 4 - + 6
Data are based on mitotic HeLa cell lysates separated by 7% SDS-PAGE, and are for the purpose of qualitative comparison only.
TABLE 2
Comparison Between MPM-2 and PTE1 Reactive Proteins
112 Fig. 14. Immunoblot analysis of polyclonal topoisomerase II peptide antibodies.
Mitotic (lanes 10-16) or interphase (lanes 2-8) HeLa cell lysates were separated by 7% SDS-PAGE, a narrow portion of the gel was cut for coomassie staining and the remaining gel was transferred to the nitrocellulose paper.
Nitrocellulose transfers were then cut into strips for immunobloting analysis with various antibodies. Lanes 1 and 9, molecular weight marker (kDa); lanes 4 and 12, preimmune serum of the antibody against the dephosphorylated topoisomerase II peptide (Ab1); lanes 5 and 13, Ab1; lanes 6 and 14, preimmune serum of the antibody against the phosphorylated topoisomerase II peptide (Ab2); lanes 7 and
15, Ab2; lanes 2 and 10: PTE1, affinity purified antibody from Ab2; lanes 3 and 11,
MPM-2 antibody; lanes 8 and 16, coomassie staining. Please note the similarities and differences between PTE1 and MPM-2 staining patterns. Also note the predominant staining of mitotic HeLa cell lysates by Ab2 and PTE1.
113 1 2 3 4 5 6 7 8
250- £ * r ' i __ /?< ■ .i.
98
64- ft. e_ ■ 50- ► t o
9 10 11 12 131415 16 250
B m
FIGURH 14
114 Fig. 15. PTE1 recognizes DNA topoisomerase II
A mitotic HeLa lysate was separated on a 4% UREA gel and transferred to nitrocellulose paper. The transfer was then cut into strips and processed for immunoblotting with various antibodies. Lane 1, molecular weight markers (kDa); lane 2, PTE1; lane 3, MPM-2 antibody; lane 4, topoisomerase II antibody. Please notice that the topoisomerase II band comigrated with a band recognized by both
MPM-2 and PTE1.
115 250—► *
. i*-
FIGURE 15
116 Fig. 16. PTE1 specifically recognizes the phosphorylated topoisomerase II peptide on immunoblot
Dephosphorylated (lanes 1 and 3) and phosphorylated (lanes 2 and 4) topoisomerase II peptides were loaded onto an 18.4% SDS-PAGE and transferred to nitrocellulose. The nitrocellulose was then treated with glutaraldehyde to retain the peptides on the blot before proceeding further with the immunological analysis.
Lanes 1 and 2, Ab1; lanes 3 and 4, Ab2. While Ab1 recognized both dephosphorylated and phosphorylated peptides, Ab2 only recognized the phosphorylated peptide.
117 1 2 3 4
30 kDa 16 kDa 4 kDa
FI CUR]- 16
1 18 Fig.17. The phosphorylated topoisomerase II peptide could inhibit binding of the PTE1 IgG.
Mitotic HeLa lysates were separated by 4% UREA gel and transferred to nitrocellulose for the peptide competition experiments. The purified IgG from Ab2
(3 ml solution containing 60 ug IgG) was preincubated with various amounts of either dephosphorylated topoisomerase II peptide (lane 2, 10 ug; lane 3, 100 ug) or phosphorylated peptide (lane 4, 10 ug; lane 5, 100 ug) at RT for 1 hr prior to incubation with the nitrocullose paper. IgG only was used as control (lane 1).
Preincubation with phosphorylated peptide eliminated the antibody’s ability to recognize most bands, including DNA topoisomerase II (arrow).
119 1 2 3 4 5
f i c ;u r i : 1 7
120 Fig. 18. The PTE1 antigens could be partially removed by alkaline phosphatase treatment
Mitotic HeLa lysates were incubated either with dephosphorylation buffer as control {lanes 1 and 3) or with alkaline phosphatase (lanes 2 and 4) at a final concentration of 200 U/ml at 37°C overnight. The samples were then separated by
7% SDS-PAGE and analyzed by either coomassie staining (lanes 1 and 2) or immunoblot by PTE1 (lanes 3 and 4). Notice limited treatment by alkaline phosphatase eliminated a large portion of PTE1 antigens.
121 FIG U R F, 18
122 Fig. 19. PTE1 recognizes rat brain MAPI
Rat brain MTs were separated by 4% UREA gel and transferred to nitrocellulose. The nitrocellulose was then cut into strips and probed with various antibodies. Lane 1, PTE1; lane 2, MPM-2; Iane3, HM-2. Notice PTE1 recognized
MAPI but not MAP2.
123 12 3
FIGURH 19
124 Fig. 20. Immunocytochemical localization of the PTE1 antibody and MPM-2
JAR cells were processed as described in the Materials and Methods for
immunostaining by various antibodies. The fluorescent staining and the corresponding phase contrast image pair of the same cells are shown here. Panels
A, C, E, G, I, K, M, O, phase contrast images; panels B, D, F, H, J, L, N, P, fluorescent staining. Panels A and B, interphase cell stained by the preimmune
serum of Ab2; panels C and D, mitotic cell stained by the preimmune serum of Ab2;
panels E and F, interphase cell stained by Ab2; panels G and H, mitotic cell stained
by Ab2; panels I and J, interphase cell stained by PTEt; panels K and L, mitotic cell
stained by PTE1; panels M and N, interphase cell stained by MPM-2; panels O and
P, mitotic cell stained by MPM-2.
125 FIGURH 20
continued on next page
126 FICIURI; 20 (continued)
127 Fig. 21. Localization of the PTE1 antigens during various mitotic stages
Immunofluorescent staining of JAR cells in different mitotic stages were shown here. Panel A, prometaphase; panel B, metaphase; panel C, early anaphase; panel D, late anaphase; panel E, early telophase; panel F, cytokinesis.
Notice the staining of midbody in panel F. However, centrosome staining was lost during telophase.
128 B
D
FIGIIR!-: 21
129 SUMMARY
Microtubule associated proteins (MAPs) are a group of proteins that bind
MTs in vivo and copurify with MTs. They are classified as either motor proteins or non-motor proteins. Motor proteins include the kinesin superfamily and the dynein family. The non-motor proteins include high molecular weight MAPs (MAPI A,
MAPI B, MAP2, MAP4) and low molecular weight MAP tau. While the motor MAPs are involved in microtubule based movement, non-motor MAPs are believed to be involved in regulating MT dynamics and a wide array of MT functions. For example, a recent report has shown that MAP1B is indispensable for neuronal development in mice (Edelmann et al., 1996). Homozygous MAP1B mutant mice died during embryogenesis, while heterozygous mice exhibited various neuronal disorders.
MAPs also play essential roles in cell division, and are thought to be responsible for the abrupt change of MT dynamics during mitosis, which helps to make the formation of the mitotic spindle possible. It is well established that MAPs are extensively phosphorylated in vivo. The phosphorylation of MAPs has been shown to alter the affinity of MAPs for MTs, and it is also believed that phosphorylation may also regulate the interactions between MAPs and MAP-associated proteins. Some of the MAP phosphorylation events are strictly regulated in a cell cycle and/or
130 neuronal development dependent fashion. One of these regulated phosphorylation sites, recognized by a mouse monoclonal antibody MPM-2, is present on MAP4 during mitosis and MAP1B during neuronal differentiation. Most significantly, the
MPM-2 phosphoepitope is also found on many other proteins during mitosis, including DNA topoisomerase II. Although extensive efforts have been made to study the MPM-2 epitope and kinases, the essential elements of the MPM-2 phosphoepitope remained elusive. Furthermore, the in vivo MPM-2 sites on MAPs and DNA topoisomerase II were not identified. It was also unclear if other potential post-translational modifications, in addition to phosphorylation, might exist on MAPs and play a role in regulating MAP functions. This study attempted to answer some of these questions.
The first chapter documents the work related to the identification of additional post-translational modification on MAPs. It is shown here that MAPI, MAP2, and
MAP4 are glycosylated. The presence of carbohydrate residues on these proteins was indicated by labeling with biotin-hydrazide following periodate oxidation, a specific and well-established method for detecting saccharide moieties on proteins.
Both MAP2 and MAP4 were also labeled in vitro by [3H]UDP-galactose in the presence of galactosyltransferase. Labeling by galactosyltransferase indicated that
MAP2 and MAP4 contained terminal non-reducing N-acetylglucosamine (GlcNAc)
residues, and they appeared to be O-linked to the proteins as shown by their sensitivity to p-elimination. Chromatographic analysis showed that the GlcNAcs
131 were directly linked to the proteins as monosaccharides. This adds MAP2 and
MAP4 to the list of intracellular O-GlcNAc modified proteins, which includes other cytoskeletal proteins such as cytokeratins 8, 13, and 18 and neurofilament proteins
NF-L and NF-M. Further analysis showed that nearly 10% of the MAP2 isolated from rat brain is modified by O-GlcNAc. This estimate is thought to reflect the minimal level of O-GlcNAc modification present on MAP2, however. Both the O-
GlcNAc and biotin-hydrazide-reactive carbohydrate moieties are shown to locate on the projection domain of MAP2. Three O-GlcNAc containing peaks were observed following FPLC chromatography of a tryptic digest of MAP2, suggesting that multiple modification sites exist. The specific modification sites and functional significance of the O-GlcNAc glycosylation on the HMW MAPs remains to be determined.
Chapter 2 and 3 describe the work related to the MPM-2 phosphoepitope.
Based upon the analysis of reported MPM-2 reactive sequences, a model for the essential elements that comprise the MPM-2 phosphoepitope was developed. This model was tested by employing a series of synthetic phosphopepttdes. A 14 amino acid synthetic phosphopeptide derived from a potential MPM-2 site on human DNA topoisomerase II is shown to be recognized by the MPM-2 antibody. This phosphopeptide was sufficient to compete for antibody binding to a native MPM-2 antigen on MAP4, indicating that it contained all of the essential elements of the
MPM-2 epitope. By substituting selected amino acids with alanine, the various contribution of different amino acids to the binding between the MPM-2 antibody
132 and the epitope were examined. Changing the amino acid that was adjacent to the phosphorylated threonine residue on the C-terminal side (the +1 position) had no effect on MPM-2 antibody binding. However, substitution of aromatic amino acids at either the -2 or +2 positions reduced antibody recognition. The aromatic amino acid at the -2 position appeared to be the most critical residue of those tested that influenced antibody binding. Chapter 2 describes the effort made to characterize
MPM-2 phosphoepitope. These results provide information required for the molecular definition of the MPM-2 epitope, and should aid in the identification of potential MPM-2 reactive sites on other mitotic phosphoproteins.
To test the hypothesis that this in vitro MPM-2 reactive DNA topoisomerase
II sequence is indeed phosphorylated in vivo, this topoisomerase II phosphopeptide was used to generate a rabbit polyclonal antibody. Like MPM-2 antibody, this antibody, PTE1, recognizes a large number of mitotic proteins. Interestingly, although some protein bands are recognized by both PTE1 and MPM-2 as shown in immunoblots, each antibody recognizes a unique subset of protein bands. Among the proteins recognized by both MPM-2 and PTE1 in mitotic HeLa lysate is a band co-migrating with DNA topoisomerase II. A large proportion of PTE1 antigens from mitotic HeLa cells could be removed after limited dephosphorylation by alkaline phosphatase, verifying PTE1 antigen probably contains a phosphate. Most importantly, the PTE1 antibody only recognized the phosphorylated topoisomerase
II peptide on the immunoblot. The specificity of PTE1 antibody and the presence of
133 a phosphate within its epitope are confirmed by the competition experiments using the purified IgG. The antibody lost its ability to recognize most bands if preincubated with the phosphopeptide, but was unaffected if preincubated with the non- phosphorylated peptide. The fact that only phosphopeptide could compete with the endogenous topoisomerase II for the binding to the polyclonal antibody indicates that this sequence is probably phosphorylated in vivo and is the site that contributes to DNA topoisomerase II MPM-2 reactivity during mitosis. One other protein recognized by both antibodies is rat brain MAPI.
As expected, immunocytochemical staining indicated that cellular PTE1 antigens could only be observed during mitosis. While the immune serum showed some background staining, the IgG and affinity purified antibody do not show any generalized background staining. PTE1 also does not show any filamentous staining as observed using the MPM-2 antibody. Analysis of the PTE1 antigens’ distribution during various mitotic stages has yielded a unique pattern as compared to MPM-2 staining pattern. Both chromosomes and centrosomes become PTE1 reactive as the cells enter prophase, and PTE1 antigens appear to be distributed evenly along all chromosomes. No other major cellular structures are stained.
During metaphase, similar staining of chromosomes and spindle poles could be observed. The staining seems to diminish when cells enter anaphase. The PTE1 antigens could also be observed during cytokinesis at the midbodies. The appearance of PTE1 antigens on chromosomes correlates in time with the
134 maximum activity of DNA topoisomerase II, which is shown to be essential for
chromosome condensation during prophase and sister chromatid separation during
metaphase. By analogy, the presence of PTE1 antigens on spindle poles and
midbodies might also reflect the regulatory events on the corresponding protein
components of these structures. It is possible that one or a group of related kinases
phosphorylates these potential PTE1 sites on various proteins during mitosis to
regulate their functions. This work should provide a viable approach to investigate
mitotic regulated proteins and their corresponding mitotic activated kinase(s).
In summary, this study documents efforts in the investigation of novel post-
translational modification on MAPs and the characterization of the phosphorylation
sites shared by MAPs and DNA topoisomerase II. HMW MAPs are shown here to
be modified by glycosylation, a previously unknown post-translational modification
on MAPs. In addition, MAP2 and MAP4 are shown to contain O-GlcNAc, a unique
cytoplasmic/nuclear specific glycosylation. Systematic work has been done to
partially characterize the MPM-2 phosphoepitope, which has revealed the
importance of the flanking aromatic amino acids for the first time. Various lines of
evidence obtained also support the hypothesis that the selected topoisomerase II
sequence is indeed the in vivo MPM-2 site on DNA topoisomerase II. Finally, PTE1, the rabbit polyclonal antibody generated against the topoisomerase II
phosphopeptide, stains important structures during mitosis and should serve as an
invaluable tool for analyzing mitotic specific phosphorylation.
135 REFERENCES
Ali, J.A., Jackson, A.P., Howells, A.J. and Maxwell, A. (1993) Biochemistry 32: 2717-2724.
Anderson, N. G., Mailer, J. L., Tonks, N. K. and Sturgill, T. W. (1990) Nature 343, 651-653.
Aquino, D.A., Margolis, R.U., and Margolis, R.K. (1984a) J. Celt Biol. 99, 1117- 1129
Aquino, D.A., Margolis, R.U., and Margolis, R.K. (1984b) J. Ceil Biol. 99,1130- 1139
Austin, C.A. and Fisher, L.M. (1990) Sci. Progress 74: 147-162.
Austin, C.A., Sng, J.-H., Patel, S. and Fisher, L.M. (1993) Biochem. Biophys. Acta 1172: 283-291.
Berger, J.M., Gamblin, S.J., Harrison, S.C. and Wang, J.C. (1996) Nature 379: 225-232.
Briones, E. and Wiche, G. (1985) Proc. Natl. Acad. Sci. USA 82, 5776-5780
Burg, M.A. and Cole, G.J. (1994) J. Neurobiol. 25, 1-22.
Burden, D.A., Goldsmith, L.J. and Sullivan, D.M. (1993) Biochem. J. 293, 297- 304.
Caceres, A. and Kosik, K.S. (1990) Nature 343, 461-463
Cardenas, M.E., Dang, Q., Glover, C. V. and Gasser, S. M. (1992) EMBO J. 11, 1785-1796.
Cardenas, M.E. and Gasser, S.M. (1993) J. Cell Sci. 104, 219-225.
136 Cardenas, M.E., Walter, R., Hanna, D. and Gasser, S.M. (1993) J. Cell Sci. 104, 533-543.
Centonze, V.E. and Borisy, G.G. (1990) J. Cell Sci. 95, 405-411.
Chakraborty, A., Saha, D., Bose, A., Chatterjee, M.t and Gupta, N.K. (1994) Biochem. 33, 6700-6706
Chou, C.-F., Smith, A.J., and Omary, M.B. (1992) J. Biol. Chem. 267, 3901-3906
Chou, C.-F. and Omary, M.B. (1993) J. Biol. Chem. 268, 4465-4472
Chung, T.D.Y., Drake, F.H., Tan, K.B., Per, S.R., Crook, S.T. and Mirabelli, C.K. (1989) Proc. Natl. Acad. Sci. USA 86: 9431-9435.
Coligan, J. E., Kruisbeek, A. M., Margulies, D. H., Shevach, E. M. and Strober, W. (1994) Current Protocols in Immunoloby. Vol. 2, Unit 9.4.1. John Wiley & Sons, Inc. USA.
Corbett, A.H., DeVore, R.F. and Osheroff, N. (1992) J. Biol. Chem. 267, 20513- 20518.
Corbett, A.H. and Osheroff, N. (1993a) Chem. Res. Toxicol. 6, 585-597.
Corbett, A.H., Hong, D. and Osheroff, N. (1993b) J. Biol. Chem. 268, 14394-14398.
Dang, Q., Alghisi, G. C. and Gasser, S. M. (1994) J. Mol. Biol. 243, 10-24.
Davis, F. M., Tsao, T. Y., Fowler, S. K. and Rao, P. N. (1983) Proc. Natl. Acad. Sci. USA 80, 2926-2930.
Davis, F. M., Wright, D. A., Penkala, J. E. and Rao, P. N. (1989) Cell. Struct. Funct. 14, 271-277. 5. Dinsmore, J.H. and Solomon, F. (1991) Cell64, 817-826
DeVore, R.F., Corbett, A.H. and Osheroff, N. (1992) Cancer Res. 52, 2156-2161.
DiNardo, S., Voelkel, K. and Sternglanz, R. (1984) Proc. Natl. Acad. Sci. USA 81, 2616-2620
Dong, D.L.-Y., Xu, Z.-S., Chevrier, M.R., Cotter, R.J., Cleveland, D.W., and Hart, G.W. (1993) J. Biol. Chem. 268, 16679-16687
Dong, D.L.-Y. and Hart, G.W. (1994) J. Biol. Chem. 269, 19321-19330
137 Downes, C. S., Clarke, D. J., Muilinger, A. M., Gimenez-Abian, J. F., Creighton, A. M. and Johnson, R. T. (1994) Nature 372, 467-470.
Drake, F.H., Zimmerman, J.P., McCabe, F.L., Bartus, H.F., Per, S.R., Sullivan, D.M., Ross, W.E., Mattem, M.R., Johnson, R.K., Crooke, S.T. and Mirabelli, C.K. (1987) J. Biol. Chem. 262: 16739-16747.
Drake, F.H., Hofmann, G.A., Bartus, H.F., Mattem, M.R., Johnson, R.K., Crooke, S.T. and Mirabelli, C.K. (1989) Biochemistry 28: 8154-8160.
Dubois, M., Gilles, K., Hamilton, J., Rebers, P., and Smith, F. (1956) Anal. Chem. 28, 350-356
Earnshaw, W.C., Halligan, B., Cooke, C.A., Heck, M.M.S. and Liu, L.F. (1985) J. Cell Biol. 100: 1706-1715.
Earnshaw, W.C. and Mackay, A.M. (1994) FASEB J. 8: 947-956.
Edelmann, W., Zervas, M., Costello, P., Roback, L„ Fischer I., Hammarback, J.A., Cowan, N., Davis, P., Wainer, B. and Kucherlapati, R. (1996) Proc. Natl. Acad. Sci. USA. 93(3), 1270-1275.
Eldik, L.J.V. and Wolchok, S.R. (1984) Biochem. Biophys. Res. Commun. 124, 752- 759
Gasser, S.M., Laroche, T., Falquet, J., Boy de la Tour E., and Laemmli, U.K. (1986) J. Mol. Biol. 188, 613-629.
Golsteyn, R.M., Mundt, K.E., Fry, A.M. and Nigg, E.A. (1995) J. Cell Biol. 129: 1617- 1628.
Haltiwanger, R.S., Kelly, W.G., Roquemore, E.P., Blomberg, M.A., Dong, L.-Y.D., Kreppel, L., Chou, T.-Y., and Hart, G.W. (1992a) Biochem. Soc. Trans. 20,264-269
Haltiwanger, R.S., Blomberg, M.A., and Hart, G.W. (1992b) J. Biol. Chem. 267, 9005-9013
Hart, G.W., Haltiwanger, R.S., Holt, G.D., and Kelly, W.G. (1989) Annu. Rev. Biochem. 58, 841-874
Heck, M.M., Hittelman, W.N. and Earnshaw, W.C. (1989) J. Biol. Chem. 264, 15161-15164.
Hirano, T. and Mitchison T. (1991) J. Cell Biol. 115, 1479-1489.
138 Hirano, T. and Mitchison T. (1993) J. Cell Biol. 120, 601-612.
Holm, C., Goto, T., Wang, J.C. and Botstein, D. (1985) Ce//41, 553-563.
Hyams, J.S. and Lloyd, C.W. (1994) Microtubules, 1st Ed., Wiley-Liss, New York.
Jenkins, J.R., Ayton, P., Jones, T., Davies, S.L., Simmons, D.L., Harris, A.L., Sheer, D. and Hickson, I.D. (1992) Nucleic Acids Res. 5587-5592.
Joly, J.C., Flynn, G., and Purich, D.L. (1989) J. Cell Biol. 109, 2289-2294
Kelly, W.G., Dahmus, M.E., and Hart, G.W. (1993) J. Biol. Chem. 268,10416-10424
Kimble, M. and Kuriyama, R. (1992) Inter. Rev. Cytol. 136, 1-50.
King, I.A. and Hounsell, E.F. (1989) J. Biol. Chem. 264, 14022-14028
Kirschner, M. and Mitchison, T. (1986) Cell45, 329-342.
Kobata, A. (1994) Methods Enzymol. 230, 200-208
Kuang, J., Zhao, J. Y., Wright, D. A., Saunders, G. F. and Rao, P. N. (1989) Proc. Natl. Acad. Sci. USA 86, 4982-4986.
Kuang, J. and Ashorn, C. L. (1993) J. Cell Biol. 123, 859-868.
Kuang, J., Ashom, C. L., Gonzalez-Kuyvenhoven, M. and Penkala, J. E. (1994) Mol. Biol. Cell 5, 135-145.
Kuriyama, R. and Borisy, G.G. (1981) J. Cell Biol. 91, 822-826.
Ledesma, M.D., Bonay, P., Colaco, C., and Avila, J. (1994) J. Biol. Chem. 269, 21614-21619.
Llamazares, S., Moreira, A., Tavares, A., Girdham, C., Spruce, B., Gondalez, C., Karess, R., Glover, D.M. and Sunkel, C.E. (1991) Genes Dev. 5: 535-544.
Lu, K.P., Kemp, B.E. and Means, A.R. (1994) J. Biol. Chem. 269: 6603-6607.
Lu, K.P. and Hunter, T. (1995) Ce//81: 413-424.
Mueller, P. R., Coleman, T. R., Kumagai, A. and Dunphy, W. G. (1995a) Science 270, 86-90.
139 Mueller, P. R., Coleman, T. R. and Dunphy, W. G. (1995b) Mol. Biol. Cell6, 119- 134.
Muramoto, K., Taniguchi, H., Kawahara, M., Kobayashi, K., Nonomura, Y., and Kuroda, Y. (1994) Biochem. Biophys. Res. Commun. 205, 1467-1473
Murthy, A. and Flavin, M. (1983) Eur. J. Biochem. 137, 37-46.
Olmsted, J.B. (1986) Annu. Rev. Cell Biol. 2, 421-457
Ookata, K., Hisanaga, S.-i., Bulinski, J.C., Murofushi, H., Aizawa, H., Itoh, T.J., Hotani, H., Okumura, E., Tachibana, K., and Kishimoto, T. (1995) J. Cell Biol. 128, 849-862
Payne, D. M., Rossomando, A. J., Martino, P., Erickson, A. K., Her, J. H., Shabanowitz, J., Hunt, D. F., Weber, M. J. and Sturgill, T. W. (1991) EMBOJ. 10, 885-892.
Pinna, L.A. (1990) Biochim. Biophys. Acta. 1054, 267-284.
Poljak, L. and Kas E. (1995) TICB 5, 348-354.
Rieder, C.L. (1982) Int. Rev. Cytol. 79, 1-57.
Roca, J. (1995) TIBS 20, 156-160.
Roquemore, E.P., Chou, T.-y., and Hart, G.W. (1994) Methods Enzymol. 230, 443- 460
Rose, D. and Holm, C. (1993) Mol. Cell. Biol. 13, 3445-3455.
Saijo, M., Ui, M. and Erromoto, T. (1992) Biochemistry 31, 359-363.
Saitoh, Y. and Laemmli, U.K. (1994) Cell76: 609-622.
Salmon, E.D., Leslie, R.J., Saxton, W.M., Karow, M.L. and McIntosh, J.R. (1984) J. Cell Biol. 99, 2165-2174.
Saxton, W.M., Stemple, D.L., Leslie, R.J., Salmon, E.D., Zavortink, M. and McIntosh, J.R. (1984) J. Cell Biol. 99, 2175-2186.
Schafer, I.A. and Sorrell, J.M. (1993) Exp. Cell Res. 207, 213-219
140 Soltys, B.J. and Gupta, R.S. (1992) Biochem. Cell Biol. 70, 1174-1186.
Swedlow, J.R., Sedat, J.W. and Agard, D.A. (1993) Cell 73: 97-108.
Taagepera, S., Rao, P. N., Drake, F. H. and Gorbsky, G. J. (1993a) Proc. Natl. Acad. Sci. USA 90, 8407-8411.
Taggepera S., Huang, K. and Gorbsky, G. J. (1993b). Mol. Biol. Cell A, 350a.
Taagepera, S., Dent, P., Her, J „ Sturgill, T. W. and Gorbsky, G. J. (1994) Mol. Biol. CellS, 1243-1251.
Theurkauf, W.E. and Vallee, R.B. (1982) J. Biol. Chem. 257, 3284-3290
Tombes, R. M., Peloquin, J. G. and Borisy, G. G. (1991) Cell Regulation 2, 861-874.
Uemura, T. and Yanagida, M. (1986) EMBOJ. 5: 1003-1010.
Uemura, T., Ohkura, H., Adachi, Y., Morino, K., Shiozaki, K. and Yanagida, M. (1987) Cell 50, 917-925.
Vallee, R. (1980) Proc. Natl. Acad. Sci. USA 77, 3206-3210
Vallee, R.B. and Collins, C.A. (1986) Methods Enzymol. 134, 124-126
Vandre, D. D., Davis, F. M., Rao, P. N. and Borisy, G. G. (1984) Proc. Natl. Acad. Sci. USA 81, 4439-4443.
Vandre, D. D. and Borisy, G. G. (1985) in: Cell Motility, Mechanism and Regulation (Ishikawa, H., Hatano, S. and Sato, H. eds), pp. 389-401, University of Tokyo Press, Tokyo.
Vandre, D. D., Davis, F. M., Rao, P. N. and Borisy, G. G. (1986) Eur. J. Cell Biol. 41, 72-81.
Vandre, D.D., Centonze, V.E., Peloquin, J., Tombes, R.M., and Borisy, G.G. (1991) J. Cell Sci. 98, 577-588
Vassetzky, Y. S., Dang, Q., Benedetti, P. and Gasser, S. M. (1994) Mol. Cell. Biol. 14, 6962-6974.
Watt, P. M. and Hickson, I.D. (1994) Biochem. J. 303, 681-691.
Wells, N. J. and Hickson I. D. (1995) Eur. J. Biochem. 231, 491-497.
141 Westendorf, J. M., Rao, P. N. and Gerace, L. (1994) Proc. Natl. Acad. Sci. USA 91, 714-718.
Whalen, A.M., McConnell, M. and Fisher, P.A. (1991) J. Cell Biol. 112, 203-213.
Wilchek, M. and Bayer, E.A. (1987) Methods Enzymol. 138, 429-442
Woessner, R.D., Chung, T.D.Y., Hofmann, G.A., Mattem, M.R., Mirabelli, M.R., Drake, F.H. and Johnson, R.K. (1990) Cancer Res. 50: 2901-2908.
Woessner, R.D., Mattern, M.R., Mirabelli, M.R., Johnson, R.K. and Drake, F.H. (1990) Cell Growth Dirrentiation 2: 209*214.
Yan, S.-D., Chen, X., Schmidt, A.-M., Brett, J., Godman, G., Zou, Y.S., Scott, C.W., Caputo, C., Frappier, T., Smith, M.A., Perry, G., Yen, S.-H., and Stern, D. (1994) Proc. Natl. Acad. Sci. USA 91, 7787-7791
Zini, N., Martelli, A.M., Sabatelli, P., Santi, S., Negri, C., Astaldi Ricotti, G.C.B., and Maraldi, N.M. (1992) Exp. Cell Res. 200: 460-466.
142 APPENDIX
THE MICROTUBULE CYTOSKELETON
IN HUMAN PHAGOCYTIC LEUKOCYTES IS
A HIGHLY DYNAMIC STRUCTURE
143 ABSTRACT
The microtubule cytoskeleton of human leukocytes has been difficult to study, in part, due to the lack of a reliable protocol for the indirect immunofluorescence staining of microtubules in these cells. We report here the development of a simple and reliable immunocytochemical labeling protocol for the examination of microtubules in leukocytes including monocytes, neutrophils, and eosinophils. The dynamic properties of microtubules in both monocytes and neutrophils were examined by indirect immunofluorescence staining of cells following exposure to nocodazole. Nocodazole-induced depolymerization is extremely rapid in both cell types, as is the regrowth of microtubules following removal of the nocodazole. Rapid reorganization of the microtubule cytoskeleton was also observed in neutrophils undergoing chemotactic stimulation. Bundling of microtubules was observed in both monocytes and neutrophils isolated from patients undergoing taxol infusion chemotherapy. The taxol-induced bundles were transient in nature as they were absent from samples collected 48 hours following the completion of the taxol infusion. These results demonstrate the unique dynamic properties of leukocyte microtubules, and indicate that they can be altered in vivo.
The development of this staining protocol should allow for the further analysis of leukocyte microtubules as related to the normal functional response of these cells and form the basis for correlating alterations in microtubule dynamics with the effects of taxol on leukocyte function. INTRODUCTION
The microtubule cytoskeleton plays important roles in cytoplasmic organization, intracellular transport, and cellular polarity (Schliwa et al., 1986). The direct visualization of the microtubule cytoskeleton by procedures such as indirect
immunofluorescence microscopy has proven to be an important technique for the elucidation of these functions. However, immunocytochemical examination of microtubules has been more difficult in leukocytes than in most other cell types.
Regardless, the involvement of microtubules in the specialized functions of leukocytes has been suggested from a number of studies.
The general organization of the cytoplasmic microtubular array appears to be involved in both directed motility and target recognition in different types of leukocytes. Repositioning of the microtubule organizing center (MTOC) has been shown to accompany target recognition of cytotoxic T-lymphocytes (Geiger et al.,
1982), and is required for target cell lysis (Knox et al., 1993). Similarly, MTOC
repositioning occurs during the interaction of neutrophils with antigen-antibody complexes (Chiplonkar et al., 1992). Repositioning of the MTOC also occurs during the directed motility of both macrophages (Nemere et al., 1985) and neutrophils
(Anderson et al., 1982; Keller and Niggli, 1993; Schliwa et al., 1982) in response to chemotactic agents, and laser irradiation of the MTOC has been shown to disrupt directed motility in the newt eosinophil (Koonce et al., 1984).
145 Directed movement of granules and neutrophil degranulation involves the microtubule cytoskeleton {Mollinedo et al., 1989), as well as the extension of a tubular lysosome complex in stimulated macrophages (Robinson and Luo, 1992;
Swanson et al., 1987). Interestingly, the microtubule based motor molecule kinesin has been shown to be present in neutrophils and associated with granules from these cells (Rothwell et al., 1993a), and the extension of lysosomes in macrophages is kinesin dependent (Swanson et al., 1992).
Reorganization of the microtubular cytoskeleton has been shown to accompany differentiation of the promyelocytic leukemic cell line HL-60 along the granulocytic pathway (Leung et al., 1992), or the monocytic pathway (Katagiri,
1993), and suggests that differentiation of leukocytes may be regulated in part by alterations in the organization of microtubules or regulation of microtubule assembly/disassembly dynamics. Microtubule alterations have also been shown to be associated with the malignant phenotype associated with certain human T-cell leukemic cell lines in comparison to normal peripheral blood lymphocytes (Anand and Chou, 1993). Mitogenic activation of normal peripheral blood T-cells is also associated with changes in microtubule distribution, and is accompanied by alterations in the expression of microtubule associated proteins (Anand and Chou,
1992). Taken together these results indicate that alterations in the microtubular cytoskeleton are common to leukocytes in different stages of differentiation or
146 activation, but it remains to be determined whether these alterations result in or from
the phenotypic changes in the cells themselves.
More recently, it has been shown that the integrity of microtubules plays a
direct role in the expression of certain genes in leukocytes. For example, disruption
of microtubules in monocytes by colchicine treatment leads to an increase in
interleukin-1 synthesis, which is involved in the activation of monocytes (Manie et
al., 1993). Conversely, stabilization of macrophage microtubules by taxol has been
shown to influence the production of various cytokines such as tumor necrosis factor (Ding et al., 1993).
Leukocytes are motile cells that have the potential to undergo rapid shape
changes; it is possible that the microtubule cytoskeleton in leukocytes has the ability
for rapid dynamic reorganization. We have recently demonstrated that microtubule
dynamics in mouse macrophages are extremely rapid, and that there is a rapid
modulation of the tyrosination state of crtubulin following phorbol ester stimulation
(Robinson and Vandre, submitted). Rapid microtubule turnover rates have also
been reported in human monocytes (Cassimeris et al., 1986). Examination of
neutrophil microtubules and a determination of their dynamic properties has been
difficult owing to the inability to visualize reliably these structures by
immunofluorescence techniques.
147 In order to characterize the dynamics of the microtubule cytoskeleton in human leukocytes we began a systematic examination of different fixation protocols for indirect immunofluorescence staining of microtubules that would be applicable to the majority if not all leukocyte cell types. We have applied a novel fixation protocol to human leukocytes that allows for the immunocytochemical localization of microtubules in all types of circulating leukocytes. Using this technique, we have documented the number of microtubules present in different human leukocytes, and demonstrated the unusually rapid dynamics of microtubules in both human monocytes and neutrophils. We have confirmed the rapid alterations that occur in the neutrophil microtubule cytoskeleton in response to chemotactic stimulation, but, in addition, we describe differences in the immunofluorescence staining properties of microtubules between stimulated and unstimulated neutrophils. Moreover, this technique has allowed, for the first time, a demonstration of alterations in monocyte and neutrophil microtubules of patients undergoing taxol chemotherapy. A preliminary account of this work has appeared in abstract form (Ding et al., 1993).
148 MATERIALS AND METHODS
Human leukocytes were isolated from freshly drawn heparinized blood obtained from healthy normal donors or ovarian cancer patients undergoing taxol chemotherapy, according to a research protocol approved by the Ohio State
University human subject review board. Blood samples were layered onto
Nycomed PolymorphoprepTM (Accurate Chemicals, Westbury, NY) and centrifuged at 2000 rpm for 30 minutes at room temperature using a HS-4 rotor in a Sorvall RC
2-B centrifuge. The cellular layer containing polymorphonuclear leukocytes and eosinophils was carefully removed and equilibrated with an equal volume of 0.45%
NaCI. The mononuclear cell layer was also removed, and then both cell fractions were equilibrated with an excess of HBSS buffer (5.36 mM KCI, 5.55 mM glucose,
0.44 mM anhydrous KH2P04, 137 mM NaCI, 0.34 mM anhydrous Na JHP044.17 mM
NaHC03,0.1% BSA, pH 7.2). Cells were pelleted by centrifugation for 5 minutes using an IEC clinical centrifuge. The cell pellets were resuspended in 1 ml of distilled H20 for 30 seconds to lyse contaminating red blood cells, and then mixed with 10-12 ml HBSS . Cells were subsequently pelleted as before.
Isolated cells were resuspended with PMN buffer (10 mM HEPES, 150 mM
NaCI, 5 mM KOH, 1.2 mM MgCI2, 1.3 mM CaCJ, 5.5 mM glucose, pH 7.5) and allowed to adhere to the coverslip for 12 minutes at 37°C. Adherent cells were fixed using a variety of different fixation protocols as described in Table 3. In protocols
149 1 -5 cells were first rinsed for 30 sec in PHEM microtubule stabilization buffer (60 mM PIPES, 25 mM HEPES, 10 mM EGTA, 2 mM MgCI2, pH 6.9) prior to fixation.
Aqueous fixatives and lysis buffers were also prepared in PHEM buffer. Protocols
6 and 7 followed the procedures described by Anderson et al., (Anderson et al.,
1982), and Rothwell et al., (Rothwell et al., 1993a; Rothwell et al., 1993b), respectively. However, the best results were obtained using protocol 8, and is described here in detail. After a brief rinse (30 sec) in PBSa (140 mM NaCI, 1.5 mM
KH2P04,2.7 mM KCI, 6.5 mM NaJHPO, 0.002% NaN ,pH 7.4), cells were fixed with
0.7% glutaraldehyde in PBSa for 15 min at 22°C. Cells were washed in PBSa followed by a lysis/extraction in PBSa containing 0.5% SDS (sodium dodecyl sulfate) for 15 minutes. In some cases, improved staining was obtained if a lysis step in PBSa containing 0.5% Triton X-100 for 15 min preceded the lysis/extraction in 0.5% SDS. Cells were washed in PBSa and unreacted aldehyde groups were reduced by treatment with 2 changes of NaBH4[1-2 mg/ml in TBS (0.9% NaCI, 10 mM Tris-base, pH 7.4)] over 20-30 minutes. The cells were washed again with three changes of PBSa.
Cells were processed for immunofluorescence staining as follows.
Non-specific antibody binding was reduced by an incubation with 4% normal goat serum for 30 minutes prior to primary antibody addition. Coverslips were washed once with PBSa and the cells were incubated with primary antibody for 1 hour, washed with PBSa 3 times, incubated with secondary antibody for another 30
150 minutes. Primary antibodies used included clone Tub1A2, a mouse monoclonal antibody specific for tyrosinated a-tubuiin (Sigma, St. Louis, MO); a mixture of clones DM1A and DM1B, mouse monoclonal antibodies specific fora- and 3-tubulin respectively (Amersham, Arlington Heights, IL); and a rabbit polyclonal antibody specific for tyrosinated a-tubulin (a gift from Dr. Chloe Bulinski, Columbia University,
New York, NY). The secondary antibodies used were FITC conjugated goat anti-mouse and goat anti-rabbit antibodies (Cappell, Durham, NC). Both primary and secondary antibodies were diluted in PBSa containing 4% normal goat serum and all antibody incubations were carried out at 37°C. Coverslips were mounted in
Mowiwol mounting medium containing p-phenylenediamine (1 mg/ml) as an anti-bleaching agent. Cells were examined using a Zeiss IM35 epifluorescence microscope with a Nikon 100X 1.4 NA objective. Photomicrographs were recorded using Kodak T-MAX 400 ASA film.
For the determination of microtubule turnover, attached cells were treated with nocodazole (Methyl-(5-[2-Thienylcarbonyl]-1H-Benzimidazol-2-YL) Carbamate) at a final concentration of either 13 or 1.3 pM for different time periods at 37°C. For microtubule regrowth experiments, cells were first treated with nocodazole (1.3 ^M) for 10 minutes and then rapidly washed with two changes of PMN buffer. Following release from the nocodazole, cells were incubated in PMN buffer at 37°C for different periods of time. Microtubule stability was also determined in cells incubated at 4°C for various time periods in PMN buffer. For cell stimulation
151 experiments, leukocytes were treated with fMLP (10~7 M) for 0-15 min after their attachment to coverslips (see above). Following treatment with the microtubule depolymerizing agent nocodazole or with fMLP the leukocytes were processed as described above.
152 RESULTS
Immunofluorescence staining of microtubules in human leukocytes.
In order to examine the microtubule cytoskeleton in isolated human leukocytes we employed a variety of fixation and lysis protocols to define a suitable and reproducible method for the visualization of microtubules by indirect immunofluorescence staining in these cells (Table 3, and see Methods). While monocyte microtubules were visualized to varying degrees using each of the fixation and lysis protocols examined, we were unable to obtain any staining of neutrophil microtubules using standard immunofluorescence protocols (Table 3, fixation protocols 1-5), which are known to give satisfactory results in a variety of other cell types. The lack of microtubule staining in neutrophils was not due to the selective sensitivity of antigenic sites to glutaraldehyde fixation, since microtubules were not observed in either methanol or paraformaldehyde fixed samples. In addition, this lack of microtubule staining in neutrophils was not due to the specificity of the tubulin antibodies employed, as we examined 13 different tubulin antibodies obtained from both commercial sources as well as individual investigators (data not presented).
More elaborate fixation and lysis protocols have been reported as being suitable for the staining of neutrophil microtubules (Anderson et al., 1982; Rothwell
153 et al., 1993a; Rothwell et al., 1993b), but these typically involve repeated washing of the cells in microtubule stabilization buffers both before and after cell lysis and/or include several postfixations (Table 3, fixation protocols 6 and 7). The results we have obtained with these staining protocols have been extremely variable.
Typically, neutrophils processed for immunofluorescence using fixation protocol 6
(Anderson et al., 1982) showed clear centrosomal staining, but microtubule staining was either absent, or only one or two microtubules were stained (data not presented; Bell et al., 1987). In neutrophils stained using fixation protocol 7
(Rothwell et al., 1993a; Rothwell et al., 1993b) microtubules were occasionally visualized; however, the staining pattern was clearly incomplete (e.g. few microtubules were observed extending to the cell margins, see below). Examination of whole mount samples of neutrophils by electron microscopy has shown that intact microtubules are present in similarly processed neutrophils (Schliwa et al., 1982), which suggested to us that microtubules were present in most if not all of the neutrophils processed using the fixation protocols presented in Table 3. Thus, the lack and/or inconsistency of microtubule staining in neutrophils could have resulted from a masking of the tubulin antigenic sites that would have reduced access of the primary antibody to the tubulin. In an effort to explore this possibility, we examined the effect of several chaotropic agents on the immunofluorescence labeling of neutrophil microtubules. This included the extraction of cells with high salt concentrations, exposure to guanidine hydrochloride, and extraction with the ionic detergent SDS. A lysis/extraction of fixed neutrophils with 0.5% SDS was shown
154 to give superior indirect immunofluorescence staining of neutrophil microtubules as compared to any of the other protocols tested (Table 3, fixation protocol 8). We have used three different anti-tubulin antibodies (see Materials and Methods) with success, and therefore, we utilized this fixation protocol as a general procedure to further examine the microtubule cytoskeleton of human leukocytes.
A comparison of the typical microtubule staining patterns obtained in monocytes and neutrophils using different fixation protocols is shown in Figure 22.
While microtubules can be observed in monocytes following detergent lysis of cells prior to fixation (Fig. 22A), the majority do not extend to the periphery of the cell boundaries and they typically have a curled or sinuous morphology. In contrast, the number of microtubules observed is much higher in samples fixed prior to lysis in
Triton X-100 (data not presented) and/or lysis/extraction in SDS (Fig. 22B). In addition, the microtubules typically are straight, and radiate from the centrosome to the boundaries of the cell. Similarly, only a few microtubules were detected in neutrophil samples using fixation protocols previously reported to be suitable for the examination of neutrophil microtubules such as protocol 7, Table 3 (Fig 22C). The microtubule pattern obtained using this fixation was incomplete, and staining was rarely observed in portions of the cytoplasm that extended beyond the region of the nucleus (Fig. 22C). In comparison, a much more complete and elaborate microtubule network was observed in neutrophils that underwent a lysis/extraction step with 0.5% SDS following fixation (Fig. 22D). Microtubule staining was clearly
155 present in regions of the cytoplasm that extended beyond the nucleus; however, staining was somewhat variable in unstimulated neutrophils in that it was sometimes difficult to trace the staining of an individual microtubule completely back to the centrosome in some cells (compare Fig. 22D with Fig. 23D).
Critical to an evaluation of the microtubule cytoskeleton in different human leukocytes is the unambiguous identification of the cell type by independent and specific morphological characteristics such as nuclear morphology. In addition to improved microtubule staining, nuclear morphology was not affected by the iysis/extraction of fixed samples with 0.5% SDS (Fig. 23). Therefore leukocyte cell types could be positively identified. Examples of a typical monocyte (Fig. 23A and
B), a neutrophil (Fig. 23C and D), and an eosinophil (Fig. 23E and F) are presented.
In each case the microtubules radiate from the single centrosome located near the nucleus. A representative field of cells from a mixed leukocyte population shows that microtubule arrays are also stained in lymphocytes and platelets following
Iysis/extraction of fixed samples with 0.5% SDS (Fig. 24). We have also found this fixation protocol to be effective for the examination of microtubules and intermediate filaments in cultured epithelial cells (Vandre, unpublished data). The average number of microtubules observed by indirect immunofluorescence staining of monocytes and neutrophils using the SDS Iysis/extraction fixation protocol is presented in Table 4, and is compared to previous reports of microtubule numbers in leukocytes. Interestingly, we observed more than twice the number of
156 microtubules in human monocytes than previously reported by indirect immunofluorescence (Cassimeris et al., 1986). This discrepancy may in part be due to the fixation conditions in the earlier report where monocytes were lysed prior to their fixation, whereas cells were fixed prior to extraction/lysis in this study. The number of microtubules in both stimulated and unstimulated neutrophils have been previously determined by electron microscopy of whole mount preparations, and a slight increase in numbers were observed in stimulated cells.
Microtubule numbers were also reported in neutrophils following immunofluorescence staining, but in this case no significant differences were observed between stimulated and unstimulated cells (Anderson et al., 1982). We have been unable to detect the entire set of microtubules using the fixation protocol presented by Anderson et al (Anderson et al., 1982). We have obtained estimates of microtubule numbers with the current SDS/extraction protocol that are more similar to those reported by Schliwa et al (Schliwa et al., 1982). We also find a slight increase in microtubule number between stimulated and unstimulated cells.
The current estimates are nearly twice those obtained from the examination of thin sections by electron microscopy (Hoffstein and Weissman, 1978); however, it should be noted that those determinations were derived from cells in suspension.
157 Microtubules in human monocytes and neutrophils are unusually dynamic.
Microtubules are dynamic structures, and have been shown to have relatively rapid turnover kinetics in human monocytes (Cassimeris et al., 1986) and mouse macrophages (Robinson and Vandre, submitted) in comparison to other cell types.
The potential for rapid microtubule dynamics in other leukocytes, similar to those of monocytes and macrophages, have not been previously examined. Therefore, the sensitivity of microtubules to nocodazole depolymerization was compared in both monocytes and neutrophils to determine if the microtubules in these two cell types shared similar dynamic properties. Cells were treated for various length of time with either 13 /uM nocodazole (monocytes) or 1.3 /u.M nocodazole (neutrophils) and then processed for indirect immunofluorescence microscopy of microtubules (Fig. 25).
The microtubules in each cell type were extremely dynamic, with the vast majority of microtubules having depolymerized within 15-30 seconds of treatment (Fig. 25, compare panels A and B with panels C and D respectively). The model of dynamic instability of microtubules in leukocytes is supported by the observation that most microtubules have depolymerized completely while others remain unchanged at the early times of nocodazole treatment (Fig. 25, panels C and D). Within minutes of exposure to nocodazole, virtually all of the microtubules had depolymerized in both cell types (Fig. 25, panels E and F). Leukocyte microtubules were also sensitive to cold treatment and showed similar depolymerization rates (data not presented).
The microtubules of neutrophils appear to be more sensitive to nocodazole
158 treatment than those of monocytes as indicated by the more rapid loss of microtubules and the greater sensitivity to lower concentrations of nocodazole.
Based upon these results, the estimated half life of microtubules in human neutrophils is between 15 and 30 seconds, which is slightly less than that of human monocytes. A more accurate determination of the half life of microtubules in neutrophils was not attempted due to rapidity of the process.
Microtubule regrowth was also compared in monocytes and neutrophils following removal of the nocodazole (Fig. 26). Cells were treated with 1.3 nocodazole for 10 minutes to ensure that microtubules were depolymerized, at which time the nocodazole was washed out and the microtubules allowed to regrow.
Regrowth was initiated within 1 minute and was restricted to nucleation of microtubules from the centrosome (Fig. 26, panels A and B). Within three minutes an elaborate array of microtubules was established, and complete regrowth was achieved between 5 and 10 minutes. Interestingly, in a significant portion of the neutrophils the initial stages of microtubule regrowth occurred at two separate nucleation sites (Fig. 26, panels B and D). This suggested that splitting and separation of the centrioles resulted from the nocodazole induced microtubule depoiymerization and/or the accompanying regrowth following release of the nocodazole. This was a transient event since very few cells were observed with two microtubule asters following complete recovery of the microtubule array (Fig. 26, panel F). In addition, it was frequently noted that the staining intensity of the
159 regrown microtubules was significantly greater than microtubule staining in untreated neutrophils {see below).
The microtubule network is rapidly rearranged following stimulation of neutrophils.
Neutrophils demonstrate the capacity to orient and migrate in response to a variety of extracellular chemotactic stimuli. Microtubule reorganization has been thought to be associated with these changes in cell behavior. We examined the reorganization of neutrophil microtubules by indirect immunofluorescence following the treatment of cells with the chemotactic agent fMLP (Fig. 27). Cells were first allowed to attach to coverslips prior to treatment, and these randomly migrating cells served as the control (Fig. 27, panel A). Within 1 minute of the application of fMLP to the culture medium cells displayed a more oriented morphology. Microtubules within these cells were reoriented along the long anterior-posterior axis of the cells, and the centrosome was typically located between the anterior or leading edge of the cell and the nucleus (Fig. 27, panel B). Following extended exposure to the chemotactic agent, cells lost their oriented morphology assuming a more flattened appearance. Microtubules were then more randomly oriented extending from a centrally located centrosome to the edges of the cell (Fig. 27, panel C).
Interestingly, the staining intensity of the microtubules also increased following fMLP stimulation (Fig. 27, compare panels B and C with panel A). While the direct cause
160 for this increased staining has not been determined, it suggests that there may be
alterations in the protein components associated with neutrophil microtubules
following stimulation such that these proteins are not as intimately associated with
the microtubules, thus allowing for increased primary antibody binding. Although there were differences in the staining intensity of the microtubules between
stimulated and unstimulated cells we observed only a slight increase in microtubule
numbers in the stimulated cells (Table 4). Preliminary studies also indicate that
there is little or no change in the turnover rate of microtubules between stimulated
and unstimulated cells (data not presented).
Microtubule bundle formation is induced in the leukocytes of patients
undergoing taxol chemotherapy.
Taxol is a drug that induces the stabilization of microtubules, and ultimately causes the formation of bundles of microtubules. Taxol exerts a direct effect on
tubulin since microtubules can be stabilized in cell extracts as well as in intact cells.
This compound is currently undergoing clinical trials as a chemotherapeutic agent,
and has been shown to have efficacy against certain tumors including ovarian
cancers (Rowinsky et al., 1990). One of the major side effects of taxol
chemotherapy is a severe neutropenia that develops several days following the
administration of the drug. With this in mind, we employed our protocol that allows
for the examination of microtubule staining patterns in leukocytes by indirect
161 immunofluorescence. We examined leukocytes isolated from patients undergoing taxol chemotherapy in an effort to determine whether the microtubules of these cells were affected by this treatment. Monocytes and neutrophils were isolated from blood samples obtained from ovarian cancer patients prior to, immediately following, and 48 hours after the completion of a 24 hour infusion of taxol (170-200 mg/m2).
Microtubule staining patterns in control cells were typical of those observed from normal healthy donors (data not presented). However, following a 24 hour infusion, taxol induced microtubule bundles were observed in both monocytes and neutrophils (Fig. 28). In monocytes, a single microtubule bundle was typically located in a position curving around the nucleus (Fig. 28, panel A), but multiple bundles were also present in some cells (data not presented). A single microtubule bundle was also present in neutrophils, and this bundle usually extended between two of the nuclear lobes (Fig. 28, panel B). One end of these microtubule bundles remained associated with the centrosome. The taxol induced bundling of leukocyte microtubules was transient since a normal microtubule array was observed in cells isolated 48 hours following the completion of the taxol infusion (data not presented).
162 DISCUSSION
The microtubule arrays of human leukocytes have been shown to be difficult to examine using conventional immunocytochemical fixation protocols, and for unknown reasons neutrophil microtubules in particular are more difficult to visualize by indirect immunofluorescence in comparison to the microtubules of other leukocytes (Schliwa et al., 1982). None of the previously reported fixation protocols used to examine neutrophil microtubules have been widely accepted as evidenced by a lack of their routine use. We have been able to consistently localize the centrosomal region in neutrophils using one of these previously published protocols
(Chiplonkar et al., 1992), but we have had difficulty in detecting the entire set of neutrophil microtubules. In the previous reports examining the microtubule cytoskeleton of neutrophils, the fixation protocols have relied upon multiple rinses in microtubule stabilization buffers either before and/or after cell lysis (Anderson et al., 1982; Rothwell et al., 1993a; Rothwell et al., 1993b). Thus, the potential existed for the depolymerization of extremely dynamic microtubules prior to cell fixation.
However, even the inclusion of taxol, a microtubule stabilizing agent that should have prevented the depolymerization of microtubules prior to cell fixation, in the rinse or lysis buffers did not improve microtubule staining in neutrophils (Chiplonkar and Robinson, unpublished observations). Therefore, while microtubule depolymerization may have contributed to the lack of staining, this potential
163 microtubule loss was not solely responsible for the lack of immunostaining of neutrophil microtubules.
After a systematic evaluation of a number of different fixation protocols designed for the indirect immunofluorescence staining of microtubules in a wide variety of different cell types, it became apparent that antigenic sites present on neutrophil microtubules were either destroyed or masked in some unknown fashion that precluded binding of primary anti-tubulin antibodies. Loss of antigenicity of proteins associated with the cytoskeleton has been reported as a result of either destruction of the antigen by fixation or stearic hindrance of antibody access to the antigen due to cross-linking of associated proteins (Bell et al., 1987; Peranen et al.,
1993). Restoration of antigenicity was achieved in these cases by either high salt extraction prior to fixation, or guanidine hydrochloride denaturation following fixation.
While these approaches improved the immunofluorescence staining of neutrophil microtubules the morphology of the cells suffered. Soltys and Gupta (Soltys and
Gupta, 1992) reported that SDS extraction was capable of rendering some antigens lost after glutaraldehyde fixation immunoreactive. We found that this approach,
SDS lysis/extraction after glutaraldehyde fixation, not only allowed for microtubule staining in neutrophils and all other leukocytes we have examined, but preserved the morphology of the cells as well. Using this new approach we have reevaluated the microtubule cytoskeleton of human leukocytes. We have shown that previous estimates of microtubule numbers in human monocytes may have underestimated
164 the actual number by nearly 50%. We determined the number and appearance of the microtubule cytoskeleton in human eosinophils for the first time. Also, based upon our experience with human neutrophils, the current SDS extraction/lysis protocol is the most reliable and consistent fixation protocol reported for the examination of neutrophil microtubules. It appears that the difficulties previously encountered with neutrophil microtubule staining were in part due to a masking of the antigenic sites. While the mechanism for restoration of antigenic sites by SDS has not been defined (Soltys and Gupta, 1992), it is probably a result of partial extraction and/or denaturation of proteins associated with the microtubules that may have sterically prevented antibody binding.
Leukocytes are motile cells that can undergo rapid shape changes in response to extracellular stimuli. Reorganization of the microtubule cytoskeleton and repositioning of the centrosome accompanies some of these responses including:binding of lymphocytes to target cells (Geiger et al., 1982; Knox et al.,
1993), binding of neutrophils to antigen-antibody complexes (Chiplonkar et al.,
1992), directed motility in macrophages (Nemere et al., 1985), and directed motility in neutrophils (Anderson et al., 1982; Schliwa et al., 1982). These results have suggested that there is a rapid turnover of microtubules in leukocytes in response to the activation state or stimulatory response of the cells. In addition, the tyrosination state of a-tubulin in neutrophils has been reported to change rapidly in response to cell stimulation (Rothwell et al., 1993b). The detyrosination/tyrosination
165 post-translational modifications of tubulin would also require rapid microtubule dynamics in neutrophils since the tubulin-specific carboxypeptidase (the enzyme responsible for removing the carboxyl terminal tyrosine residue of a- tubulin) is only active against the microtubule polymer and the tubulin tyrosine ligase (the enzyme responsible for retyrosination of a-tubulin) is only active against the tubulin dimer
(Bulinski and Gunderson, 1991). While rapid microtubule turnover rates have been observed in human monocytes (Cassimeris et al., 1986), many of the microtubules were probably lost during lysis of the cells prior to fixation as indicated by comparing the total number of microtubules reported in these earlier results with that obtained using the current SDS fixation protocol. Thus, the results presented here document for the first time that microtubules of normal human neutrophils are extremely dynamic, and appear to be more dynamic than even those found in monocytes.
These rapid dynamic properties could provide the necessary mechanism with which the leukocyte microtubule based cytoskeleton is reorganized in response to cellular stimulation. Furthermore, it should be noted that, while all microtubule systems are dynamic, the half-life of microtubules in most non-dividing mammalian cells is 5-20 minutes compared to the approximately 30 seconds in neutrophils and monocytes.
The regulation of microtubule dynamics in leukocytes, therefore, may prove to be uniquely designed to participate in the functional activities of these cells.
Microtubule reorganization accompanies the directed motility or chemotaxis of neutrophils in response to fMLP (Anderson et al., 1982; Schliwa et al., 1982).
166 Associated with this reorganization is the repositioning of the centrosome. Previous reports are conflicting regarding the position of the centrosome relative to the cell nucleus in these polarized cells. Using whole mount electron microscopy of detergent lysed neutrophil samples Schliwa et al (Schliwa et al,, 1982) reported a repositioning of the centrosome between the anterior lamellipod and the nucleus, whereas the previous immunofluorescence staining of neutrophil microtubules by
Anderson et al (Anderson et al., 1982) indicate a position posterior to the nucleus.
The results we have obtained are consistent with a position of the centrosome anterior to the nucleus in migrating neutrophils. In addition, we show that there is an increased staining intensity of the microtubules in fMLP stimulated cells. One possible explanation for this enhanced fluorescence is that following stimulation, as the microtubules are rapidly reorganized, there are concomitant alterations in the conformation or binding affinity of proteins, such as microtubule associated proteins, which may have been responsible for masking the tubulin antigenic sites in unstimulated cells. Leukocyte microtubuJe associated proteins have not been well characterized, but at least 11 different potential microtubule associated proteins have been reported by two-dimensional electrophoresis in resting T-lymphocytes
(Anand and Chou, 1992). Therefore, it is quite possible that different leukocytes contain cell-type specific microtubule associated proteins with unique functional properties. Alternatively, direct post-translational modification of tubulin may account for some of the increased staining. For example, rapid changes in the tyrosination state (Rothwell et al., 1993b) or increases in the
167 tyrosine-phosphorylation state (Katagiri et al., 1993) of tubulin may account for conformation changes in the microtubules allowing for increased antigen accessibility. We also observed increased staining of microtubules following their regrowth in neutrophils and monocytes recovering from nocodazole induced depolymerization. Disruption of microtubules has been shown to induce both polarity and locomotion in neutrophils (Keller and Niggli, 1993), and nocodazole treatment may, therefore, mimic certain aspects of leukocyte stimulation. In addition, a transient centrosome splitting was often observed in neutrophils undergoing microtubule regrowth, similar to the splitting of centrosomes in neutrophils undergoing activated random locomotion or chemokinesis (Schliwa et al., 1982).
A unique aspect of these studies has been our ability to examine the microtubules in leukocytes obtained from patients undergoing taxol chemotherapy.
Previous reports have shown that taxol treatment induced the formation of microtubule bundles in cultured leukemic cell lines or bone marrow blast cells isolated from leukemia patients, and that bundling correlated with the level of cytotoxicity and cytoreduction of tumor respectively (Rowinsky et al., 1989;
Rowinsky et al., 1988). Microtubule bundling also occurred following taxol treatment of isolated normal human lymphocytes , but the formation of microtubule bundles did not prevent MTOC repositioning (Knox et al., 1993), or target cell binding
(Chuang et al., 1994). The direct effect of taxol on the microtubules of cells isolated
168 from patients undergoing taxol chemotherapy, however, has not been previously determined. The effect of clinically achievable levels of taxol on the microtubules of circulating leukocytes obtained from patients undergoing taxol infusion is of
interest since neutropenia is one of the major dose-limiting factors related to taxol therapy. We show that microtubules are bundled in both the monocytes and neutrophils isolated from patient blood obtained immediately following the completion of a 24 hour infusion of taxol at doses ranging from 170-200 mg/m2.
This bundling of microtubules was a transient event with recovery of normal microtubule arrays within 48 hours after taxol therapy ceased. The precise time course for the recovery of normal microtubule morphology after a 24 hour infusion has not been determined, nor have we determined whether alterations in microtubule dynamics occur prior to the bundling or whether the bundling and subsequent recovery of leukocyte microtubule arrays are affected by the time course of taxol infusion (e.g. 1, 3, or 24 hr). The mechanism by which taxol affects tumor cells may not be restricted to its direct stabilization of microtubules in these cells. Recently, taxol has been shown to mimic the effects of lipopolysaccharides on macrophage cytokine production and has been shown to stimulate tumoricidal activity of macrophages (Ding et al., 1993; Manie et al., 1993; Manthey et al., 1994).
Thus, taxol modulation of microtubule dynamics in leukocytes may be an important factor in activating cytokine production and the immune response. The ability to visualize leukocyte microtubules, as afforded by the immunocytochemical staining protocol reported here, should also be useful in correlating changes in gene
169 expression as a result of taxol treatment with the modulation of microtubule dynamics in these cells.
170 ACKNOWLEDGEMENTS
This work was supported by NSF grant DCB-9096261 to DDV, Ohio State
University Comprehensive Cancer Center Core grant CA 16058 to BCB, and The
Council for Tobacco Research grant 2065 to JMR.
171 REFERENCES
Anand, B., I.N. Chou: Microtubules and microtubule-associated proteins in resting and mitogenically activated normal human peripheral blood T cells. J. Biol. Chem. 267, 10716-10722 (1992).
Anand, B., I.N. Chou: Microtubule network and microtubule-associated proteins in leukemic T lymphocytes. Leukemia7, 51-57 (1993).
Anderson, D.C., L.J. Wible, B.J. Hughes, C.W. Smith, B.R. Brinkley: Cytoplasmic microtubules in polymorphonuclear leukocytes: effects of chemotactic stimulation and colchicine. Cell 31, 719-729 (1982).
Bell, P.B., I. Rundquist, I. Svensson, V.P. Collins: Formaldehyde sensitivity of a GFAP epitope, removed by extraction of the cytoskeleton with high salt. J. Histochem. Cytochem. 35, 1375-1380 (1987).
Bulinski, J.C., G.G. Gunderson: Stabilization and posttranslational modification of microtubules during cellular morphogenesis. BioEssays 13, 285-293 (1991).
Cassimeris, L.U., P. Wadsworth, E.D. Salmon: Dynamics of microtubule depolymerization in monocytes. J. Cell Biol. 102, 2023-2032 (1986).
Chiplonkar, J.M., D.D. Vandre, J.M. Robinson: Stimulus-dependent relocation of the microtubule organizing center in human polymorphonuclear leukocytes. J. Cell Sci. 102, 723-728 (1992).
Chuang, L.T., E. Lotzova, J. Heath, K.R. Cook, A. Munkarah, M. Morris, J.T. Wharton: Alteration of lymphocyte microtubule assembly, cytotoxicity, and activation by the anticancer drug taxol. Cancer Res. 54,1286-1291 (1994).
Ding, A., E. Sanchez, C.F. Nathan: Taxol shares the ability of bacterial lipopolysaccharide to induce tyrosine phosphorylation of microtubule-associated protein kinase. J. Immunol. 151, 5596-5602 (1993).
Ding, M., J.M. Robinson, R.W. Burry, D.D. Vandre: Human phagocytic leukocytes have an extremely dynamic microtubule array. Mol. Biol. Cell4, 451a (1993).
Geiger, B., D. Rosen, G. Berke: Spatial relationships of microtubule-organizing centers and the contact area of cytotoxic T lymphocytes and target cells. J. Cell Biol. 95, 137-143 (1982).
172 Hoffstein, S., G. Weissman: Microfilaments and microtubules in calcium ionophore-induced secretion of lysosomal enzymes from human polymorphonuclear leukocytes. J. Cell Biol. 78, 769-781 (1978).
Katagiri, K., T. Katagiri, K. Kajiyama, T. Yamamoto, T. Yoshida: Tyrosine-phosphorylation of tubulin during monocytic differentiation of HL-60 cells. J. Immunol. 150, 585-593 (1993).
Keller, H.U., V. Niggli: Colchicine-induced stimulation of PMN motility related to cytoskeletal changes in actin, alpha-actinin, and myosin. Cell Motil. Cytoskel. 25, 10-18 (1993).
Knox, J.D., R.E.J. Mitchel, D.L. Brown: Effects of taxol and taxol/hypothermia treatments on the functional polarization of cytotoxic T lymphocytes. Cell Motil. Cytoskel. 24, 129-138 (1993).
Koonce, M.P., R.A. Cloney, M.W. Berns: Laser irradiation of centrosomes in newt eosinophils: evidence of centriole role in motility. J. Cell Biol. 98, 1999-2010 (1984).
Leung, M.-F., J.A. Sokoloski, A.C. Sartorelli: Changes in microtubules, microtubule-associated proteins, and intermediate filaments during the differentiation of HL-60 leukemia cells. Cancer Res. 52, 949-954 (1992).
Manie, S., A. Schmid-Alliana, J. Kubar, B. Ferrua, B. Rossi: Disruption of microtubule network in human monocytes induces expression of interleukin-1 but not that of interleukin-6 nor tumor necrosis factor-alpha. J. Biol. Chem. 268, 13675-13681 (1993).
Manthey, C.L., P.-Y. Perera, C.A. Salkowski, S.N. Vogel: Taxol provides a second signal for murine macrophage tumoricidal activity. J. Immunol. 152,825-831 (1994).
Mollinedo, F., J.M. Nieto, J.M. Andreu: Cytoplasmic microtubules in human neutrophil degranulation: reversible inhibition by the colchicine analogue 2-methoxy-5-(2',3‘,4,-trimethoxyphenyl)-2,4,6-cycloheptatrien-1-one. Mol. Pharmacol. 36, 547-555 (1989).
Nemere, I., A. Kupfer, S.J. Singer: Reorientation of the Golgi apparatus and the microtubule-organizing center inside macrophages subjected to a chemotactic gradient. Cell Motil. 5, 17-29 (1985).
Peranen, J., M. Rikkonen, L. Kaariainen: A method for exposing hidden antigenic sites in paraformaldehyde-fixed cultured cells, applied to initially unreactive antibodies. J. Histochem. Cytochem. 41, 447-454 (1993).
173 Robinson, J.M., Z. Luo: The association of microtubules with an acidic phosphatase-containing tubular compartment in macrophages. Acta Histochem. Cytochem. 25, 223-229 (1992).
Rothwell, S.W., C.C. Deal, J. Pinto, D.G. Wright: Affinity purification and subcellular localization of kinesin in human neutrophils. J. Leuk. Biol. 53, 372-380 (1993a).
Rothwell, S.W., J. Nath, D.G. Wright: Rapid and reversible tubulin tyrosination in human neutrophils stimulated by the chemotactic peptide, fMet-Leu-Phe. J. Cell. Physiol. 154, 582-592 (1993b).
Rowinsky, E.K., P.J. Burke, J.E. Karp, R.W. Tucker, D.S. Ettinger, R.C. Donehower: Phase I and pharmacodynamic study of taxol in refractory acute leukemias. Cancer Res. 49, 4640-4647 (1989).
Rowinsky, E.K., L.A, Cazenave.R.C. Donehower: Taxol: a novel investigational antimicrotubule agent. J. Natl. Cancer Inst. 82, 1247-1259 (1990).
Rowinsky, E.K., R.C. Donehower, R.J. Jones, R.W. Tucker: Microtubule changes and cytotoxicity in leukemic cell lines treated with taxol. Cancer Res. 48,4093-4100 (1988).
Schliwa, M.: Microtubules. In: M. Alfert, W. Beerman, L. Goldstein (eds.): Cytoskeleton an Introductory Survey, pp 47-81. Springer Verlag, New York 1986.
Schliwa, M., K.B. Pryzwansky, U. Euteneuer: Centrosome splitting in neutrophils: an unusual phenomenon related to cell activation and motility. Ce//31, 705-717 (1982).
Soltys, B.J., R.S. Gupta: Interrelationships of endoplasmic reticulum, mitochondria, intermediate filaments, and microtubules - a quadruple fluorescence labeling study. Biochem. Cell Biol. 70, 1174-1186 (1992).
Swanson, E., A. Bushnell, S.C. Silverstein: Tubular lysosome morphology and distribution within macrophages depend on the integrity of cytoplasmic microtubules. Proc. Natl. Acad. Sci. USA 84, 1921-1925 (1987).
Swanson, J.A., A. Locke, P. Ansel, P.J. Hollenbeck: Radial movement of lysosomes along microtubules in permeabilized macrophages. J. Cell Sci. 103,201 -209 (1992).
174 Fixative Microtubule Staining
Monocytes Neutrophils
1. -20°C Methanol 6 min +
2. 3.7% paraformaldehyde 15 min - 0.2% Triton 15 min +
3. 0.7% glutaraldehyde 15 min - 0.2% Triton X-100 15 min ++
4. 0.7% glutaraldehyde + 0.2% Triton X-100 15 min ++
5. 0.2% Triton X-100 90 sec - 0.7% glutaraldehyde 15 min +
6. 0.5% Triton X-100 90 sec - multiple rinses in 4% polyethelene glycol - 3% paraformaldhyde + 1% DMSO + +
7. 0.5% Triton X-100 2-8 min - 2% paraformaldehyde + 0.1% glutaraldehyde 10 min - -20°C methanol 6 min - -20°C acetone 1 min + +
8. 0.7% glutaraldehyde 15 min - 0.5% SDS 15 min +++ ++
TABLE 3 Fixation Protocols
175 Monocyte Neutrophil Eosinophil
Number observed: A. 35.4+7.7 32.4+4.5 (unstim.) previous reports1 35.1±5.0 (fMLP stim.)
B. 32.0+4.0 (unstim.) - 40.0±5.0 (fMLP stim.)
C. 22.3+2.0 (unstim.) -
Number observed: 80.0+6.5 36.2+4.0 (unstim.) 28.0+4.6 this report3 41.8+4.6 (fMLP stim.)
1 Previous reports refers to the microtubule numbers reported by:A) indirect immunofluorescence staining of monocytes (Cassimeris et al., 1986) and neutrophils (Anderson et al., 1982); B) electron microscopy whole mount preparations of neutrophils (Schliwa et al., 1982); and C) electron microscopy of thin sections of neutrophils (Hoffstein and Weissman, 1978). immunofluorescence staining of microtubules in newt eosinophils has been reported, but numbers were not determined (Koonce et al., 1984). 3The microtubules in a minimum of 15 cells obtained from four donors were counted.
TABLE 4 Microtubule Numbers in Human Leukocytes
176 Fig. 22. Comparison of microtubule staining patterns obtained in human leukocytes using different fixation protocols.
Indirect immunofluorescence staining of microtubules in monocytes (A and
B) and neutrophils (C and D) are compared. An incomplete microtubule array is detected in monocytes processed using fixation #5, Table 3(A), whereas a much more extensive microtubule array is observed using fixation #8, Table 3 (B). Only a few neutrophil microtubules are detected near the nucleus in cells processed using fixation #7, Table 3 (C). Again, a more extensive microtubule array is observed in neutrophils processed using fixation #8, Table 3 (D). Due to the autofluorescence surrounding the nuclear region, some microtubules cannot be traced back to the centrosome in the neutrophil samples. All samples were incubated with clone Tub1A2 tubulin monoclonal antibody, which gave the best staining in samples that were not lysed/extracted with SDS. Bar = 5 /^m.
177 FIGURE 22
178 Fig. 23. Microtubule staining obtained in different human leukocytes.
Phase contrast and immunofluorescence of the same cells: monocyte (A and
B), neutrophil (C and D), and eosinophil (E and F) processed for staining using the
SDS lysis/extraction staining protocol (fixation #8, Table 3). The morphology of the leukocytes is maintained using this fixation protocol, and complete microtubule arrays can be detected. This fixation protocol was utilized to obtain the microtubule staining presented in all subsequent Figures. Bar = 5 ^m.
179 F
FIGURE 2 3
180 Fig. 24. Representative field of microtubule staining in a mixed population of leukocytes.
Microtubules were stained using a mixture of DM1A and DM1B monoclonal anti-tubulin antibodies. This survey micrograph illustrates labeling of platelets
(arrowheads), a lymphocyte (small arrowhead), and a monocyte (large arrowhead).
In addition, eight neutrophils (large unlabeled cells) are also present in this field.
This shows that the SDSIysis/extraction protocol provides suitable microtubule labeling of leukocytes. Bar =10 ^m.
181 FIGURE 24 Fig. 25. Microtubule turnover in monocytes and neutrophils.
Monocytes (A, C, and E) were treated with 13 ,uM nocodazole, while neutrophils (B, D, and F) were treated with 1.3 /uM nocodazole. Panels A and B, represent untreated control cells; panel C, cells treated with nocodazole for 30 seconds; panel D, cells treated with nocodazole for 15 seconds; and panels E and
F, cells treated with nocodazole for 5 minutes. There is both a rapid and essentially complete depolymerization of microtubules in monocytes and neutrophils following exposure to nocodazole. Bar = 5 ^m.
183 FIGURE 25 Fig. 26. Regrowth of microtubules in monocytes and neutrophils following release from nocodazole.
Monocytes (A, C, and E) and neutrophils (B, D, and F) were treated with 1.3
/^M nocodazole for 10 minutes to ensure that most microtubules were depolymerized. The nocodazole was washed-out and microtubule regrowth was allowed for 1 min (A and B), 3 min (C and D), or 5 min {E and F). Microtubule regrowth was evident within 1 min from the time of nocodazole release and was complete within 5 min. Many neutrophils showed two foci of microtubule nucleation during early periods of regrowth (B and D). Anti-tubulin antibodies DM1A and
DM1B were used as the primary antibodies. Bar = 5 ^m.
185 %
FIGURE 26 Fig. 27. Reorganization of microtubules in neutrophils following stimulation with the chemotactic peptide fMLP.
Neutrophils were allowed to attach to the coverslips for 12 min prior to the addition of fMLP (10'7 M) for 0, 1, or 15 min (A, B, and C respectively). Microtubules rapidly reoriented along the anterior-posterior axis of the migrating cell, with the centrosome located between the leading edge of the cell and the nucleus (B). A
random microtubule orientation was observed following longer exposure to fMLP when cell no longer exhibited directed migration (C).Each sample was treated in an identical manner, including the recording of the photomicrographs and printing on the figures. Interestingly, the intensity of the microtubule staining increased following stimulation with fMLP. Bar = 5 ^m.
187 F I G U R P 2 7
1 8 8 Fig. 28. Taxol induced microtubule bundles are present in the leukocytes of patients undergoing taxol chemotherapy.
Monocytes (A) and neutrophils (B) were isolated from patient blood samples immediately following a 24 hr infusion of taxol (200 mg/m2). Microtubules bundles were observed by indirect immunofluorescence in both monocytes and neutrophils isolated from this sample. Bar = 5 ^m.
189 FIGURE 2 8