IIIHHIHII WI 1 7 115 977 THS ON THE {SQLATION C1? SASAL 533213 FROM TETEQAHYMEHA ?\“RE?©RMIS The“: For Hm 9'3qu 05 M. 5. MICHEGRN STATE UNWERSITY Elwood A. Manes! 11967 LIBRARY “mus Mlchigan State University ABSTRACT ON THE ISOLATION OF BASAL BODIES FROM TETRAHYMENA PYRIFORMIS by Elwood A. Linney The problem of isolating basal bodies (kinetosomes) from the ciliate protozoan Tetrahymena pyriformis was examined. The structural relationships between the three components of the kinetid were described and the possible similarity between the fibrils composing these three com— ponents and those found in mitotic spindles, tobacco mosaic virus, and other biological structures were mentioned. The isolation technique employed iSOpycnic ultra- centrifugation with electron microscopic identification of the isolated fractions. Parameter changes examined qualita— tively for their effect upon the resultant isolation in- cluded: (1) pH of isolation medium; (2) EDTA concentration; (3) method of adding material to ultracentrifugal gradient; (4) length of ultracentrifugal run; (5) amount of material added to the gradient; (6) ultracentrifugal sucrose density range; and (7) temperature of material preparation. The resultant ultracentrifugal separation allowed discrimination of the material into four bands--one mito— chondrial band and three other bands all apparently similar Elwood A. Linney in structure appearing as membranous vesicles in electron microscopic observations. No basal bodies were identified in any of the ultracentrifugally separated fractions. The presence or lack of EDTA in the homogenization medium did not appear to affect structures observed after the homogenate was prepared for electron microsc0py. Basal bodies were positively identified in only two cases: (1) the pellet fraction obtained from slow centrifugation immediately after homogenization; and (2) homogenized material prepared at roomtemperature. In both cases of observation the basal bodies were observed either aligned in rows as observed in situ or had membrane or other material still attached to them. From these results it is suggested that the basal bodies come apart under the conditions used if they are not still attached to other material. Suggestions for further work on isolation are made: (1) separation in deuterium oxide (D20) for possible stabili- zation of the basal body structure; (2) prefixing cells be- fore separation of parts; (3) examining the chemical com- position, structure, and bonds of peripheral ciliary doublets for possible information concerning basal body fibrils. ON THE ISOLATION OF BASAL BODIES FROM TETRAHYMENA PYRIFORMIS BY Elwood A. Linney A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of BiOphysics 1967 ACKNOWLEDGMENTS I wish to thank the members of my thesis committee: Dr. R. Neal Band, Dr. John I. Johnson, and Dr. Barnett Rosenberg for the aid given me while I was present at Michigan State University. As my major professor I wish particularly to thank Dr. Johnson for encouraging the development of my own in- terests. As the "unofficial" advisor of my research program I thank Dr. R. Neal Band who most generously contributed his advice, time, and technical skills to a student from a "foreign" department. ii TABLE OF CONTENTS Page ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . ii LIST OF FIGURES . . . . . . . . . . . . . . . . . . iv I. INTRODUCTION . . . . . . . . . . . . . . . . 1 II. THE KINETID . . . . . . . . . . . . . . . . . 4 A. Cilium . . . . . . . . . . . . . . . . . . 5 B. Kinetosome . . . . . . . . . . . . . . . . 7 C. Fibrils . . . . . . . . . . . . . . . . . 9 III. MICROTUBULES . . . . . . . . . . . . . . . . 10 IV. THE KINETID AS A DEVELOPING UNIT . . . . . . 14 V. INTENT OF WORK . . . . . . . . . . . . . . . 18 VI. METHODS . . . . . . . . . . . . . . . . . . . 22 A. Culturing . . . . . . . . . . . . . . . . 22 B. Separation . . . . . . . . . . . . . . . . 22 C. Electron Microscopy . . . . . . . . . . . 24 VII. RESULTS . . . . . . . . . . . . . . . . . . . 27 VIII. DISCUSSION . . . . . . . . . . . . . . . . . 34 IX. BIBLIOGRAPHY . . . . . . . . . . . . . . . . 38 X. FIGURES O O O O O O O O O O O O O O O O C O O 42 iii LIST OF FIGURES Isolation method . . . . . . . . . . . . . Description of material in various fraction Membranous vesicles . . . . . . . . . . . . Longitudinal section of Cilium . . . . . . Cross section of basal bodies . . . . . . . Longitudinal section of basal bodies . . . Sketch of Cilium cross section . . . . . . Sketch of basal body cross section . . . . Sketch of ciliary peripheral doublet . . . Sketch of flagellar fibril . . . . . . . . iv Page 42 43 44 44 46 46 48 48 48 48 I . INTRODUCTION Correlative study of cellular fine structure and function has been given added impetus by the recent findings of DNA in some cellular organelles. The question of the means of develOpment of organelles has been studied by several investigators with little in the way of answers towards questions such as membrane development, mitochon— drial development (27), etc. DNA has now been found in both mitochondria (49) and chlorOplast (8,47), extracted (28), partially physically and chemically characterized (28,34,36), and shown to take up radioactive RNA precursors (28) such taking up has also been shown to be inhibited by the addition of actinomycin D). The DNA structure is circular--similar to that found in bacteria (35,53). Such similarities with bacterial DNA has lent credence to the old speculation that_mitochondria and chloroplasts might at one time have been parasites which have evolved and lost their autonomy (26). The present investigation began with an interest in such problems as possible function of organellar DNA and proceeded to the more physical problem of isolating an organelle suggested by some investigators to have DNA-— the kinetosome (basal body) of ciliate protozoans. The basal body is the base for the cilia of ciliate pro— tozoan; characteristically the cilia are aligned in rows, called kineties (5), along the cellular surface. The number and arrangement of such structures is constant for a given cellular cycle stage and is species specific (39). When a ciliate divides to form two daughter cells the cilia and kinetosomes double in number before the division occurs (42). Such develOpment, accompanied by complex fibrillar develOpment just beneath the surface of the cell, and intimately associated with the kineto- somes, is still not understood. The extent of this as- sociation will be described after some words of caution. The following descriptions will take advantage of published work of electron microsc0py of cellular fine structure. The use of the electron microsc0pe for the study of cellular fine structure appears to be increasing but it can still be said that the results obtained are highly dependent on the art of the investigator and not necessarily upon his scientific rationale. The two most common techniques employed are the thin sectioning technique whereby the material is fixed, embedded in a resin, sec- tioned, and then perhaps stained with a heavy metal solu- tion before observation with the electron micrOSCOpe; while the second technique involves the addition of a heavy metal stain to the material and then allowing the solution to dry around the material as a glass (thus the electron microscopist will "see" the material by the absence of the stain). Both techniques have limitations, the principal one in both cases is the necessity of killing the organism. Also, for fine sectioning technique there are still ques- tions regarding where the most common fixative, osmium tetroxide, reacts (22,23). Fixing methods usually involve fixing at 0-4°C, though current studies indicate that some cellular components disappear at this temperature (2,3,4, 21,46,57). With such precautions immediately mentioned and considered, the following account of fine structure can be described without the continual addition of "assuming this is the in vivo structure." The development and function of organelles may be intimately bound to their composition, particularly in the light of the previously mentioned findings of DNA in chloro- plast and mitochondria. That is, if such DNA proves to be functional DNA such function might encompass both develOp- ment and function known and unknown. The object of this study was to isolate a cellular organelle from the ciliate protozoan Tetrahymena pyriformis. This organism was chosen after preliminary autoradiographic studies with the ciliate Stentor Coeruleus revealed many difficulties in the culturing of organisms upon other living organisms. With Tetrahymena pyriformis culturing is pos- sible upon only proteose peptone and water. The organelle studied was the kinetosuomeaor that component of the larger unit called the kinetid (5) which consists of the Cilium, the kinetosome, and fibrillar structures lying close to the kinetosome. II . THE KINETID The kinetid is a composite of three structures found near the cortex (pellicle, cell surface) of ciliate proto- zoans--it is composed of the Cilium, that structure which normally functions to prOpel the cell through its liquid environment; the basal body, the base structure of the Cilium lying beneath the surface of the cell; and fibrils-- in Tetrahymena pyriformis 4 types of fibrils have been iden- tified which run below the surface of the cell in close proximity to the basal bodies (39). The kinetids in Tetra- hymena are arranged in 18 longitudinal rows (12) along the surface of the cell--each row being defined a kinety by Chatton (24). Daughter cells are produced by the growth of a single cell with accompanying doubling of the number of kinetids before the cell divides into two daughter cells-- one daughter cell having the mother's anterior end and the other having the mother's posterior end. The kinetid's components have been studied by various investigators, many limiting their work to just one of its three components. Considering this, each will be treated separately below. A. Cilium The Cilium has been the most extensively studied of the three kinetid components-~the most probable reason being the ease at which it can be isolated in pure quantities-- either by agitation and differential centrifugation to re- move cell bodies (56) or by the addition of CaCl2 in suit- able quantities (58) to the media which causes the cilia to separate from the cell bodies. Cilia and flagella of most cells are approximately 0.25 microns in diameter (17) with varying lengths depend— ing upon the species (see Figures 7 and 9). The length of Tetrahymena pyriformis cilia is approximately 8 microns (58). Electron microscopic examination of cilia and flagella have shown a characteristic cross sectional structure of a mem- brane surrounding a fibrillar core called an axoneme (14). In Tetrahymena pyriformis this membrane is continuous with the outer cellular membrane (39). The axoneme cross section shows the characteristic structure of 9 peripheral doublet fibrils plus two central fibrils surrounded by a sheath (see Figure 4). The two central fibrils are approximately 240A in diameter and 360A apart-~they, as are the peripheral fibrils, are apparently tubular in nature. Each peripheral doublet forms a dense figure of eight (in cross section) with their common wall shared--their tubular wall is approxi- mately 45A in thickness and the doublet is approximately 370A by 250A in cross section. On one of each peripheral doublet fibril (called the A fibril) are two projections seen in cross section to be 150A in length and 50A thick. Longi- tudinally they are about 70A long and spaced 170 to 220A apart (1?). Preliminary chemistry of the Cilium has been aided by the finding that once the ciliary membrane is removed the Cilium can be selectively dissected chemically and also partially reconstituted (15). That is, the cilia can be isolated, the membrane extracted with digitonin, and the resultant being the above described axoneme (ciliary membrane can also be obtained by dialyzing cilia against an EDTA solution). Most of the ciliary ATPase was found to be in the axonemal fraction (16) and this ATPase can be obtained in solution by treating the axonemes with 0.6M KCl or by dialyzing them against a che— lating agent at low ionic strength. The unsolubilized fraction consists primarily of peripheral doublets with no projections. The ATPase components were isolated in an analytical ultracentrifuge and checked for ATPase activity (theisolated components were 4Svedberg units, 14 units, and 30 units). The 48 component proved to be relatively inactive while the 148 and 308 were active and appeared to have a monomer-polymer relationship-—later shown to be this in the electron microscope (17). The ATPase component was added back to the projectionless doublets and electron microsc0py indicated that the projections reappeared showing that at least some of the ATPase protein was located at these pro- jections. Follow up work indicated that the relative amounts of amino acids in this ATPase (called dynein) were similar in nature to that found in the muscular ATPase actomyosin (43). Further work with negatively stained fibrils from flagella suggest that the fibrils apparently consist of 10 to 13 beadlike filaments which are proteinaceous in nature and form the tubular wall (19,37) (see Figure 10). Such fine structure will be discussed later in the section on microtubules. B. Kinetosome The basal body of Tetrahymenaypyriformis is roughly cylindrical with a length of approximately 0.4 microns and a diameter of 0.22 microns (41). It consists of nine periph- eral triplet fibrils—-two of the three fibrils apparently continuous with the ciliary peripheral doublet fibrils (40) (see Figure 8). The remaining fibril of each triplet tapers away as the fibrils reach the cell surface and the Cilium (40). Between the Cilium and the basal body is a darkly staining basal plate—-the other end of the kinetosome is apparently open to the cytOplasm of the cell. There is no membrane about the kinetosome but there are fibrils that are both connected and extremely close to it. Because of its presence within the cell, the kineto- some has proven harder to isolate than the Cilium. Seaman (52) has reported a method of isolation using a digitonin solubilization technique originally suggested by Child and Mazia (6) whereby whole cells are pretreated with cold ethanol, solubilized with digitonin, and homogenized. The kinetosomal isolate was obtained by centrifugal techniques from this homogenate and the percentage dry weight was de- termined (52): protein 49.6% DNA 3.1% RNA 2.1% Lipid 5.2% Carbohydrate 6.2% Unfortunately the apparent criterion for purity was a phase micrograph of ungiven magnification; considering the small size of the kinetosome, this does not appear sufficient. Others (1,20,48) have repeated and varied Seaman's technique with accompanying electron microscopic check of the isolate. In all cases the isolate was shown to consist of some kineto- somes with much unidentifiable background material and, or, kinetosomes with fibrils and other material still attached. Randall (42) has investigated the possibility of kinetosomes having their own DNA with the use of acridine orange staining. Using the synchronizing technique of Scherbaum and Zeuthen (50), a culture of Tetrahymena pyri— formis was subjected to a series of heat shocks causing 50 to 80% of the cells to divide at one time. By taking out samples during the synchronization and staining pellicular pieces with the nucleic acid binding dye acridine orange, an increase of acridine orange fluorescence indicative of DNA was found at the basal body sites with a peak in time corresponding to the time at which the cells were synchro- nized to divide--such a correlation between dividing cells and acridine orange fluorescence is the best indication to date of DNA being at least very close to the kinetosome (from comparison of fluorescence of basal bodies with that from T2 phage Randall suggested that the amount of DNA per basal body is approximately the same or 10—16g, however the qualitative nature of the stain suggests that this value is probably not very accurate). Smith—Sonnenborn and Plaut (54) have found similar fluorescence of DNA binding stains from the pellicle of Paramecium aurelia. The specific chemistry and function of the kineto- some most probably awaits its isolation—-however, its cen- tral position in the kinetid indicates that it is probably at least an organizing center for kinetid structures such as the Cilium and the fibrils. C. Fibrils Four different sets of fibrils have been observed via the electron microsc0pe near the pellicle of Tetra- hymenagpyriformis (39): (1) one set is striated in nature arising from the kinetosome and tapering to the right and anterior end of the cell; one arises from each kinetosome 10 and they overlap to form a bundle along each kinety. (2) a set of tubular fibrils running transversely from the left side of each kinetosome and passing to the left closely under the pellicle. (3) a set of tubular fibrils passing posterior from the right posterior edge of each kinetosome. (4) individual overlapping tubular fibrils running longi- tudinally beneath the pellicle and to the right of the kinetosomes of a kinety. All the tubular fibrils are approximately 200A in diameter in thin sections--no connection has been observed between any of the four different sets of fibrils. I I I . MI CROTUBULES Structures with a cylindrical shape found in various cells and having a diameter roughly between 180A and 300A have been referred to as microtubules (3). The dimensions of the fibrils which make up the Cilium, the basal body, and which lie next to such structures are within this range. Mitotic spindle fibrils (46), tubules associated with blood platelets (2,4), tubules within the ax0podia of Actino- sphaerium nucleofilium (57), plus several others lie within this range. When such structures are observed for fine structure with negative staining techniques (4,19,37) it is evident that at least flagellar fibrils (19,37) and blood platelet fibrils (4) are composed of 10-13 beadlike filaments which ll compose the tubular wall of these structures (see Figure 10). The size of the bead has been reported to be 35-50A in di— ameter (4,19,37) though this value may not be accurate con- sidering the techniques used. Optical diffraction examina- tion of electron micrographs of negatively stained flagellar fibrils partially and wholly intact lead to the suggestion that the subunits (beads) are probably transversely bonded along with a longitudinal bond. The subunits are staggered from each other giving the fibril a helical appearing struc- ture. Though one might expect a structural difference be— tween the two fibrils of flagellar peripheral doublets, no difference was observed (a site for the ATPase projections was not apparent) (19). The dynamic prOperties of the mitotic apparatus have allowed several studies to be made upon its associated microtubules regarding the effects of changes in its sur- roundings to their structure. Temperature studies indicate that the mitotic spindle microtubules disappear when the temperature is lowered to 0—4°C in a number of cells (21,46). Such a process appears reversible since the microtubules re- form if the cell is brought back to room temperature (21,46). The presence or absence of divalent cations in the electron microscopic fixative appears to effect what is seen (with- out a specific concentration of divalent cations the micro- tubules associated with the metaphase Spindle of the ameba Pelomyxa carolinensis are not observed in the electron 12 microscope) (45). The addition of 0.8M urea to a culture of Chaos carolinensis whose mitotic microtubules had dis- appeared by cooling treatment inhibited their reformation at room temperature (46). The whole mitotic apparatus be- comes disorganized under pressures as high as 12,000psi (31) though no subsequent electron microsc0pic study of its micro- tubules has been published. Normally birefringence is used (21) as a method to detect the presence of microtubules in living cells--Inoue has found that such birefringence in- creases as the temperature is increaSed after initial cool- ing of cells. Such an observation leads to the question of the nature of the polymerization reaction that forms the microtubules from subunits. Considering this observation alone the reaction must be endothermic (57). Similar observations have been made concerning the polymerization reaction of the protein subunits of tobacco mosaic virus (24) (i.e., the reaction was deterred with cold treatment). The tobacco mosaic virus is tubular with a length of 3000A and a diameter of 180A. Its tubular wall width is approximately 50A. It is considered to be composed of a giant helix of over 2000 protein subunits with RNA bound to them in a specific manner. The structure will form spontaneously under specific conditions, but the polymerization is temperature sensitive in a manner similar to the microtubules of the mitotic apparatus. It was sug- gested that the subunits must release something during 13 polymerization because of this temperature sensitivity and this something was experimentally determined to be water (25). These studieSIare relevant to the microtubular work in that cellular mitotic spindles have been shown to be "frozen" when D20 is substituted for H20 in the medium (30). This would appear to suggest that hydrogen bonds play an important role in microtubular structure since it is known that the D-bond is stronger than the H-bond. This, however, does not suggest the nature of the water-subunit bond if mitotic microtubular subunits are similar to TMV protein subunits. All the following has involved possible similarities between microtubules--a recent study has classified micro- tubules in accordance with their differential reactions to a variety of treatments: (1) fixing at 0°C; (2) fixing at 50°C; (3) negative staining; (4) colchicine treatment; and (5) digestion with pepsin. From the results four classes of microtubules were suggested: (1) cytoplasmic micro- tubules; (2) sperm tail central and accessory fibrils; (3) sperm tail peripheral doublet fibrils that have ATPase projections; and (4) sperm tail peripheral doublet fibrils that do not have ATPase projections. Specifically perti- nent were the findings that the Sperm tail fibrils do not disappear with cold treatment, that pepsin digests all fi- brils eventually but does not apparently digest cellular membrane (indicating a chemical difference in fibrils and 14 cellular membrane), and that there were differential effects along the length of at least three of the four classes of fibrils (3). IV. THE KINETID AS A DEVELOPING UNIT The complex structure of the individual components of the kinetid plus the complexity involved in their ar— rangement with respect to each other has fascinated several investigators--particularly when one questions how such structures might develOp. The central position of the kinetosome in the kinetid has led to several speculations on its develOpment. As the base of the Cilium it is the site from which the Cilium de- velops and either the site or near the site of the fibril development. It has been suggested by Lwoff that kinetosomes di- vide to develop two from one and that they might also have their own genetic material. The divisional process was suggested from light microscopic observations of cells in development (29). Pertinent to this suggestion have been the studies mentioned previously regarding the isolation of basal bodies and subsequent determination of DNA content (l,20,52)--none of these studies have been definitive for reasons mentioned previously. Randall's acridine orange study (42) indicates that DNA is most probably associated at least very closely 15 to the kinetosomes. Ehret and DeHaller (11), in an exten— sive electron microscopic study of cortical develOpment in the ciliate Paramecium bursaria, found no dividing kineto- somes and suggest that they might develop from smaller structures they observed below the surface of the cell. Because of the structural similarity of the kineto- some and the centriole (that organelle which forms the pole of the mitotic spindle and which also has a nine peripheral triplet fine structure) and because some centrioles do form the base for eventual flagella (13), it has been suggested that the basal body may develop in a similar way. Gall (13) has observed that new centrioles form from a smaller pro— centriole found lying near mature centrioles. The pro- centriole is similar to the mature centriole in that it has nine peripheral triplets, but the length is shorter. The procentriole and mature centriole are aligned with their cylindrical axii perpendicular to one another. Mazia et_§l. have suggested a generative model of centriole forma— tion whereby a centriole has a germ or seed within it which directs the growth of a replica similar in manner to the production of new bacteriOphage (33). Another suggestion involves the possible duplication of the triplet fibrils within the kinetosomal structure fol- lowed by a sliding away of one set of triplets resulting in two kinetosomes (41). 16 All of these models are suggestions, some based upon observation, but little is actually known about ki- netosomal development in ciliate protozoans. However, certain questions can be asked to provide the framework for investigation: (1) how does the basal body develOp; (2) what are the relative developments of the fibrils and the Cilium with respect to the basal body; (3) are there structural and chemical similarities between the micro— tubules that make up the Cilium, the kinetosome, and the three tubular fibril sets? Studies of Cilium and flagellus regeneration and growth have been made on Tetrahymena pyriformis (7), a number of flagellate protozoans (44), and the ameba— flagellate Naegleria'gruberi (10). Tetrahymena cilia ap- peared to regenerate fully in 80 minutes after an initial lag period of 30 minutes (7). A Similar but more quanti- tative study with three flagellates studied the growth of flagella with time after the flagella had been removed by various means (44). There was also a lag period of 30 minutes followed by regeneration at a decreasing rate taking at least five hours for the flagellum to reach its full length. In the ameba-flagellate Naegleria gruberi the flagellum was observed to grow to its natural length in less than 10 minutes after the appearance of a basal body (1m. Gradually fibrils and cytoplasm still covered by the cellular membrane protruded from the surface until 17 the flagellum reached its full length (10). The flagellate study indicated that the growth zone of the growing flagel- lum might be the end, i.e., the flagellum becomes longer by adding on material to its end. Autoradiography with tritiated leucine indicated this possibility. Furthermore the turnover of flagellar protein is quite rapid--36-37% of the radioactive label was lost after 18 hours (44). Nothing is known concerning the development of the fibrillar structures in Tetrahymena-—this is unfortunate because this organism provides an excellent system for such a study. In Tetrahymena it is possible to synchro- nize a large number of cells to divide at one time (50)-- fixing of such cells during various phases of synchrony for electron microscopy would allow the Opportunity for such a fibrillar developmental study. In Naegleria it has also been observed that newly functioning kinetids appear to have kinetosomes and fla— gella but no fibrils (51). From the known information it thus appears as though the basal body must be present for flagellar develOpment and that such development occurs possibly by the addition of new material to the end. Since the turnover of protein in flagellum was shown to be fairly rapid, there is probably some specific pathway that new material follows when passing to the Cilium or flagellum from the cell body. If the elec- tron dense borders of the ciliary and basal body microtubules 18 are indicative of actual size there remains the physically possible pathway up the core of the peripheral tubules of subunits necessary for flagellar maintenance and growth. V. INTENT OF WORK After the preceding review of some of what is known about the kinetid, a description of the object of this in- vestigation follows. The intricate fine structure of the components of the kinetid plus the observation that such fine structure must be reproduced in some manner during every cell cycle led this investigator to the problem of the isolation of the center of the system--the kinetosome. Initial interest was developed via the implications of the findings of DNA in other organelles such as mitochondria and chloroplasts. It has been reported by Sonnenborn that a Paramecium which accidentally received a cortical piece of another Paramecium proceeded to develop a new clone which had developed an extra set of cortical structures from this original piece (55). Thus if there is not genetic material within the kinetosomes there are at least some structural constraints apart from nuclear inheritance that effect the development of pellicular organelles. With such thoughts in mind an attempt to isolate the kinetosome was made with the eventual plan being to determine its chemical composition. 19 Seaman's (52) published technique of kinetosomal isolation consisted basically of a prefixing of the cell, followed by solubilization of parts of the cell, and finally homogenization and isolation of the unsolubilized parts. Unfortunately such unsolubilized parts were not all kineto- somes. The basic technique chosen to attempt the isolation might be called classical—-homogenization of cells followed by layering of the homogenate above a linear sucrose gra- dient and ultracentrifuging so that an approximate equilib— rium is eventually reached in the ultracentrifuge tube whereby each cellular particulate eventually reaches a level of sucrose density equivalent to its own average den- sity. If there are many particulates of one kind having the same inherent density, they will form a band if their density lies within the range of gradient densities (9). Thus, for success, one assumes that the particulates he wished to separate all have the same density, and that this average density is separably different from the density of other cell particulates in the homogenate. Actually is it not as simple as that-~some of the possible problems that occur will be described below. Most quantitative ultracentrifugal technique in- volves work with simple or macromolecules--with cellular particulates many more problems occur and effect the results to the extent that the technique becomes basically a qualita- tive one. 20 With an equilibrium technique particle shape and size do not play as important a role compared to techniques which involve the recording of Speed of movement of material through a gradient, i.e., size and shape might effect fric- tional forces a particle would experience in travelling through a liquid but if sufficient ultracentrifugal time is allowed the particle will reach its sucrose density level. There are problems such as water of hydration of the different chemicals making up the particulate--different sizes and Shapes here might effect the average density. Also, particulates such as mitochondria are naturally osmo- tic and this will effect the average density of the par- ticulate (9). In using this ultracentrifugal technique there are a variety of gradients materials that might be used. Sucrose is most commonly used and has been used in this experimental work; however chemicals such as cesium chloride are also commonly used. With sucrose a gradient is made with a small machine which mixes the appropriate sucrose solutions, but with cesium chloride the gradient is formed in the ultra— centrifuge; this second technique is more accurate and has been used extensively for DNA density averages but involves many hours of preliminary ultracentrifugation to first form the gradient. Heavy water, D O, has been used to separate par— 2 ticulates that normally have the same density in a sucrose gradient. If the particulates have sufficiently different 21 hydration levels separation will be possible due to an increased density of the particle with a higher hydration level (9). The simplest and most commonly used technique of sucrose gradient ultracentrifugation was first used and continued--the Specific methods will be given later but the rationale behind the basic steps will be given here. After the cells are homogenized the suspension might contain small particles up to large pieces of cells or even whole cells. To eliminate the larger pieces the homogenate is resuspended and centrifuged Slowly to sedi- ment these pieces--this pellet is thrown out if small or rehomogenized if large. The supernatant from this first centrifugation is then spun quickly to remove particles of the Size range of mitochondria and lysosomes. The supernatant from this centrifugation contains ribosomes and other small particles which are thrown away. The pellet is washed again and then added to a linear sucrose density gradient. After an empirically determined number of hours of ultracentrifuging the particulates will reach a level in the ultracentrifuge tube with a sucrose density equivalent to their average density in that gradient solu- tion. If there is enough material the resultant separation will Show up as a number of bands each representing an ac- cumulation of particulates at a Specific sucrose density range. 22 AS might be apparent, for reproducibility and good separation one must be able to produce consistently accurate lineargradients. With this work the gradient was formed in a small plastic chamber to which was added the densest sucrose solution. Within this chamber was a stirrer. When the gradient was being made the less dense sucrose solution was injected Slowly into the chamber with an equivalent volume of air and the resultant gradient was forced out another hole into the ultracentrifuge tube. VI . METHODS All the work in this study was concerned with the amicronucleate strain of the ciliate protozoan Tetrahymena pyriformis. A. Culturing The cells were grown in 1.5 liter quantities in low form culture bottles on 1% proteose peptone plus 0.1% yeast extract. The autoclaved media was innoculated with 1-2 ml of a two day old test tube culture and allowed to grow for 48—72 hours. B. Separation Many of the parameters were changed in the course of this work and such changes will be described in the re- sults section--for explanation a typical separation run will be described below (also refer to Figure l). 23 After 48 to 72 hours of growth, 3 liters of media was centrifuged in a Foerst continuous flow centrifuge to sediment the cells. The cells were then washed in distilled water at 2,500 rpm in the SS34 rotor of a Sorvall RC2 re— frigerated centrifuge. The cells were resuspended and washed again in the homogenization buffer (0.4M sucrose, lmM EDTA, SmM tris, pH 8.2). The resultant pellet was re- suspended to creamy viscosity and homogenized with a hand operated colloidal homogenizer. The homogenate was then centrifuged at 2000 rpm for 5 minutes at 0°C in the Sorvall to sediment unhomogenized cells and larger pieces. If this pellet was large it was rehomogenized--otherwise it was normally discarded. The supernatant was then centrifuged in the Sorvall for 30 minutes at 0°C at 15,000 rpm. The supernatant was discarded and the pellet was resuspended again in the 0.4M sucrose buffer and centrifuged again at 15,000 rpm for 30 minutes. The pellet was resuspended to a creamy viscosity and layered over a linear sucrose gra- dient formed in 3 Spinco 5 ml ultracentrifuge tubes. The tubes were centrifuged for at least 4 hours in a Spinco Model L preparative ultracentrifuge, swinging bucket rotor SW39 at 39,000 rpm. The resultant separation was removed from the tubes in one of two ways: (1) either the tube was capped and punctured at the bottom to allow a very dense sucrose to be pumped through the bottom hole so that the gradient 24 could be collected from a tube connected to the cap; or (2) the material was collected with a fine U-tipped pipette placed in the tube below each band and connected by tubing to a micrometer controlled spring tensioned syringe for fine control of liquid uptake. The separated material was then observed with the light microsc0pe or prepared for electron microscopy. C. Electron Microscopy The material was prepared for the electron micro- scope using the two most common techniques: (1) negative staining, 2% phosphotungstic acid was adjusted to pH 6.8 by adding concentrated NaOH (0.01% albumin was also added for better Spreading of material). A small drop of the homogenate was dropped upon a drop of the staining mixture-- carbon and parlodion coated grids were passed through the drop and allowed to Sit for a minute or two on parafilm—- then the remaining liquid was removed with filter paper. (2) fine sectioning, the separated material was first fixed in cold 1.5% gluteraldehyde buffered with the phosphate buffer of Sabatini, Bensch, and Barrnett (38) overnight. The material was then washed several times in plain buffer and allowed to sit for at least one day in the buffer. This was then followed by fixation in Zetterquist buffered osmium textroxide at 0°C for 40 minutes. The material was then rinsed in distilled water and the following dehydra- tion and embedding schedule was followed. 25 l. 15 minutes in 50% ethanol 2. 15 minutes in 75% ethanol 3. 15 minutes in 95% ethanol 4. 15 minutes in 100% ethanol 5. 15 minutes in 100% ethanol 6. 15 minutes in 100% ethanol 7. 30 minutes in 100% ethanol 8. 20 minutes in propylene oxide 9. 20 minutes in propylene oxide 10. 50/50 mixture of propylene oxide and the embedding resin Araldite and Maraglas were both used as embedding resins. The Araldite formula used follows (18): \ Araldite 27 ml DDSA 23 ml DMP 30 1 ml The Maraglas formula used (18): Maraglas 34 ml Cardolite NC 513 10 m1 Dibutyl phthalate 5 ml BDMA 1 ml In both cases the mixtures were stirred for at least 10 minutes before use. For Araldite the embedding schedule follows: 11. place in 2:1 Araldite:pr0pylene oxide overnight 12. add Araldite mixture to dried Size 00 gelatin capsules and add material on top of capsule-—allow material to Sink 26 to bottom of capsule and then place in oven at 48°C over— night 13. temperature is raised to 60°C and the capsules are left for at least 48 hours V For Maraglas 11. after in 50/50 for 1/2 hour place in pure mixture and keep for 12 hours at 10°C 12. material was placed in dried Size 00 gelatin capsules and kept at 60°C for at least 48 hours to harden When sufficiently hard the gelatin was removed with warm water and the material was faced off with razor blades. Silver sections were obtained from an LKB Ultra- tome and a Porter-Blum MT-2 microtome. The sections were cut with glass knives, floated on water, flattened with chloroform vapors, and collected on parlodion coated-carbon coated grids. They were then double stained with uranyl and lead stains. The uranyl acetate stain was prepared by mixing 15 grams of uranyl acetate with 25 ml of methanol. This was left overnight and filtered before use the next day. The grids were placed face up in watchglasses of the stain for 20 minutes. They were rinsed briefly in plain methanol once, then 50/50 methanol/distilled water, and finally in distilled water. Reynolds lead citrate stain was used for the second stain (38). 27 The electron microsc0py was done by Dr. R. Neal Band on a Hitachi HU ll electron microscope at 75kv ac— celerating voltage. The micrographs were taken on Kodak electron image plates. VII. RESULTS Though the technique that was used for separation-- commonly called isopycnic ultracentrifugation-—has been used by several workers on many different types of bio- logical material, the beginning of this study was spent determining the various parameters involved empirically so that a separation might best be made with the Specific ma— terial of interest. The nature of the eventual separation varied de- pending upon a number of factors--those studied at least qualitatively were: time of ultracentrifuge run, pH of homogenization medium, quantity of material added to the sucrose gradient, EDTA concentration, nature of the addi- tion of the material to the gradient, temperature of pro- cedures, and density range of the gradient. A thorough study of the effect of changes in all of these parameters would have taken years--especially if the results were all to be observed with the electron micro- scope. Such time and patience were not available so the criterion for use of a specific value or technique was an apparent clearer separation as judged visually from the 28 Sharpness of the resultant bands found in the tube after the ultracentrifugation. Since this ultracentrifugal technique involves an approximate equilibrium between the buoyant force upon the material and the imposed centrifugal force, the density range of the gradient was chosen first. It was assumed that the material was at least partially proteinaceous in nature and the first attempt used a gradient sufficient to suspend mitochondria but wide enough to cover a range of densities beyond this one (i.e., sucrose molarities from 0.4 to 2.2 covering a density range of approximately 1.05 to 1.29 at 0°C). Apparent bands lay somewhat within this range (see Figure 2) so that the gradient was eventually reduced to the range of 0.6M to 1.9M. The solution chosen affected somewhat the resultant quantity of material that was available for ultracentrifugal separation e.g., with the divalent cation chelating agent ethylene diamine tetra- acetic acid (EDTA) there was less clumping of material in the differential centrifugations and finer suspension of material for layering (i.e., if two different cellular particulates of different average densities clumped to- gether because of the presence of divalent cations, they would move down to a sucrose density equivalent to the average density of the two particulates together and not their individual average densities). Affecting this che- lating action is the pH of the medium. It has been Shown (32) 29 that EDTA works best as a divalent cation chelating agent at a pH above 8.0. The nature of the addition of the material to the gradient was changed using the three most common methods: (1) layering of the homogenate above the gradient in a Solution of lesser density than the lowest gradient density; (2) forming the sucrose gradient above the homogenate—-the homogenate being in a solution of sucrose denser than the densest gradient solution; and (3) adding the material con- tinuously to the gradient while it is being formed. All the methods have advantages--by forming the gradient above the homogenate one eliminates the possibility of Side pack- ing of material when the homogenate is layered above (i.e., since the centrifugal force is radial some material near the parallel sides of tube may be forced to the Side and be packed together). The addition of the material to the gradient continuously allows for best initial separation of the suspension and thus lessens the possibility of clumping. By far the easiest and shortest method to use is the tOp layering technique. With this method the mate— rial has just to follow the centrifugal force whereas in the bottom layering and continuous addition methods some of the homogenate has to float up against this force--in both of these the ultracentrifugal time necessary is rela- tively quite longer. These three techniques were all used with the best apparent results coming from the top layering 30 and continuous addition methods (there was so much more material in the lowest band that it seemed that more clump- ing took place at this level and prevented the material normally from higher bands to reach a higher level when the homogenate was first placed on the bottom). The easier and faster technique of top layering was generally used since there was no appreciable difference in the observed separations of the tOp layering and continuous addition techniques. The EDTA—pH effect was observed by comparing sepa- rations of two EDTA-pH 7.0 tubes with one EDTA-pH 8.2 tube. More material was recovered from differential centrifuga- tion via the pH 8.2 technique and an apparent better sepa- ration (the other two tubes were not as clear) was obtained. Variations of EDTA concentration from 0 to lOmM were tried with lmM at pH 8.2 being not noticeably different from 10mM at pH 8.2 (lmM thus was generally used). The time of the ultracentrifuge run depended upon the way in which the material was added. For continuous addition and bottom layering techniques the run was at least 12 hours. For the tOp layering technique there generally appeared to be no change in separation if the ultracentri- fugation was run more than 4 hours so 4-6 hours the normal length of time used (the time varying according to the time at which the investigator would be free to turn off the ultracentrifuge and separate the material in the tubes). 31 The amount of homogenate added to the gradient varied from time to time depending upon the other selected parameters mentioned above. It was found that too little material would cause visible loss of some bands (particu- larly the top bands--see Figure 2) whereas too much would help increase clumping of material and thus give poor sepa- ration. Therefore when appropriate parameters were finally selected (see flow sheet-Figure l) the best results were obtained when 3 liters of a three day old culture were homogenized, differentially centrifuged, and layered over three 5 ml ultracentrifuge tubes. The first runs characteristically showed one large yellowish band at a density calculated to be approximately 1.23 with some rather dispersed material above it. Follow- ing the above described procedure the dispersive material eventually was discriminated into three separate bands (see Figure 2) all with much less material than the lower yellowish band. There was also a small amount of material floating at the tOp of the tubes and sedimented at the bottom. Once these bands were discriminated the above mentioned parameters were changed with the intention of obtaining the sharpest separation before electron micro— scopic examination (this was done for the reason that the electron microsc0pic observation would be made by Dr. R. Neal Band and would be made via the use of another depart- ment's electron microscopic facility--this limited 32 availability meant that scope time Should not be wasted upon poor separations). Negative staining was attempted for its ease of preparation-—electron microscopic observations of all four bands indicated an.amorphousnature. Assuming that the negative stain might be affecting the material or that its structure might not be adequately shown with negative stain, it was decided to fix, embed the material in resin, and fine section it. Electron microscopic observations Showed that the first three bands indicated in Figure 2 were all somewhat similar and composed of membranous vesicles as shown in Figure 3. The fourth band was positively identified as mitochondria. In only one case during the electron micro- scopy was an apparent basal body observed (apparent Since the section was thick and the peripheral fibrils could not be resolved as triplets or doublets--there were no central fibrils and no membrane) and this was amidst the membranous vesicles of a section of the third band material. Considering these negative results it was assumed that either the basal bodies were disappearing (falling apart into fibrils or their subunits) during one of the preparative steps or they were in a fraction that had been thrown away. A normal run was then made and the material in the supernatant of the fast centrifugal runs (15,000 rpm for 33 30 minutes) was sedimented in the ultracentrifuge and pre- pared for electron microscopy. .The tubes which contained the normal separation were frozen quickly and sliced into sections. The material in each section was collected and prepared for electron microsc0py. Observations showed that the material from the top of the tube and from the bottom of the ultracentrifuge tube looked very small and granular and quite unlike basal bodies or their fibrils. The mate- rial normally left in suspension after the fast 15,000 runs appeared to look the same as that collected from the top and the bottom of the ultracentrifuge tube. The other sections of the tube showed material Similar to what would be expected from the band they included. To examine the possibility that the lack of diva- lent cations might be causing a loss of structure, material was washed, homogenized, and prepared for layering over the sucrose gradient in the normal fashion described earlier but with no EDTA present in the buffer. Instead of layer- ing over the gradient the homogenate was prepared for elec— tron microscopic examination along with a normal EDTA con- taining homogenate. Both homogenates looked similar in the electron microsc0pe--Several membranous vesicles, some mitochondria, and no basal bodies. Two final possibilities were then checked: (1) the material normally sedimented after the 2000 rpm run immedi- ately after homogenization was collected and prepared for 34 electron microscopy; and (2) considering the published find- ings of loss of microtubular structures when organisms are brought to 0-4°C it was decided to examine such a possibility with the homogenate material. Cells were collected, washed, and homogenized at room temperature. Immediately after homogenization the homogenate was fixed in 1.5% buffered gluteraldehyde described earlier at room temperature over- night. This was followed with washing in plain buffer and then normal fixing in osmium tetroxide at 0°C as described earlier. Basal bodies were observed in the Slow run sediment but in all cases they were observed in patches including several basal bodies (see Figure 5). It appeared as though the basal body seen were always attached in some way to other basal bodies--most probably by membrane. Basal bodies were also observed in the room tempera- ture fixed material. Since this was a total homogenate and not larger pieces of cells, basal bodies were observed more infrequently than those seen in the above described fraction. These observations also indicated that basal bodies were to be found either in patches with other basal bodies or at least still attached to pieces of membrane (see Figure 6). VIII. DISCUSSION The results are somewhat negative in that individual basal bodies in large numbers were not isolated. 35 Looking quickly at what has been found: (1) three bands of membranous vesicles appearing quite similar in structure in the electron microscope were separated at three different sucrose densities all lower in density to that of mitochon— dria; and (2) when basal bodies were observed via the elec— tron microscope they were Shown either arranged in a definite pattern indicating their attachment to larger pieces of cells or if observed singly, still having some membrane attached to them (though in this situation the section might have been through only one basal body of a larger patch of basal bodies connected together--no basal bodies were positively identified in any fraction remaining after the Slow cen- trifugation following the homogenization). The first finding is interesting but apparently not involved with the intended purpose of this study—-how— ever it is intriguging to note that Tetrahymena pyriformis has been shown to have three membranes composing its pel— licle (39). The second observation might in some way relate to the microtubular work described earlier. The Slow run sediment fraction most probably does not include all the basal bodies originally in the homogenized cells. That basal bodies were not observed in the later fractions could be explained in at least two ways: (1) a few thin sections of material on an electron microscopic grid is not an ade- quate statistical sample for what is in the original 36 separated fraction. For example, when material is eventually to be fixed for electron microscopic observations it is cen- trifuged into a pellet and this pellet might contain layers of different particulates--since only small parts of this pellet will be observed one might observe only some of the particulates of the fraction; and (2) it may be that under the conditions used in this study additional structures are necessary to hold the basal body together after homogeniza- tion. That the first possibility might have occurred is highly unlikely. The quantity of material collected from the ultracentrifuge tube was usually just sufficient to see with naked eye--since sections were taken from several pieces of each pellet it is highly probable that the mate— rial observed with the electron microsc0pe was representa- tive of what the total fractions consisted. The second suggestion is more probable and in a sense allows me to close this study at this point. Under the conditions used it appears as though the basal body ‘ in some way falls apart if not attached to membrane and perhaps other structures. Whether this falling apart in— volves dissolution into fibrils or fibril subunits is not known (though fibrils were not observed in later fractions either). The next direct step might be to attempt an iso- lation of the larger pellicular pieces from the slow run pellet and if possible work with these. The use of nonpolar 37 solvents might possibly allow a retention of structure if such larger pieces were homogenized. Perhaps D 0 might 2 stabilize the kinetosomal structure as it apparently does with the mitotic Spindle structure (30). Two other possibilities of exploring the problem of basal body isolation might be: (1) a direct method of working with the method of Seaman (52) or designing a similar prefixing prOcedure before homogenization; or (2) an indirect method involving first the assumption that peripheral ciliary doublet fibrils are identical to basal body triplet fibrils. Since such peripheral doublets can be isolated in large quantities (15), a study of structure, chemical composition, and bonding forces in a way Similar to the techniques used for analysis of the tobacco mosaic virus (25) might provide suggestions for possible condi- tions and techniques to be used for isolation of intact basal bodies from cells. 10. 11. 12. 13. 14. 15. 16. 17. IX. BIBLIOGRAPHY Argetsinger, J., J. Cell Biol. 24, 154 (1965) Behnke, 0., J. Cell Biol. 35, 697 (1967) Behnke, O. and A. Forer, J. Cell Sci., 2, 169 (1967) Behnke, O. and T. Zelander, J. Ultrastructural Res., lg, 147 (1967) Chatton, E., Compt. Rend. Soc. Biol" 91, 577 (1924) Child, F. M. and D. Mazia, Experientia, 12, 161 (1956) Child, F. M., J. Cell Biol., 21, 18A (1965) Chun, E. H., M. H. Vaughn, Jr.,and A. Rich, J. Mol. Biol., 1, 30 (1963) DeDuve, C., J. Berthet, and H. Beaufay, Prog. Biophys. Biophys. Chem., 9, 325 (1959) Dingle, A. D., and C. Fulton, J. Cell Biol., 31, 43 (1966) Ehret, C. F., and G. DeHaller, J. Ultrastructural Res. Suppl., 6 (1963) Furgason, W. H., Arch. Protistenk., 94, 224 (1940) Gall, J. G., J. Biophys. Biochem Cytol., 19, 163 (1961) Gibbens, I. R., and A. V. Grimstone, J. Biophys. Bio- chem. Cytol., Z, 697 (1960) Gibbons, I. R., Arch. Biol. (Liege), 16, 317 (1965) Gibbons, I. R., and A. Rowe, Science, 149, 424 (1965) Gibbons, I. R. in "Molecular Organization and Biological Function," ed. by J. M. Allen, Harper and Row, New York (1967): p. 211 38 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 39 Glauert, A. M. in "Techniques for Electron Microsc0py," ed. by D. H. Kay, Blackwell Scientific Publications, Oxford, Great Britain (1965), p. 166 Grimstone, A. V., and A. Klug, J. Cell Sci., 4, 351 (1966) Hoffman, E. J., J. Cell Biol., 25, 217 (1965) Inoue, S. in "Primitive Motile Systems in Cell Biology," ed. by R. D. Allen and N. Kamiya, Academic Press, New York (1964), p. 549 ‘ Korn, E. D., Science, 153, 1491 (1966) Korn, E. D., J. Cell Biol., 34, 627 (1967) Lauffer, M. A., A. T. Ansevin, T. E. Cartwright, and C. C. Brinton, Jr., Nature, 181, 1338 (1958) Lauffer, M. A. in "The Molecular Basis of NeOplasia," University of Texas Press, Austin (1962), p. 180 Lehninger, A. L., "The Mitochondrion: Molecular Basis of Structure and Function, W. A. Benjamin, Inc., New York (1964) Luck, D. J., J. Cell Biol., 44, 483 (1963) Luck, D. J., and E. Reich,’Proc.'Nat. Acad. Sci., 5;, 931 (1964) Lwoff, A., "Problems of Morphogenesis in Ciliates," Wiley and Sons, New York (1950) Marsland, D., and A. Zimmerman, Exptl. Cell Res., 39, 23 (1963) Marsland, D. in "Primitive Motile Systems in Cell Biology," ed. by R. D. Allen and N. Kamiya, Academic Press, New York (1964), p. 173 Martell, A. E., and M. Calvin, "Chemistry of Metal Che- late Compounds," Prentice-Hall, Inc., New York (1952), p. 494 Mazia, D., P. J. Harris, and T. Bibring, J. Biophys. Biochem. Cytol., Z, l (1960) Nass, M. M. K., and S. Nass, J. Cell Biol., 49, 593 (1963) 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 40 Nass, M. M. K., Proc. Nat. Acad. Sci., 33, 1215 (1966) Nass, S., and M. M. K. Nass, J. Cell Biol., (1963) Pease, D. C., J. Cell Biol., 33, 313 (1963) Pease, D. C., "Histological Techniques for Electron 1_9_, Microsc0py," Academic Press, New York (1964) Pitelka, D. R., J. Protozool., 3, 75 (1961) 613 Pitelka, D. R., and F. M. Child in "Biochemistry and Physiology of Protozoa," vol. Academic Press, New York (1964), p. 131 Randall, J., J. M. Hopkins, J. M. Eadie, and R. W. Butcher, Proc. Linnean Soc. London, 174, 31 (1962) 3, ed. by S. H. Hutner Randall, J., and C. Disbrey, Proc. Roy Soc. London, B, 162, 473 (1965) Renaud, F. L.--information given in Zoology department seminar on work done with I. R. Gibbons, Michigan State University, February 3, 1967 Rosenbaum, J. L., and F. M. Child, 345 (1967) Roth, L. E., and E. W. Daniels, J. J. Cell Biol., Cell Biol., (1962) Roth, L. E., J. Cell Biol., 34, 47 Sager, R., and M. R. Ishida, Proc. (1967) Nat. Acad. 1_2, Sci., _5_o_, 725 (1963) Satir, B., and J. L. Rosenbaum, J. Protozool., 397 (1965) 17;: Schatz, G., E. Haslbrunner, and H. Tuppy, Biochem. Biophys. Res. Comm., 33, 127 (1963) Scherbaum, 0., and E. Zeuthen, Exptl. Cell Res., 3, 221 (1954) Schuster, F. L., J. Protozool., 33, 297 (1963) Seaman, G. R., Exptl. Cell Res., 33, 292 (1960) Sinclair, J. H., and B. J. Stevens, 218A (1966) J. Cell Biol., a. 57 lb 54. 55. 56. 57. 58. 41 Smith-Sonnenborn, J., and W. Plaut, J. Cell Sci., 3, 225 (1967) Sonnenborn, T. M. in "The Nature of Biological Diversity," ed. by J. M. Allen, McGraw—Hill Co., Inc., New York (1963), p. 165 Tibbs, J., Biochim. Biophys. Acta, 33, 275 (1957) Tilney, L. G., and K. R. Porter, J. Cell Biol., 34, 327 (1967) Watson, M. R., and J. M. HOpkins, Exptl. Cell Res., _3, 280 (1962) X. FIGURES Figure 1. Sketch of isolation method grow 3 liters of Tetrahymena pyriformis in 1% proteose peptone plus 0.1% yeast extract for 48-72 hours 1 harvest in Foerst continuous flow centrifuge I wash twice: (1) distilled water and (2) buf- fer at 25000 rpm for 5 minutes in Sorvall centrifuge at 0°C 1 homogenize cells in hand colloidal homogenizer centrifuge at 2000 rpm for 5 minutes at 0°C supernatant is centrifuged \\\\‘s at 15,000 rpm for 30 :3 A: rehomogenize pellet minutes at 0°C \. resuspend pellet and resediment at 15,000 rpm for 30 minutes at 0°C 1 add sufficient buffer to make creamy suspension _of pellet 1 layer suspension over the linear sucrose density gradient 1 centrifuge tubes in Spinco model L preparative ultracentrifuge, roter 39SW (swinging bucket) for 4-6 hours at 39,000 rpm separate bands from ultracentrifuge tubes I prepare material for electron microscope 42 43 Figure 2. Description of Material in Various Fractions 1. slow run pellet--patches of basal bodies, parts of whole cells, cilia, other material, see Figure 5 2. material left in suspension after 15,000 rpm for 30 minute run--small granular, unlike tubules or basal bodies 3. pellet after two 15,000 rpm runs a. with EDTA--membranous vesicles, mitochondria b. without EDTA--membranous vesicles, mitochondria 4. material from ultracentrifuge tubes a. material floating on top of tube—- small and granular b. bands material on top band l-—membranous vesicles band 2—-membranous vesicles-- see Figure 3 band 1 band 2 band 3--membranous veSlcleS band 3 band 4--mitochondria band 4 c. pellet on bottom--dirt, also small and granular material 5. homogenate prepared and fixed at room temperature--cilia, see Figure 4, basal bodies, see Figure 6, mitochondria, vesicles, and other material 44 Figure 3. Membranous vesicles from band 2 of the ultra— centrifugal separation. The material in bands 1 and 3 appeared quite similar. Figure 4. A longitudinal section of a Cilium from the slow run pellet. Notice as the Cilium moves out of the plane of the section that peripheral doublet fibrils can be seen at both ends. FIGURE 4 46 Figure 5. A cross section of several basal bodies obtained from the Slow run sediment immediately following homogeniza- tion. Figure 6. A longitudinal section of two basal bodies with material still attached to them. This section was from room temperature prepared material. noun: a d -“ ' "V o L . 1,.: . m ' F. v ._ 5- . q |. I. . ' h ‘ . ‘ .f 7 1 FIGURE 0 48 Figure 7. A sketch of a cross section of a Cilium. Dimen- sions taken from Gibbons (17). Figure 8. A Sketch of a cross section of a basal body. Drawn from Pitelka and Child (40). Figure 9. A sketch of a ciliary peripheral doublet Showing ATPase projections (longitudinal view). Dimensions taken from Gibbons (17). Figure 10. A sketch of a longitudinal view of a Single flagellar peripheral fibril Showing the beadlike filamentous structure of its tubular wall. From Grimstone and Klug (l9). C 2 mlcrons J 1 AOstcronL I I I ”R\ Sb .: &:b 4“ ' (n) 9'82“: .9 I B . . . - >Dernpheral doublefs S §>triplets FIGURE 7 FIGURE 8 I glmagrgn§ ' E .0. 00...... ”omuuoooououuuuoo “mommoooooumfi >ATPon prOJocHons mumoomooouooooooohouoo .0. 00.000000 ”0.000000... Wane-00000000000000.0000 “Gomoonuog FIGURE 9 FIGURE IO